arXiv:1701.01373v3 [astro-ph.EP] 1 Aug 2017

Atmospheric Constraints on the Surface UV Environment of Mars at 3.9 Ga Relevant to Prebiotic Chemistry
Sukrit Ranjan1,2, Robin Wordsworth3,4, Dimitar D. Sasselov1
August 2, 2017
ABSTRACT
Recent ﬁndings suggest Mars may have been a clement environment for the emergence of life, and may even have compared favorably to Earth in this regard. These ﬁndings have revived interest in the hypothesis that prebiotically important molecules or even nascent life may have formed on Mars and been transferred to Earth. UV light plays a key role in prebiotic chemistry. Characterizing the early Martian surface UV environment is key to understanding how Mars compares to Earth as a venue for prebiotic chemistry.
Here, we present two-stream multi-layer calculations of the UV surface radiance on Mars at 3.9 Ga, to constrain the surface UV environment as a function of atmospheric state. We explore a wide range of atmospheric pressures, temperatures and compositions, corresponding to the diversity of Martian atmospheric states consistent with available constraints. We include the eﬀects of clouds and dust. We calculate dose rates to quantify the eﬀect of diﬀerent atmospheric states on UV-sensitive prebiotic chemistry.
We ﬁnd that for normative clear-sky CO2-H2O atmospheres, the UV environment on young Mars is comparable to young Earth. This similarity is robust to moderate cloud cover: thick clouds (τcloud ≥ 100) are required to signiﬁcantly aﬀect the Martian UV environment, because cloud absorption is degenerate with atmospheric CO2. On the other hand, absorption from SO2, H2S, and dust is nondegenerate with CO2, meaning if these constituents build up to signiﬁcant levels, surface UV ﬂuence can be suppressed. These absorbers have spectrally variable absorption, meaning that their presence aﬀects prebiotic pathways in
1Harvard-Smithsonian Center for Astrophysics, Cambridge, MA 02138, USA 260 Garden Street, Mail Stop 10, Cambridge, MA 02138, USA; sranjan@cfa.harvard.edu; 617-495-5676 3Harvard Paulson School of Engineering and Applied Sciences, Harvard University, Cambridge, MA 02140, USA 4Department of Earth and Planetary Sciences, Harvard University, Cambridge, MA 02140, USA

–2–
diﬀerent ways. In particular, high SO2 environments may admit UV ﬂuence that favors pathways conducive to abiogenesis over pathways unfavorable to it. However, better measurements of the spectral quantum yields of these pathways are required to evaluate this hypothesis deﬁnitively.
Subject headings: Radiative Transfer, Origin of Life, Mars, UV Radiation, Prebiotic Chemistry
1. Introduction
Recent ﬁndings suggest that young Mars may have been a clement environment for the emergence of life. Analysis of Curiosity imaging of sedimentary rock strata deposited 3.2-3.6 Ga (Grotzinger et al. 2015) suggest individual lakes were stable on ancient Mars for 100-10,000 years, with ﬂuvial features laid down over 10,000-10 million years assuming formation rates corresponding to modern Earth. Similarly, Curiosity measurements of olivine and magnetite at Yellowknife crater are consistent with aqueous conditions at near-neutral pH for thousands to hundreds of thousands of years in the Noachian, with an oxidant supply that could be an energy source (Bristow et al. 2015). In general, the geologic evidence is compelling that liquid water, a requirement for life as we know it, was present on Mars at least transiently in the Noachian (Wordsworth 2016).
The young Mars may also have been a favorable environment for prebiotic chemistry (chemistry relevant to the origin of life). Meteorite analysis has detected boron in Martian clays, important for abiogenesis since borate minerals can stabilize ribose and catalyze other prebiotic chemistry reactions (see Stephenson et al. 2013 and sources therein). Mars may also have enjoyed greater availability of prebiotically important phosphate than Earth (Adcock et al. 2013). Climate models suggest liquid water was transient on Mars (Wordsworth et al. 2013), which suggests the evidence of wet/dry cycles. Such cycles are useful for prebiotic chemistry: aqueous eras are beneﬁcial for the formation of biotic monomers, while dry eras tend to concentrate feedstock molecules and aid monomer polymerization (Benner et al. 2015), relevant to the formation of nucleotides and amino acids (Patel et al. 2015). Finally, the putative dryness of Mars and the potential acidity of its early aqueous environment owing to dissolved carbonic acid from a CO2-dominated atmosphere, suggest molybdate, which is suggested to catalyze formation of prebiotically important sugars such as ribose, may have been stable on Mars (Benner et al. 2015, 2010). Hence, there is growing interest in the possibility that prebiotically important molecules may have been produced on Mars (Benner 2013), and even the hypothesis that life may have originated on Mars and been seeded to Earth (Kirschvink and Weiss 2002; Gollihar et al. 2014; Benner et al. 2015).

–3–
Ultraviolet (UV) light plays a key role in prebiotic chemistry. UV photons can dissociate molecular bonds, produce ionic species, and excite molecules. These properties mean that UV light can stress prebiotic molecules (Sagan 1973), but also that UV light can power synthetic prebiotic photochemistry. UV light has been invoked in prebiotic chemistry as diverse as the origin of chirality (Rosenberg et al. 2008), the synthesis of amino acid precursors (Sarker et al. 2013), and the polymerization of RNA (Mulkidjanian et al. 2003).
The last decade has seen breakthroughs in long-standing problems in prebiotic chemistry such as the discovery of plausible mechanisms for the abiotic formation of activated pyrimidine ribonucleotides (Powner et al. 2009), the synthesis of short (2- and 3-carbon) sugars (Ritson and Sutherland 2012), and a reaction network generating precursors for a range of prebiotically important molecules including lipids, amino acids, and ribonucleotides (Patel et al. 2015). These pathways all require UV light to function. In experiments, line sources such as low-pressure mercury lamps with monochromatic 254 nm emission are often used to simulate the incident UV radiation. However, prebiotic UV radiance in CO2-dominated terrestrial-type atmospheres should instead be characterized by access to broadband ﬂuence (Ranjan and Sasselov 2017). The diﬀerence can have a signiﬁcant impact on prebiotic chemistry (Ranjan and Sasselov 2016), and there is a growing awareness in the prebiotic community of the importance of characterizing the wavelength dependence of proposed prebiotic pathways and/or using broadband sources in simulations (Rapf and Vaida 2016). Consequently, it is important to constrain the UV environment on the surface of Mars at epochs relevant to potential prebiotic chemistry on a spectral (wavelength-dependent) basis.
In this work, we use a two-stream multilayer radiative transfer model to constrain the surface UV environment on young Mars (3.9 Ga). We calculate the surface radiance as a function of solar zenith angle (SZA), surface albedo (A), and atmospheric composition. Our model can calculate absorption and scattering due to 8 gaseous species (CO2, H2O, CH4, SO2, H2S, O2, and O3) and 3 particulate species (H2O ice, CO2 ice, and Martian dust). Earlier analyses have focused on clear-sky ”case studies” for the atmospheric composition; we instead explore the full range of Martian atmospheric states consistent with available geological data and climate/photochemical modelling for the Martian atmosphere. We convolve the calculated surface radiance spectra against action spectra corresponding to two diﬀerent simple photochemical reactions (one useful to prebiotic chemistry, and one detrimental) that may have been important during the era of abiogenesis, and integrate the result to compute the biologically eﬀective dose rate (BED) and estimate the impact of these parameters on prebiotic chemistry.
In Section 2, we discuss previous work on this topic. In Section 3, we describe our radiative transfer model and its inputs and assumptions. Section 4 presents the surface

–4–
radiances calculated from our model as a function of Martian atmospheric state, and Section 5 discusses the implications for prebiotic chemistry. Section 6 summarizes our ﬁndings.
2. Background
Recognizing the importance of UV light to life (though mostly in the context of a stressor), previous workers have placed constraints on the primitive Martian surface UV environment. In this section, we present a review of previous work on this topic, and discuss how our work diﬀers from them.
Cockell (2000) calculate the Martian surface ﬂux at 3.5 Ga assuming solar input of 0.75× modern, and an atmosphere composed of 1 bar CO2 and 0.1 bar N2. They compute their radiative transfer in a cloud-free atmosphere using a monolayer two-stream approximation with Delta-Eddington closure. They ignore water absorption, but a dust optical depth of 0.1 is assumed, as is a Lambertian surface with surface albedo A = 0.1. Cockell (2000) report the total irradiance as a function of solar time (equator at equinox) as well as the biologically-weighted1 irradiance for DNA inactivation and photosystem damage, but not the spectral irradiance (for the early Mars case). They ﬁnd the DNA inactivation-weighted irradiance to be comparable for their models of Early Mars and Early Earth, leading them to suggest that from a UV perspective the two worlds were comparably habitable.
Ronto´ et al. (2003) calculate the Martian surface ﬂux at 3.5 Ga from 200-400 nm. They assume a 1-bar CO2 atmosphere overlying volatilizable surface H2O, and ran it through the PHOEBE photochemical model to generate atmospheric proﬁles for the other molecules that would be generated. They found a signiﬁcant population of spectrally absorbing O2, O3, and NO2 would be generated, including an ozone shield comparable to the modern Earth. They evaluated radiative transfer for both a pure CO2 atmosphere, as well as for an atmosphere with the trace species calculated in their photochemical model. Their UV radiative transfer models assumes pure absorption and ignores scattering. In this formulation, Ronto´ et al. (2003) calculated Rayleigh scattering cross-sections but treated them as absorption crosssections. This approach strongly overestimates attenuation at scattering wavelengths. It is consequently unsurprising that Ronto´ et al. (2003) report strong attenuation of surface UV ﬂuence for both model atmosphere cases. in the full photochemical model case, ﬂuence shortward of 290 nm is completely removed, due to the buildup of an ozone layer from CO2 photolysis. However, Segura et al. (2007) note that this photochemical model neglects supply of reducing gases to the atmosphere due to volcanism and sinks of oxic gases due to
1A measure of reaction rate; see Section 3.4

–5–
processes involving rainout, and including either of these eﬀects prevents the formation of an ozone layer. Geological evidence (e.g. the Tharsis plateau) indicates that young Mars had signiﬁcant volcanism, which would have prevented formation of an ozone layer of the type calculated by Ront´o et al. (2003).
Cnossen et al. (2007) calculate the Martian surface ﬂux from ∼ 3.5 − 4 Ga assuming a 5-bar CO2, 0.8 bar N2 clear-sky atmosphere. They used shortwave observations of the solar analog κ1 ceti combined with a scaled solar spectrum at longer wavelengths as their top-of-atmosphere (TOA) solar input. To calculate radiative transfer, they partition the atmosphere into 40 layers. They compute absorption using the Beer-Lambert law. To account for scattering, they compute the ﬂux scattered in each layer, assume half of it proceeds downwards and half of it proceeds upwards, and iterate this process to the surface. This approach implicitly neglects multiple scattering and assumes a surface albedo of 0, and hence tends to overestimate atmospheric attenuation of incoming radiation. Hence, Cnossen et al. (2007) report broadband suppression of the TOA ﬂux by multiple orders of magnitude.
Our work builds on these previous eﬀorts. Like Cockell (2000), Ronto´ et al. (2003), and Cnossen et al. (2007), we consider the eﬀects of a denser CO2 atmosphere; however, we consider a broader range of surface atmospheric pressures permitted by available constraints, ranging from 2 × 10−5 − 2 bar. We build on Cockell (2000)’s use of a monolayer two-stream approach to radiative transfer by using a multiple-layer two stream model, which consequently accounts for the eﬀects of multiple scattering. Such a treatment is essential because of the unique radiative transfer regime unveiled in thick anoxic atmospheres (e.g. multibar CO2 atmospheres) at UV wavelengths, characterized by the atmosphere being simultaneously optically thick and scattering-dominated. In this regime, multiple-scattering dominates and it is critical to account for its eﬀects in order to accurately compute surface radiation environments. As a corollary, the radiative transfer treatments of Ronto´ et al. (2003) and Cnossen et al. (2007) are not valid in this regime2.
In addition to varying levels of CO2, our work also explores the impact of other potential atmospheric constituents on the surface UV environment. In particular, we focus on the eﬀect of enhanced concentrations of volcanogenic gases (e.g., SO2, H2S), which may have been present at elevated levels on the young Mars (Halevy et al. 2007; Halevy and Head 2014) and if present could have had a dramatic eﬀect on the surface ﬂuence (Ranjan and Sasselov 2017). We also explore the radiative impact of varying levels of dust and CO2 and H2O clouds in the Martian atmosphere, which may have been abundant (Wordsworth et al. 2013;
2This regime is not available on the modern Earth or Mars, due to oxic absorption in the former and a thin atmosphere in the latter.

–6–
Halevy and Head 2014). Finally, previous workers reported the surface ﬂux. However, as pointed out by other workers, while the ﬂux is the relevant quantity when computing energy deposition, when computing molecular reaction rates the spherically integrated intensity, or actinic ﬂux, is the more relevant quantity (Madronich 1987; Kylling et al. 1995). For a particle lying at the planet surface, ﬂuence below the horizon is blocked by the surface. Therefore, we report instead the integral of the intensity ﬁeld at the planet surface over the hemisphere deﬁned by elevations > 0 (i.e. that part of the sky not blocked by the planet surface). We term this quantity the surface radiance. For more details, see Ranjan and Sasselov (2017).
3. Methods
In this section, we describe the methods used to calculate the surface UV environment of early Mars. All software associated with this project is available for validation and extension at https://github.com/sukritranjan/ranjanwordsworthsasselov2016.
3.1. Radiative Transfer Model
We use a multilayer two-stream approximation to compute the 1D radiative transfer of UV light through the early Martian atmosphere. Our code is based on the radiative transfer model of Ranjan and Sasselov (2017). In brief, we follow the two-stream treatment of Toon et al. (1989), and we use Gaussian (single) quadrature to connect the diﬀuse intensity to the diﬀuse ﬂux, since Toon et al. (1989) ﬁnd Gaussian quadrature closure to be more reliable than Eddington or hemispheric mean closure at short (solar) wavelengths. We include absorption and scattering due to N2, CO2, H2O, CH4, O2, O3, SO2, and H2S. For reasons of numerical stability, we assign a ceiling on the per-layer single-scattering albedo ω0 of 1−10−12. In Ranjan and Sasselov (2017), we included thermal emission from the atmosphere and surface, to enable application of our code to situations where planetary UV emission might be important. Early Mars is not such a case, so here we omit these features.
Our model requires the user to specify the partition of the atmosphere into homogenous layers, and to provide the temperature, pressure, and composition (gaseous molar concentrations) as a function of altitude. Section 3.2 describes our calculation of these quantities. Our model also requires the user to specify the wavelength bins over which the radiative transfer is to be computed; all spectral parameters are integrated over these wavelength bins using linear interpolation in conjunction with numerical quadrature. The user also must specify

–7–
the solar zenith angle (SZA) and albedo. The albedo may be speciﬁed as either a ﬁxed value (e.g. Rugheimer et al. 2015), or as a wavelength- and SZA-dependent user-determined mix of the albedos corresponding to diﬀerent terrestrial physical surface media (new snow, old snow, desert, tundra, ocean).
We take the top-of-atmosphere (TOA) ﬂux to be the solar ﬂux at 3.9 Ga, computed at 0.1 nm resolution from the model of Claire et al. (2012) and scaled to the Martian semimajor axis of 1.524 AU. We choose 3.9 Ga for the prebiotically-relevant era 1) because of evidence for at least transient liquid water on Mars around this time (Bristow et al. 2015; Grotzinger et al. 2015; Wordsworth 2016), 2) it postdates the potentially-sterilizing Late Heavy Bombardment (Maher and Stevenson 1988; Sleep et al. 1989), and 3) it predates the bulk of the evidence for the earliest terrestrial life (see Ranjan and Sasselov 2016 and sources therein). If one hypothesizes terrestrial abiogenesis was aided by transfer of prebiotically relevant compounds from Mars (Benner 2013; Gollihar et al. 2014; Benner et al. 2015), then the synthesis of these molecules and their transfer must have occurred concomitantly with the origin of life on Earth. We note that, unlike the XUV, solar output varies only modestly (within a factor of 2) from 3.5-4.1 Ga in the > 180 nm wavelength range unshielded by atmospheric CO2 or H2O. Therefore, our results are insensitive to the precise choice of epoch for abiogenesis.
In Ranjan and Sasselov (2017), we did not include scattering and absorption due to atmospheric particulates. However, clouds have been suggested to play a major role in Martian paleoclimate (Forget and Pierrehumbert 1997; Colaprete and Toon 2003; Wordsworth et al. 2013). Therefore, we updated our model to allow the user to emplace CO2 and H2O cloud decks of user-speciﬁed optical depth (at 500 nm) in the atmosphere. The cloud decks are assumed to uniformly span the atmospheric layers into which they are emplaced. Section 3.3 discusses the calculation of the particulate optical parameters (per-particle cross-section σ, asymmetry parameter g, and ω0). We use delta-scaling with Henyey-Greenstein closure (Joseph et al. 1976) to correct for the eﬀects of highly forward-peaked particulate scattering phase functions.
The fundamental output of our code is the surface radiance as a function of wavelength. The surface radiance is the integral of the intensity ﬁeld at the planet surface over the unit hemisphere deﬁned by elevations greater than zero, i.e. the intensity ﬁeld integrated over all parts of the sky not blocked by the planet surface. As we argue in Ranjan and Sasselov (2017), this is the relevant quantity for calculating reaction rates of molecules at planet surfaces (as compared to the actinic ﬂuxes for molecules suspended in the atmosphere, see Madronich 1987). In the two-stream formalism, this quantity is
Isurf = FN↓ /µ1 + FNdir/µ0,

–8–
where FN↓ is the downward diﬀuse ﬂux at the planet surface, FNdir is the direct ﬂux at the planet surface, and µ0 = cos(SZA) is the cosine of the√solar zenith angle. In Gaussian quadrature for the n = 1 (two-stream) case, µ1 = 1/ 3 (Toon et al. 1989); it can be interpreted as the eﬀective zenith angle for the diﬀuse ﬂux.
3.2. Atmospheric Proﬁle
We assume a CO2-dominated Martian atmosphere at 3.9 Ga. We take the atmosphere to be fully saturated with H2O. Typical calculations of Noachian climate call for steady-state local surface temperatures of T0 273K across the planetary surface, and more typically T0 ∼ 210−250K (Forget et al. 2013; Wordsworth et al. 2013).Both one- and three-dimensional calculations of Noachian climate produce global mean surface temperatures of 240 K or less (Forget et al. 2013; Wordsworth et al. 2013; Ramirez et al. 2014). At such cold temperatures, the H2O saturation pressure is very low, and H2O is a trace gas in the atmosphere. Therefore, we approximate the thermodynamic properties of the Martian atmosphere by the thermodynamic properties of CO2. We take cp = cp,CO2 and R = RCO2, where cp and R are the heat capacity at constant pressure and the speciﬁc gas constant respectively. We assume the heat capacity to be constant, with cp = cp(T0), where T0 is the surface temperature. We calculate cp = cp,CO2 from the Shomate relation, taking the coeﬃcients from Pierrehumbert (2010, page 115). We tested the eﬀects of permitting the heat capacity to vary with temperature, and found minimal impact on our results.
Martian paleoclimate models have been propose that invoke eﬀects like enhanced volcanism (Halevy and Head 2014) and high H2 abundance (Ramirez et al. 2014) to argue for global mean temperatures in excess of 273K. However, these models also require pCO2 in excess of 1 bar, meaning that H2O remains a trace atmospheric constituent. Regardless, our results are insensitive to the precise thermal properties of the atmosphere because of the modest variation of the absorption cross-sections of the gases in our model with temperature at UV wavelengths.
For a given surface pressure, surface temperature (P0, T0), we let the temperature decrease as a dry adiabat until it reaches the CO2 saturation temperature, at which point it follows the CO2 saturation curve. We use the empirical saturation curve of Fanale et al. (1982), as in Wordsworth et al. (2013). To avoid the need for a full radiative-convective climate model, which is tangential to our objectives in this paper, we assume a stratosphere starting at 0.1 bar, following the observation of Robinson and Catling (2014) that atmospheres dominated by triatomic gases tend to become optically thin and hence radiatively dominated around that pressure. We follow other workers (e.g. Kasting 1991; Hu et al. 2012;

–9–
Halevy and Head 2014) assuming the stratosphere to be isothermal; we conduct sensitivity studies demonstrating our results are not sensitive to this assumption.
To calculate the H2O saturation pressure, we use the empirical formulation of Wagner et al. (1994) via Wagner and Pruß (2002) for the vapor pressure of water overlying a solid reservoir (as would be the case for T0 < 273 K). We assume the atmosphere to be fully saturated in H2O until the tropopause, and we assumed the molar concentration of water in the stratosphere to be equal to its concentration at the tropopause throughout.
Our model requires temperature, pressure, and molar concentrations as functions of altitude. To obtain a mapping between pressure and altitude, we approximate the atmosphere as a series of 1000 layers, each individually isothermal, evenly spanning P0 − P0 × exp(−10). We then use the equation for an isothermal atmosphere in hydrostatic equilibrium,
P (z)/P (z0) = exp(−(z − z0)/H),
to calculate the change in altitude across each pressure layer, and sum to obtain a mapping between z and P . Here, P is pressure, z is the altitude of the layer top, z0 is the altitude of the layer bottom, and H = kT /(µg) is the scale height of the layer, with T being the layer temperature, µ the mean molecular weight of the atmospheric layer, and g the acceleration due to Martian gravity. We also considered numerically integrating the hydrostatic equilibrium equation,
dP/dz = −ρg,
where the density ρ = µP/(kT ) for an ideal gas, directly to obtain z(P ). We found this approach to agree within 1%; we consequently elected to use the simpler isothermal partition approach for our calculation. Appendix A presents sample atmospheric proﬁles derived using our methods.
3.3. Particulate Optical Parameters
In this section, we discuss our calculation of the optical parameters (σ, ω0, and g) associated with interaction of radiation with the CO2 and H2O ice particles that constitute clouds.
We approximate the particles as spherical, and compute their optical parameters using Mie theory at 0.1 nm resolution, following the treatment outlined in Hansen and Travis (1974). At each wavelength, we compute ω0, g, and the scattering eﬃciency Qs. We numerically integrate these parameters over a log normal size distribution with eﬀective radius reff and eﬀective variance veff , weighted by πr2n(r), where r is the particle radius and n(r)

– 10 –

is the size distribution, and demand a precision of 1% in the distribution-averaged mean values. We obtain the per-particle total extinction cross-section by

σ = (Qs/ω0)G,

where Qs is the distribution-averaged mean value of Qs, ω0 is the distribution-averaged mean

value of ω0, and

r2
G = πr2n(r)dr

r1

is the ”geometric cross-sectional area of particles per unit volume” (Hansen and Travis

1974) computed at 1 ppm precision. For our numerical integral, we integrated from r1 = reff 10−10veff to r2 = reff 104veff ; we found n(r) < 10−4 beyond these limits.

We take the index of refraction for H2O ice from the compendium of Warren and Brandt (2008) 3. We take the index of refraction for CO2 ice from the compendium given in Pierrehumbert (2010)4. This compendium was formed by G. Hansen by subjecting the absorption spectra of Hansen (1997) and Hansen (2005) to Kramers-Kronig analysis to obtain selfconsistent spectra of real and imaginary indices of refraction (S. Warren, private communication). We take the index of refraction for Martian dust from Wolﬀ et al. (2009) 5. The data of Wolﬀ et al. (2009) truncate at 263 nm. Measurements of the imaginary index of refraction down to 194 nm are available from Pang and Ajello (1977). We adopt the values of Pang and Ajello (1977) for wavelengths < 263 nm. Zurek (1978) present a compendia of the real index of refraction of Martian dust. Their study indicates that the index of refraction changes by only ∼ 0.1 from ∼ 200 − 263 nm. Their values at 263 nm are 0.4 higher than Wolﬀ et al. (2009); to avoid a discontinuity, we subtract 0.4 from the real indices of refraction of Zurek (1978).

Figure 1 presents the CO2 and H2O ice optical parameters as a function of wavelength for size distributions with veff = 0.1 and reff = 1, 10, 100 microns. Previous work has assumed CO2 cloud particle sizes to be in the 1-100 µm range (e.g., Forget and Pierrehumbert 1997). Microphysical modelling by Colaprete and Toon (2003) suggests that primitive Martian CO2 clouds may have been characterized by large particle sizes, as high as ref f = 100µm. The optical properties of CO2 and H2O ice particles for reff ≥ 10µm insensitive to reff ; we

3accessed via http://www.atmos.washington.edu/ice_optical_constants
4accessed via http://geosci.uchicago.edu/~rtp1/PrinciplesPlanetaryClimate/Data/ WorkbookDatasets/Chapter5Data/co2i4a.rfi.txt
5accessed via http://spacescience.arc.nasa.gov/mars-climate-modeling-group/documents/ Dust_Refractive_Indicies.txt

– 11 –

attribute this to size parameter x = 2πreff /λ being large in this regime, meaning such particles approach the large-particle limit. (x > 12 for reff ≥ 10µm and λ ≤ 500 nm). For wavelengths satisfying 195 < λ < 500 nm, CO2 and H2O ice are characterized by ω0 ≈ 1. This means that CO2 and H2O clouds do not signiﬁcantly absorb at wavelengths unshielded by H2O (< 198 nm) or CO2 (< 204 nm). By contrast, dust absorbs across 100 − 500 nm, meaning dust particles can supply absorption at wavelengths not shielded by CO2 or H2O.

Qext

ω0

2.4 2.2 2.0 1.8 1.6 1.4 1.2 1.0 1.0
0.9
0.8
0.7
0.6
0.5 1.0
0.9
0.8
0.7
0.6100

CO2 Ice

2.4 2.2 2.0 1.8 1.6 1.4 1.2 1.0 1.0

0.9

reff = 1µm vareff = 0. 1
reff = 10µm vareff = 0. 1
reff = 100µm vareff = 0. 1

0.8 0.7 0.6 0.5 1.0

H2O Ice

2.4 2.2 2.0 1.8 1.6 1.4 1.2 1.0 1.0

0.9

reff = 1µm vareff = 0. 1
reff = 10µm vareff = 0. 1
reff = 100µm vareff = 0. 1

0.8 0.7 0.6 0.5 1.0

0.9 0.9

0.8 0.8

0.7 0.7

300 5000.6100 300 5000.6100 Wavelength (nm)

Dust
reff = 1. 5µm vareff = 0. 5
300 500

g

Fig. 1.— Scattering eﬃciency Qext, single-scattering albedo ω0, and asymmetry parameter g as a function of wavelength for CO2 ice, H2O ice, and modern Martian dust, integrated over the speciﬁed log-normal size distributions.

3.4. Action Spectra and Calculation of Dose Rates
To quantify the impact of diﬀerent surface UV radiation environments on prebiotic chemistry, we follow the approach of Cockell (1999), Ronto´ et al. (2003), and Rugheimer et al. (2015) in computing Biologically Eﬀective Dose rates (BEDs). Speciﬁcally, we compute
λ1
D = ( dλA(λ)Isurf (λ)),
λ0
where A(λ) is an action spectrum, λ0 and λ1 are the limits over which A(λ) is deﬁned, and Isurf (λ) is the UV surface radiance. An action spectrum parametrizes the relative impact

– 12 –
of radiation on a given photoprocess as a function of wavelength, with a higher value of A meaning that a higher fraction of the incident photons are being used in the photoprocess. Hence, D is proportional to the reaction rate rate of a given photoprocess for a single molecule at the surface of a planet.
As D is a relative measure of reaction rate, a normalization is required to assign a physical interpretation to its value. In this paper, we report
D = D/D⊕,
where D⊕ is the dose rate on 3.9 Ga Earth. The atmospheric model for 3.9 Ga Earth is taken from Rugheimer et al. (2015), who use a 1D coupled climate-photochemistry model to compute the atmospheric proﬁle (T, P, composition) for the Earth at 3.9 Ga, assuming modern abiotic outgassing rates and a background atmosphere of 0.9 bar N2, 0.1 bar CO2, with SZA=60◦ and A = 0.2. Consequently, D > 1 means that the photoprocess is proceeding faster on the Martian surface under the speciﬁed atmosphere than it would on the surface of the Rugheimer et al. (2015) ﬁducial Earth. Note this normalization is diﬀerent from what we chose in Ranjan and Sasselov (2017), because here we are trying to assess how Mars compares to the Earth as a venue for prebiotic chemistry.
Previous workers used action spectra of UV stress on modern biology (e.g. the DNA inactivation action spectrum) (Cockell 2000, 2002; Cnossen et al. 2007; Rugheimer et al. 2015) as a gauge of the level of stress imposed by UV ﬂuence on the prebiotic environment. However, these action spectra are based on modern life. Modern organisms have evolved sophisticated methods to deal with environmental stress, including UV exposure, that would not have been available to the ﬁrst life. Further, this approach presupposes that UV light is solely a stressor, and ignores its potential role as a eustressor for abiogenesis. In this work, we follow the reasoning of our previous eﬀorts in Ranjan and Sasselov (2017) in formulating action spectra corresponding to simple photoreactions that are expected to have played major roles in prebiotic chemistry. We consider two reactions: a stressor process, to capture the stress UV light places on nascent biology, and an eustressor process, to capture the role of UV light in promoting prebiotic chemistry. A detailed description of these processes and their corresponding action spectra is given in Ranjan and Sasselov (2017); a brief outline is presented below.
3.4.1. Stressor Process: Cleavage of N-Glycosidic Bond of UMP
For our stressor process, we chose the cleavage of the N-glycosidic bond in the RNA monomer uridine monophospate (UMP). UV radiation can cleave the N-glyocosidic bond

– 13 –
which joins the sugar to the nucleobase (Gurzadyan and G¨orner 1994), irreversibly destroying this molecule. Hence, this process represents a stressor to abiogenesis.
The action spectrum is equal to the product of the absorption spectrum (fraction of incident photons absorbed) and the quantum yield curve (QY, fraction of absorbed photons that lead to the photoreaction). We take our UMP absorption spectrum from the work of (Voet et al. 1963) (pH=7.6). Detailed spectral measurements of the QY of glyocosidic bond cleavage have not been obtained. However, Gurzadyan and Go¨rner (1994) found the QY of N-glycosidic bond cleavage in UMP in neutral aqueous solution saturated with Ar (i.e. anoxic) to be 4.3 × 10−3 at 193 nm and (2 − 3) × 10−5 for 254 nm. We therefore represent the QY curve as a step function with value 4.3 × 10−3 for λ ≤ λ0 and 2.5 × 10−5 for λ > λ0. We consider λ0 values of 193 and 254 nm, corresponding to the empirical limits from Gurzadyan and G¨orner (1994). We also consider λ0 = 230 nm, which corresponds to the end of the broad absorption feature centered near 260 nm corresponding to the π − π∗ transition and also to the transition to irreversible decomposition suggested by Sinsheimer and Hastings (1949). As shorthand, we refer to this photoprocess under the assumption that λ0 =X nm by UMP-X. Figure 2 presents these action spectra.
The absorption spectra of the other RNA monomers are structurally similar to UMP (Voet et al. 1963), and the quantum yield of N-glycosidic bond cleavage in adenosine monophospate (AMP) increases at short wavelengths like UMP’s does (Gurzadyan and Go¨rner 1994), leading us to argue that action spectra for N-glycosidic bond cleavage of the other RNA monomers should be broadly similar to that for UMP. Therefore, results derived using the action spectrum for UMP N-glycosidic bond cleavage should be broadly applicable to the other RNA monomers: if a UV environment is destructive for UMP, it should be bad for the other RNA monomers, and hence for abiogenesis in the RNA world hypothesis, as well.
3.4.2. Eustressor Process: Production of Aquated Electrons from Photoionization of Cyanocuprate
For our eustressor process, we choose the production of aquated electrons from the irradiation of a tricyanocuprate (CuCN32−) complex. We chose this process because it underlies the selective 2- and 3-carbon sugar (glycolaldehyde and glyceraldehyde) synthesis pathway of Ritson and Sutherland (2012), which is the best candidate proposed so far for a selective prebiotic synthesis of these sugars. These sugars are required for the synthesis of RNA, and hence abiogenesis in the RNA world hypothesis. This process is also important to the prebiotic reaction network of Patel et al. (2015). More generally, aquated electrons are useful for a broad range of reductive prebiotic chemistry, e.g., the reduction of nitriles to amines, alde-

– 14 –
hydes to hydroxyls, and hydroxyls to alkyls6. Therefore, this process represents a eustressor to abiogenesis. While other UV-sensitive processes conducive to abiogenesis doubtless exist, we argue this process is of particular interest because of its unique role in the most promising plausibly prebiotic pathways to the RNA monomers.
We again form the action spectrum by multiplying the absorption spectrum and the quantum yield curve. We take the cyanocuprate absorption spectrum from the work of Magnani (2015), via Ranjan and Sasselov (2016). The spectral QY of aquated electron production from cyanocuprate irradiation is not known. However, Horv´ath et al. (1984) measure a QY of 0.06 for this process at 254 nm. Following Ritson and Sutherland (2012)’s hypothesis that photoionization of the complex drives aquated electron production, we assume the QY to be characterized by a step function with value 0.06 for λ ≤ λ0 and 0 otherwise. We empirically know λ0 ≥ 254 nm. To explore a range of λ0, we consider λ0 = 254 nm and λ0 = 300 nm. As shorthand, we refer to this photoprocess under the assumption that λ0 =Y nm by CuCN3-Y. Figure 2 presents these action spectra.
Action spectra typically encode information about relative, not absolute, reaction rates. Consequently, they are generally arbitrarily normalized to 1 at some wavelength (see, e.g., Cockell 1999 and Rugheimer et al. 2015). We normalize these spectra to 1 at 190 nm.
6J. Szostak, private communication, 2/5/16

– 15 –

Relative Sensitivity

102 Action Spectra

101

UMP UMP UMP

Gly Gly Gly

Bond Bond Bond

Cleavage Cleavage Cleavage

(((λλλ000

= = =

193) 230) 254)

100

CCuuCCNN33

2 2

− −

Photoionization Photoionization

((λλ00

= 254) = 300)

10-1

10-2

10-3

10-4

10-5

10-1680 200 220 240 260 280 300 320 340 360 nm

Fig. 2.— Action spectra for photolysis of UMP−λ0 and photoionization of CuCN3 −λ0, assuming a step-function form to the QE for both processes with step at λ0. The spectra are arbitrarily normalized to 1 at 190 nm.

4. Results
4.1. Clear-Sky H2O-CO2 Atmospheres
We evaluated the UV surface radiance for a range of (pCO2 , T0) for pure H2O-CO2 atmospheres in the clear-sky case (no clouds, dust or other particulates). We considered pCO2 = 0.02 − 2 bar, corresponding to the range of surface pressure for which Wordsworth et al. 2013 reported at least transient local temperatures above 273 K, and T0 = 210 − 300 K. We took an SZA of 0, corresponding to noon at equatorial latitudes. We took the surface albedo to correspond to desert (diﬀuse albedo of 0.22); we adopted this albedo because 1) young Mars is thought to have been dry and desertlike in conventional climate models, and 2) the desert diﬀuse albedo corresponds roughly to modern Mars’s surface albedo (e.g., Kasting 1991). We note that variations in surface albedo and SZA can drive variations in the spectral surface radiance of up to a factor of ∼ 20, and overall variations in prebiotically-relevant reaction rates of a factor of 10 (Ranjan and Sasselov 2017). These surface radiances are shown in Figure 3.

– 16 –

Integrated Intensity (photons/s/cm2/nmI)ntegrated Intensity (photons/s/cm2/nmI)ntegrated Intensity (photons/s/cm2/nm)

1015 pCO2=0.02bar

1014

1013

1012 TOA

1011 1010

210K 250K 300K

109

1015 pCO2=0.2bar
1014 1013 1012 1011 1010 109

1015 pCO2=2bar
1014 1013 1012 1011 1010 101900 150 200 250 300 350 400 450 500
Wavelength (nm)
Fig. 3.— Surface radiance as a function of wavelength for varying pCO2 and T0, for SZA=0 and albedo corresponding to desert. Also plotted for scale is the TOA solar ﬂux. The surface radiance is insensitive to atmospheric and surface temperature.
In the scattering regime (λ > 204 nm), our surface radiances fall oﬀ only slowly with pCO2. This is a consequence of random walk statistics in the context of multiple scattering: transmission through purely scattering media go as 1/τ (see, e.g., Bohren 1987 for a discussion with application to clouds). This result stands in contrast to the calculations of Cnossen et al. (2007) and Ronto´ et al. (2003), who ignore multiple scattering in their ra-

– 17 –
diative transfer treatments, and illustrates the importance of self-consistently including this phenomenon when considering dense, highly scattering atmospheres.
We ﬁnd our surface radiance calculations to be insensitive to T0. This is because 1) the total atmospheric column is set by P0 and independent of T0, 2) the rapid increase of CO2 cross-sections for λ < 204 nm, which means that the the atmosphere rapidly becomes optically thick in the UV, and 3) increased water vapor abundance with increasing T0 does not drive an increase in opacity because water vapor absorption is degenerate with CO2 absorption in the UV (Ranjan and Sasselov 2017).
We considered the hypothesis that including the eﬀect of variations in CO2 cross-section with temperature might impact our results. We followed the approach of Hu et al. (2012) in interpolating between cold (∼ 195 K) and room-temperature datasets for CO2 absorption to estimate the eﬀects of temperature dependence on CO2 cross-section. We used the dataset of Stark et al. (2007) from 106.5-118.7 nm, Yoshino et al. (1996) from 118.7-163 nm, and Parkinson et al. (2003) from 163-192.5 nm. We did not ﬁnd cold-temperature crosssections for CO2 at longer wavelengths. We found our results were not altered by including temperature-dependence of CO2 cross-sections for pCO2 = 0.02 − 2 bar, because the CO2 UV absorption is already saturated by 192.5 nm, where our temperature dependence kicks in. We considered lower values of pCO2 = 2 × 10−3 − 2 × 10−5 bar (below PAL), where the CO2 absorption does not saturate until wavelengths shorter than 192.5 nm. Even in the low pCO2 case, including temperature-dependence only changed the onsite of CO2 absorption saturation by 1 − 2 nm. We attribute this to the rapidity of the rise in CO2 absorption cross section with decreasing wavelength for λ 204 nm. We conclude that even including the eﬀects of temperature on CO2 UV cross-section, the UV surface ﬂuence is insensitive to T0. We consequently elected to ignore the temperature dependence of CO2 cross-sections in the remainder of this study.
Figure 4 presents the surface radiances in the pCO2 = 2 × 10−5 bar cases, calculated for SZA=0, A corresponding to desert, and T0 = Teq ≈ 200K, with and without CO2 crosssections included. Such low atmospheric pressures have been suggested based on atmospheric escape arguments (Tian et al. 2009), followed by a buildup of the atmosphere after escape rates subsided with shortwave solar output. In such a case, aqueous prebiotic chemistry could only have proceeded in environments kept warm by non-climatological means, e.g. geothermal reservoirs. Even for such low pCO2, EUV ﬂuence shortward of 185 nm is shielded out by atmospheric CO2.

– 18 –

Integrated Intensity (photons/s/cm2/nm)

1015

1014

1013

1012

1011

1010

TOA No Temp Dependence

Temp Dependence

101900 150 200 250 300 350 400 450 500

Wavelength (nm)

Fig. 4.— Surface radiance as a function of wavelength for pCO2 = 2 × 10−5 bar , T0 = 200K, SZA=0, A corresponding to desert, with and without temperature dependence of the CO2 cross-sections included. Also plotted for scale is the TOA solar ﬂux. The surface radiance is insensitive to inclusion of temperature dependence of the CO2 cross-sections, even for very low pCO2.

4.2. Eﬀect of CO2 and H2O Clouds
We considered the eﬀect of CO2 and H2O clouds in a CO2-H2O atmosphere. Such clouds have been detected on modern Mars (see, e.g., Vincendon et al. 2011), and GCM results suggest they should have been present on early Mars as well(Wordsworth et al. 2013). Figure 5 presents the UV surface ﬂuence for a 0.02-bar CO2-H2O atmosphere with H2O and CO2 cloud decks of varying optical depths emplaced in the atmosphere. This low surface pressure is chosen in order to isolate the eﬀects of the clouds as opposed to atmospheric CO2. The surface albedo corresponds to desert, and SZA=0. The optical depths are speciﬁed at 500 nm. The H2O and CO2 cloud decks are emplaced from 3-4 km and 20-21 km of altitude, respectively, corresponding approximately to the altitudes of peak cloud formation identiﬁed in Wordsworth et al. (2013). For both types of clouds, we varied the cloud deck altitudes between 0.5-60.5 km, and found the surface ﬂuence to be insensitive to the cloud deck altitude. We also experimented with partitioning the clouds into two decks, and found the surface ﬂuence to be insensitive to the partition.

– 19 –

Integrated Intensity (photons/s/cm2/nm)

1015 TOA

CO2 Cloud Deck

Cloud OD=0.1

1014

Cloud OD=1 Cloud OD=10

Cloud OD=100

1013

Cloud OD=1000 Cloud OD=10000

1012

1011

1010

109

108

1015 H2O Cloud Deck

TOA

Cloud OD=0.1

1014

Cloud OD=1 Cloud OD=10

Cloud OD=100

1013

Cloud OD=1000 Cloud OD=10000

1012

1011

1010

109

101800 150 200 250 300 350 400 450 500 Wavelength (nm)

Integrated Intensity (photons/s/cm2/nm)

Fig. 5.— Surface radiance as a function of wavelength for pCO2 = 0.02 bar, T0 = 250K, SZA=0, A corresponding to desert, and CO2 and H2O cloud decks of varying thicknesses inserted from 20-21 and 3-4 km respectively. Also plotted for scale is the TOA solar ﬂux.

CO2 and H2O ice clouds have similar impact on the surface ﬂuence. This is because both types of ice have similar optical parameters for 195 < λ < 500 nm and reff ≥ 1µm.

– 20 –
The λ < 195 nm regime, where they have diﬀerent optical parameters, is shielded out by gaseous CO2 absorption.
CO2 and H2O ice particles are pure scatterers for 195 < λ < 500 nm and reff ≥ 1µm. Consequently, it is unsurprising that surface ﬂuence falls oﬀ only slowly with increasing optical depth. In fact, the transmission of a purely scattering cloud layer varies as ∼ 1/τ ∗, where τ ∗ is the delta-scaled optical depth of the cloud layer (Bohren 1987). This is a consequence of the random-walk nature of radiative transfer in the optically-thick purelyscattering limit.
The particle size reff can have an impact on surface radiance. For reff = 1µm, the surface ﬂuence at 500 nm is ∼ 40% lower compared to reff = 100µm ( A corresponding to desert, SZA=0). This is because as reff decreases in this regime, so does g, meaning that the rescaled optical depth in the delta-scaling formalism τ ∗ = τ × (1 − g2) is higher for small particles than large particles.
Suppressing surface radiance to 10% or less of TOA ﬂux requires cloud optical depths of 100. This is comparable to the optical depth of a terrestrial thunderstorm (Mayer et al. 1998). We can compute the mass column of ice particle required to achieve this optical depth by
u = ( τ )( 4π r3ρ), πr2Qext 3
where ρ is the mass density of the ice, r is the ice particle radius, τ is the cloud optical depth, and Qext is the extinction eﬃciency. For CO2 ice, ρ = 1.5 g/cm3 7, r = 10µm, and approximating Qext = 2, we ﬁnd τ = 100 corresponds to u = 1 kg/m2. Wordsworth et al. (2013) ﬁnd in their 3D simulation CO2 ice columns of up to 0.6 kg/m2 in patches, suggesting CO2 clouds may have signiﬁcantly aﬀected the surface UV environment. For water ice, taking ρ = 0.92 g/cm3 (Miller 2009), r = 10µm, and approximating Qext ≈ 2, we ﬁnd τ = 100 corresponds to u = 6 × 10−1 kg/m2. By contrast, Wordsworth et al. (2013) ﬁnds expected ice columns of 2 × 10−3 kg/m2. Consequently, barring a mechanism which can increase the H2O cloud levels above that considered by Wordsworth et al. (2013), H2O ice clouds by themselves are unlikely to signiﬁcantly alter the surface radiance environment on early Mars.
7http://terpconnect.umd.edu/~choi/MSDS/Airgas/CARBON%20DIOXIDE.pdf

– 21 –
4.3. Eﬀect of Elevated Levels of Volcanogenic Gases
So far, we have considered pure CO2-H2O atmospheres. However, other gases may have been present in the early Martian atmosphere. In particular, Mars at ∼ 3.9 Ga was characterized by volcanic activity, which emplaced features like the Tharsis igneous province (Halevy and Head 2014). Such volcanism could have injected elevated levels of volcanogenic gases like SO2 and H2S into the atmosphere. SO2 and H2S are also strong and broad UV absorbers, and at elevated levels they can completely reshape the surface UV environment (Ranjan and Sasselov 2017). Consequently, we consider the impact of elevated levels of SO2 and H2S on the surface UV environment.
We consider SO2 levels up to 2 × 10−5 bar. For scale, Halevy and Head (2014) compute that an SO2 level of 1 × 10−5 bar in a 1 bar CO2-dominated atmosphere requires volcanic outgassing at 100× the current terrestrial outgassing. Figure 6 presents the surface radiance for varying pSO2 and pCO2 = 0.02−2 bar (SZA=0, A corresponding to desert, and T0 = 250 K).

– 22 –
Fig. 6.— Surface radiance as a function of wavelength for T0 = 250K, SZA=0, A corresponding to desert, pCO2 = 0.02 − 2 bar, and pSO2 = 2 × 10−9 − 2 × 10−5 bar. Also plotted for scale is the TOA solar ﬂux.
Note that for the same pSO2, the surface radiance varies as a function of pCO2. This is because at high pCO2, the scattering optical depth of the atmosphere exceeds unity. In such a regime, the impact of trace absorbers is ampliﬁed due to higher eﬀective path length caused by multiple scattering events (Bohren 1987). This eﬀect has been seen in studies of UV transmittance through thick clouds on Earth as well (Mayer et al. 1998). Consequently,

– 23 –
the surface UV environment is a feature of both pCO2 and pSO2. For pSO2 ≥ ×10−5 bar, UV ﬂuence < 330 nm is strongly suppressed.
H2S is also a major volcanogenic gas that may have been emitted at rates greater than or equal to SO2 on young Mars, based on studies of the oxidation state of Martian basalts (Herd et al. 2002; Halevy et al. 2007). While Halevy and Head (2014) do not calculate the H2S abundance as a function of volcanic outgassing, the calculations of Hu et al. (2013) (T0=288K, modern solar irradiance) suggest that pH2S>pSO2 in a CO2 dominated atmosphere. Consequently, we also consider the impact of elevated levels of H2S, up to pH2S=2 × 10−4 bar in atmospheres with pCO2 = 0.02 − 2 bar. Figure 7 presents the surface radiance calculated over this range of atmospheres (SZA=0, A corresponding to desert, and T0 = 250 K). For pH2S ≥ ×10−4 bar, UV ﬂuence < 370 nm is strongly suppressed.

– 24 –

Integrated Intensity (photons/s/cm2/nm)

pCO2=0.02 barTOA
pH2S = 2 × 10−9 bar pH2S = 2 × 10−8 bar

pH2S = 2 × 10−7 bar pH2S = 2 × 10−6 bar

pH2S = 2 × 10−5 bar pH2S = 2 × 10−4 bar

1015

1014

1013

1012

1011

1010

109

108

107

106

1015 pCO2=0.2 bar
1014 1013 11001112 1010 109 108 107 106

1015 pCO2=2 bar
1014 1013 1012 1011 1010 109 108 107 101600 150 200 250 300 350 400 450 500
Wavelength (nm)

Fig. 7.— Surface radiance as a function of wavelength for T0 = 250K, SZA=0, A corresponding to desert, pCO2 = 0.02 − 2 bar, and pH2S= 2 × 10−9 − 2 × 10−4 bar. Also plotted
for scale is the TOA solar ﬂux.

Scattering due to thick cloud decks can also enhance absorption by trace pSO2 and pH2S. For pSO2 = 2 × 10−7 bar and pCO2=0.02 bar8 (SZA=0, A corresponding to new snow

8At pCO2=0.02 bar, the gaseous scattering optical depth is less than unity for wavelengths longer than 204 nm, meaning that we can attribute this ampliﬁcation primarily to the cloud deck (as opposed to gaseous

– 25 –
(near 1), T0 = 250 K, reff = 10µm), inclusion of CO2 clouds with optical depth of 1000 ampliﬁes attenuation by a factor of 10 at 236.5 nm (optically thin in SO2 absorption) and by a factor of 105 at 281.5 nm (optically thick in SO2 absorption) compared to what one would calculate by multiplying the transmission from the SO2 and cloud deck individually. This eﬀect is weaker for low-albedo cases, because there are fewer passes of radiation through the atmosphere due to bouncing between the surface and cloud deck. This nonlinear variance in transmission due to multiple scattering eﬀects illustrates the need to specify surface albedo, cloud thickness and absorber level when calculating surface radiances. Figure 8 and Figure 9 present the surface radiance at the base of a pCO2=0.02 bar atmosphere with varying levels of SO2 and H2S respectively, with CO2 cloud decks of optical depths 1 − 1000 emplaced from 20-21 km (SZA=0, A corresponding to desert, T0 = 250 K, reff = 10µm).
scattering).

– 26 –

Integrated Intensity (photons/s/cm2/nm)

111111111111000000000000111111456789012345 111111111111000000000000111111456789012345 111111111111000000000000111111456789012345 1111111111110000000000001111111456789012345 00

TOA pSO2 = 2 × 10−9 bar

pSO2 = 2 × 10−8 bar pSO2 = 2 × 10−7 bar
τcloud = 1

τcloud = 10

τcloud = 100

τcloud = 1000 150 200 250 300 350
Wavelength (nm)

pSO2 = 2 × 10−6 bar pSO2 = 2 × 10−5 bar
400 450 500

Fig. 8.— Surface radiance as a function of wavelength for T0 = 250K, SZA=0, A corresponding to desert, pCO2 = 0.02 bar, pSO2 = 2 × 10−9 − 2 × 10−5 bar, and CO2 cloud decks of varying optical thickness emplaced from 20-21 km altitude. Also plotted for scale is the
TOA solar ﬂux.

– 27 –

Integrated Intensity (photons/s/cm2/nm)

111111111111000000000000111111456789012345 111111111111000000000000111111456789012345 111111111111000000000000111111456789012345 1111111111110000000000001111111456789012345 00

TOA pH2S = 2 × 10−9 bar pH2S = 2 × 10−8 bar

pH2S = 2 × 10−7 bar pH2S = 2 × 10−6 bar
τcloud = 1

τcloud = 10

τcloud = 100

τcloud = 1000 150 200 250 300 350
Wavelength (nm)

pH2S = 2 × 10−5 bar pH2S = 2 × 10−4 bar
400 450 500

Fig. 9.— Surface radiance as a function of wavelength for T0 = 250K, SZA=0, A corresponding to desert, pCO2 = 0.02 bar, pH2S= 2 × 10−9 − 2 × 10−4 bar, and CO2 cloud decks of varying optical thickness emplaced from 20-21 km altitude. Also plotted for scale is the
TOA solar ﬂux.

4.4. Eﬀect of Dust
The atmosphere of modern Mars is dusty, with typical dust optical depths varying from τd ∼ 0.2 − 2 at solar wavelengths (Smith et al. 2002; Lemmon et al. 2015), with the higher

– 28 –
values achieved during global dust storms. Mars’s dryness contributes to its dustiness, through the availability of desiccated surface to supply dust and the lack of a hydrologic cycle to quickly scrub it from the atmosphere. If one assumes that Mars were similarly dry in the past, dust levels in the Martian atmosphere may have been signiﬁcant. We may speculate that if the atmosphere were thicker, it could have hosted even more dust than the modern Martian atmosphere due to slower sedimentation times. However, detailed study of atmospheric dust dynamics, including analysis of how dust lofting scales with pCO2, is required to constrain this possibility. Dust absorbs at UV wavelengths (see Figure 1), so dust could play a role similar to volcanogenic gases in scrubbing UV radiative from the UV surface environment.
We explored the impact of including dust in our calculation of surface UV ﬂuence. We assumed that, similar to the modern Mars, the dust followed an exponential proﬁle with scale height height Hd = 11 km, similar to the atmospheric pressure scale height (Hoekzema et al. 2010; Mishra et al. 2016). For surface temperatures (and hence scale heights) comparable to modern Mars, this corresponds to the assumption by previous workers that the dust mixing ratio is constant in the lower atmosphere (Forget et al. 1999). Then, the dust optical depth across each atmospheric layer of width ∆z is
C exp [−z/Hd],
where z is the altitude of the layer center. The parameter
C = τd(exp [∆z/(2Hd)])(1 − exp [−∆z/Hd])
, is chosen such that the column-integrated dust optical depth is τd.
Figure 10 presents the surface radiance for varying τd and pCO2 = 0.02 − 2 bar (SZA=0, A corresponding to desert, and T0 = 250 K). As for SO2 and H2S, highly scattering atmospheres amplify the impact of trace absorbers. τd = 1 only marginally suppresses UV ﬂuence for pCO2 = 0.02 bar, but for pCO2 = 2 bar shortwave ﬂuence is suppressed due to enhanced Rayleigh scattering. In the absence of scattering ampliﬁcation, τd 10 is required to strongly suppress UV ﬂuence.

– 29 –

TOA
τd = 0. 1
1015 1014 1013 1012 1011 1010 109 108 107 106 105

pCO2=0.02 bar

τd = 1 τd = 10

Integrated Intensity (photons s−1cm−2 nm−1)

1015 pCO2=0.2 bar
1014 1013 1012 1011 1010 109 108 107 106 105

1015 pCO2=2 bar
1014 1013 1012 1011 1010 109 108 107 106 1015 00 150 200 250 300 350 400 450 500
Wavelength (nm)
Fig. 10.— Surface radiance as a function of wavelength for T0 = 250K, SZA=0, A corresponding to desert, pCO2 = 0.02 − 2 bar, and τd = 0.1 − 10. Also plotted for scale is the TOA solar ﬂux.
Figure 11 presents the surface radiance at the base of a pCO2=0.02 bar atmosphere with varying τd, with CO2 cloud decks of optical depths 1 − 1000 emplaced from 20-21 km (SZA=0, A corresponding to desert, T0 = 250 K, reff = 10µm). The impact of a cloud deck is less than the impact of increasing overall atmospheric pressure, because of the limited column of absorber contained within the cloud itself. If the cloud deck is extended, then the

– 30 –

surface ﬂuence is reduced (for the same total cloud and dust optical depth).

IntensityIntegrated (photons s−1cm−2 nm−1)

111111111110000000000011111156789012345
111111111110000000000011111156789012345
15
10
14
10
13
10
12
10
11
10
10
10
9
10
8
10
7
10
6
10
5
10
15
10
14
10
13
10
12
10
11
10
10
10
9
10
8
10
7
10
6
10
5
10 100

TOA
τd = 0. 1

150

200

τcloud = 1

τcloud = 10

τcloud = 100

τcloud = 1000

250

300

350

Wavelength (nm)

400

τd = 1 τd = 10

450

500

Fig. 11.— Surface radiance as a function of wavelength for T0 = 250K, SZA=0, A corresponding to desert, pCO2 = 0.02 bar, τd = 0.1 − 10, and CO2 cloud decks of varying optical thickness emplaced from 20-21 km altitude. Also plotted for scale is the TOA solar ﬂux.

– 31 –
5. Discussion
5.1. CO2-H2O Atmosphere
Normative CO2-H2O climate models predict the steady-state early Martian climate to have been cold, with global mean temperatures below freezing (Forget et al. 2013; Wordsworth et al. 2013; Wordsworth et al. 2015). In this scenario, aqueous prebiotic chemistry would proceed in meltwater pools, which could have occurred transiently during midday due to the diurnal cycle, during summer during a seasonal cycle, or a combination of both. Aqueous prebiotic chemistry could also have proceeded in geothermally heated pools, which may have been abundant during the more volcanically active Noachian.
Figure 12 presents the dose rates Di corresponding to irradiation of prebiotically relevant molecules through a clear-sky CO2-H2O atmosphere with SZA=0 and A corresponding to desert. The dose rates Di decline by only an order of magnitude across 5 orders of magnitude of pCO2 ≤ 2 bar, meaning that the prebiotic photochemistry dose rates are only weakly sensitive to pCO2 across this range. Di is within an order of magnitude of unity across this range, meaning that the Martian dose rates are comparable to the terrestrial rates. This is consistent with the results of Cockell (2000), who found that UV stress as measured by DNA damage were comparable for 3.5 Ga Earth and Mars.

– 32 –

Relative Dose Rate ¯Di

UMP UMP UMP

Bond Bond Bond

Cleavage Cleavage Cleavage

(((λλλ000

= = =

193) 230) 254)

3.5

CCuuCCNN33

Photoionization Photoionization

((λλ00

= =

254) 300)

3.0

2.5

2.0

1.5

1.0

0.5

0.100-5 10-4 10-3 10-2 10-1 100 101 pCO2 (bar)

Fig. 12.— UV dose rates Di for clear-sky CO2-H2O atmospheres as a function of pCO2 (T0 = 250K, SZA=0, A corresponding to desert).

However, the sky was not necessarily clear during this epoch. In fact, formation of H2O and CO2 clouds are expected based on GCM studies (Wordsworth et al. 2013). Thick (though patchy) CO2 cloud decks in particular are expected for thick CO2 atmospheres. Figure 13 presents the dose rates Di as a function of CO2 cloud optical depth for pCO2=0.02 bar, SZA=0, and A corresponding to desert. As we might expect from Figure 5, the dose rate drops oﬀ only modestly with τcloud. τcloud 1000 is required to suppress dose rates by more than an order of magnitude. For reff = 10µm, this corresponds to 20× the maximum cloud column calculated by Wordsworth et al. (2013). Overall, the impact of clouds on their own on UV-sensitive photochemistry is expected to be modest.

– 33 –

Relative Dose Rate ¯Di

UMP UMP UMP

Bond Bond Bond

Cleavage Cleavage Cleavage

(((λλλ000

= = =

193) 230) 254)

CCuuCCNN33

Photoionization Photoionization

((λλ00

= =

254) 300)

101

100

10-1

10-2

10-3

10-140-1 100 101 102 103 104 τcloud (unscaled)

Fig. 13.— UV dose rates Di for CO2-H2O atmospheres with a CO2 cloud deck emplaced from 20-21 km altitude, as a function of cloud deck optical depth (T0 = 250K, pCO2=0.02 bar, SZA=0, A corresponding to desert).

If young Mars were indeed cold and dry, then one might expect it to have been dusty, as it is today. The modern Martian dust optical depth ranges from τd ∼ 0.2 − 2, with higher values achieved during dust storms (Smith et al. 2002; Lemmon et al. 2015). Higher dust loadings may have been possible when the atmosphere was thicker. Figure 14 presents the dose rates as a function of τd and pCO2, for SZA=0 and A corresponding to desert. Figure 15 presents the dose rates as a function of τd and CO2 cloud optical depth, for SZA=0 and A corresponding to desert.
Unlike CO2 and H2O ice and gas, dust absorbs across the UV waveband. Further, this absorption can be dramatically increased in highly scattering atmospheres due to enhanced eﬀective path length due to multiple scattering eﬀects. Consequently, dust can dramatically suppress UV ﬂuence and hence photochemistry, especially for thick and/or cloudy atmospheres. For τd ≥ 10, dose rates are suppressed by 2-8 orders of magnitude, depending on atmospheric pressure and cloud thickness. Dust of thickness τd = 1 provides minimal suppression for pCO2 ≤0.2 bar or τcloud < 100, but for pCO2 ≥2 bar or τcloud ≥ 1000, can suppress dose rates by orders of magnitude. Dust of thickness τd ≤ 0.1 does not signiﬁcantly alter dose rates across the explored parameter space. Overall, dust can signiﬁcantly alter

– 34 –

dose rates if it is present at high levels (greater than that seen for modern Martian dust storms) or if it is present at levels comparable to the modern average, but embedded in a thick atmosphere or underlying thick clouds.

Relative Dose Rate ¯Di

Relative Dose Rate ¯Di

UMP Bond Cleavage (λ0 = 193) UMP Bond Cleavage (λ0 = 230) UMP Bond Cleavage (λ0 = 254)

CuCN3 Photoionization (λ0 = 254) CuCN3 Photoionization (λ0 = 300)

101 pCO2=0.02 bar

100

10-1

10-2
101
100
10-1
10-2
10-3
100 10-1 10-2 10-3 10-4 10-5 10-6 10-7 10-180-1

pCO2=0.2 bar
pCO2=2 bar
100 τd (unscaled)

101

Relative Dose Rate ¯Di

Fig. 14.— UV dose rates Di for dusty CO2-H2O atmospheres as a function of pCO2 and τd (T0 = 250K, SZA=0, A corresponding to desert).

– 35 –

Relative Dose Rate ¯Di Relative Dose Rate ¯Di Relative Dose Rate ¯Di Relative Dose Rate ¯Di

UMP Bond Cleavage (λ0 = 193) UMP Bond Cleavage (λ0 = 230) UMP Bond Cleavage (λ0 = 254)

CuCN3 Photoionization (λ0 = 254) CuCN3 Photoionization (λ0 = 300)

101 τcloud = 1 (unscaled)

100

10-1

10-2
101 100 10-1 10-2 10-3
100 10-1 10-2 10-3 10-4
10-1 10-2 10-3 10-4 10-5 10-160-1

τcloud = 10 (unscaled)
τcloud = 100 (unscaled)
τcloud = 1000 (unscaled)
100 τd (unscaled)

101

Fig. 15.— UV dose rates Di for dusty CO2-H2O atmospheres with a CO2 cloud deck emplaced, as a function of τd and τcloud (T0 = 250K, pCO2=0.02, SZA=0, A corresponding to desert).

We considered the hypothesis that attenuation due to dust might diﬀerentially aﬀect the stressor and eustressor pathways. That is, we considered the possibility that the eustressor reaction rates might fall oﬀ faster (or slower) than the stressor reaction rates with attenuation due to dust. Since the stressor pathway measures destruction of RNA monomers and the eustressor pathway measures a process important to the synthesis of key RNA precursors,

– 36 –
this means that environments that favor the eustressor pathway over the stressor pathway are a more favorable venue for abiogenesis than the reverse, in the RNA world hypothesis. This argument assumes that these particular stressor and eustressor processes were important in the prebiotic world. They might not have been. However, we expect other photochemical stressor and eustressor processes to behave in generally similar ways to these processes. For example, we generally expect the quantum yield of prebiotic molecular destruction to decrease with increasing wavelength because of decreased photon energy compared to bond strength. Similarly, regardless of the solvated electron source (e.g. HS− as opposed to tricyanocuprate), we expect the quantum yield to go approximately as a step function in wavelength. We therefore suggest that results derived from these pathways may generalize to other processes, though a detailed comparison is required to rule on this hypothesis.
To assess the hypothesis that a dusty Mars might be less (or more) clement for abiogenesis than a non-dusty Mars as measured by our stressor (UMP-X) and eustressor (CuCN3-Y) pathways, we calculated DUMP −X/DCuCN3−Y . We calculated this quantity for pCO2 = 2 bar (no clouds) and τcloud = 1000 (pCO2=0.02 bar) for τd = 0.1 − 10. If these ratios rise with τd, it means that the stressor pathway is relatively favored by dusty atmospheres; if they fall, it means that the stressor pathway is relatively disfavored by dusty atmospheres.
Figure 16 presents these calculations. Dust attenuation on its own is relatively ﬂat at CO2-scattering wavelengths, as is cloud attenuation. Consequently dusty/cloudy atmospheres tend to reduce UV ﬂuence in a spectrally ﬂat manner, and favor neither the stressor nor the eustressor pathway. On the other hand, in a thick CO2 atmosphere, the scattering optical depth, and hence ampliﬁcation of dust absorption, increases as wavelength decreases. Consequently, the dose rate ratio does change with increasing τd. However, the direction of the change is sensitive to the value of λ0, the ionization threshold for the tricyanocuprate ionization process, and the magnitude is further sensitive to the value of λ0 for the UMP cleavage process. We consequently conclude it is possible that thick, dusty atmospheres might be more or less clement for abiogenesis than non-dusty atmospheres, but determining which requires wavelength-dependent measurements of the QYs of these chemical processes in the laboratory.

– 37 –

UMP-193/CuCN3-254 UMP-230/CuCN3-254 UMP-254/CuCN3-254

UMP-193/CuCN3-300 UMP-230/CuCN3-300 UMP-254/CuCN3-300

τ103 pCO2=2 bar, cloud = 0

102

101

100

10-1

10-2
101 pCO2=0.02 bar, τcloud = 1000 (unscaled)

¯DUMP−X/ ¯DCuCN3−Y

¯DUMP−X/ ¯DCuCN3−Y

100

10-110-1

τd

100 (unscaled)

101

Fig. 16.— Ratio of stressor dose rates UMP-X divided by eustressor dose rates UMP-Y for dusty CO2-H2O atmospheres, as a function of τd (T0 = 250K, SZA=0, A corresponding to desert). The atmospheres are highly scattering, either because of high pCO2 or thick CO2 clouds. Lower ratios imply a more favorable environment for abiogenesis as measured by these two photoprocesses.

– 38 –
5.2. Highly Reducing Atmospheres
Recent work suggests that if the reduced gases H2 and/or CH4 were present at elevated levels in a thick (∼ 1 bar) atmosphere, collision-induced absorption (CIA) due to the interaction of these gases with CO2 might provide enough greenhouse warming to elevate mean Noachian temperatures above freezing (Ramirez et al. 2014; Wordsworth et al. 2017). Ramirez et al. (2014) found that global mean surface temperatures exceeded 273K for P0 ≥ 3 bar and [H2]≥ 0.05, with higher concentrations of H2 required for lower P0. More recently, Wordsworth et al. (2017) used new ab initio calculations of H2-CO2 and CH4-CO2 CIA to show earlier estimates of the CIA were underestimated, and that 2-10% levels of CH4 or H2 in a > 1.25 bar atmosphere could elevate planetary mean temperatures over freezing. While it is unclear if such high reducing conditions can be sustained in the steady state, this scenario remains an intriguing avenue to a Noachian Mars with conditions at least transiently globally clement for liquid water and prebiotic chemistry (Batalha et al. 2015; Wordsworth et al. 2017).
H2 is spectrally inert at UV wavelengths compared to CO2. Based on the constraints on H2 absorption we found (Backx et al. 1976; Victor and Dalgarno 1969), the contribution of H2 to atmospheric absorption and scattering are negligible for H2 mixing ratios of 0 − 0.1. Similarly, CH4 does not absorb at wavelengths longer than 165 nm (Au et al. 1993; Chen and Wu 2004). Hence, it’s absorption is highly degenerate with CO2, and its presence at the levels suggested in Wordsworth et al. (2017) does not impact the UV surface environment. Photochemically-generated hydrocarbon hazes require CH4/CO2 ratios of > 0.1, and are consequently expected to be thin or nonexistent in this scenario (DeWitt et al. 2009). Consequently, the UV surface environment in an H2 or CH4-rich atmosphere should be similar to the pCO2=2 bar case discussed in Section 5.1.
5.3. Highly-Volcanic Mars (CO2-H2O-SO2/H2S Atmosphere)
We have so far considered atmospheres with CO2 and H2O as their dominant photoactive gaseous species. However, other gases have been proposed as signiﬁcant constituents of the Martian atmosphere. In particular, Halevy et al. (2007) suggest that the lack of massive carbonate deposits on Mars could have been explained if, during epochs of high volcanism on young Mars, SO2 built up to the ∼ 1 − 100 ppm level. At such levels, Halevy et al. (2007) ﬁnd that SO2 would supplant CO2 as the agent regulating global chemistry and climate, inhibiting massive carbonate precipitation in the process. Halevy and Head (2014) further argue that enhanced radiative forcing from high SO2 levels could transiently raise mean surface temperatures at the subsolar point (assuming no horizontal heat transport)

– 39 –
above the freezing point of water, explaining the observed ﬂuvial features. Halevy and Head (2014) calculate that SO2 mixing ratios 10 ppm (1 bar atmosphere) could have been possible during, e.g., the emplacement of the Martian volcanic plains. While the impact of SO2 on Martian carbonates and climate remains debated (e.g., Niles et al. 2013, Kerber et al. 2015), it remains plausible that Noachian Mars may have been characterized by at least transiently high SO2 levels due to higher volcanic outgassing rates.
We consequently sought to explore the impact of elevated levels of SO2 on the UV surface environment and hence prebiotic chemistry. Figure 17 presents the dose rates calculated for a clear-sky atmosphere with varying pSO2 and pCO2. Figure 18 presents the dose rates calculated for an atmosphere with pCO2=0.02 bar (optically thin at scattering wavelengths), but varying levels of CO2 clouds. In both cases, SZA=0 and A corresponds to desert.
SO2 is a far broader, stronger UV absorber than either CO2 or H2O, and consequently its presence can exert a dramatic impact on UV surface radiance and photochemistry rates. As with dust, multiple scattering from other atmospheric constituents can amplify SO2 absorption. For pCO2 ≤ 0.2 bar, pSO2 ≥ 2 × 10−6 bar is required to suppress dose rates to Di < 0.1, whereas for pCO2 ≥ 2 bar, pSO2 ≥ 2 × 10−7 bar is suﬃcient. Similarly, for τcloud ≤ 10, pSO2 ≥ 2 × 10−6 bar is required to suppress dose rates to Di < 0.1, whereas for τcloud ≥ 100, pSO2 ≥ 2 × 10−7 bar is suﬃcient. For pSO2 ≥ 2 × 10−5 bar, UV-sensitive prebiotic photochemistry is strongly quenched.

– 40 –

Relative Dose Rate ¯Di

Relative Dose Rate ¯Di

111111111101100000000000-1---------019876543210 11111111111001100000000000--11---------0198765432110
1111111111100000000000------211111-----1975311975310-9

UMP UMP UMP

Bond Bond Bond

Cleavage Cleavage Cleavage

(((λλλ000

= = =

193) 230) 254)

CCuuCCNN33

Photoionization Photoionization

((λλ00

= =

254) 300)

pCO2=0.02 bar

pCO2=0.2 bar

pCO2=2 bar
10-8 10-7 10-6 10-5 10-4 pSO2 (bar)

Relative Dose Rate ¯Di

Fig. 17.— UV dose rates Di for CO2-H2O-SO2 atmospheres as a function of pCO2 and pSO2 (T0 = 250K, SZA=0, A corresponding to desert).

– 41 –

Relative Dose Rate ¯Di Relative Dose Rate ¯Di Relative Dose Rate ¯Di Relative Dose Rate ¯Di

111111111101100000000000-1---------019876543210 11111111111001100000000000--11---------0198765432110 11111111111111100000010000000000------111111---------0987654321543210 11111111110000000000-----11111-----197531975310-9

UMP UMP UMP

Bond Bond Bond

Cleavage Cleavage Cleavage

(((λλλ000

= = =

193) 230) 254)

CCuuCCNN33

Photoionization Photoionization

((λλ00

= =

254) 300)

τcloud = 1(unscaled)

τcloud = 10 (unscaled)

τcloud = 100 (unscaled)

τcloud = 1000 (unscaled)
10-8 10-7 10-6 10-5 10-4 pSO2

Fig. 18.— UV dose rates Di for CO2-H2O-SO2 atmospheres with a CO2 cloud deck emplaced, as a function of pSO2 and τcloud (T0 = 250K, pCO2=0.02SZA=0, A corresponding to desert).

As with dust, we considered the hypothesis that attenuation from SO2 might have a differential impact on the eustressor and stressor pathways. We calculated DUMP −X/DCuCN3−Y for pCO2 = 2 bar (no clouds) and τcloud = 1000 (pCO2=0.02 bar) for a broad range of pSO2. This calculation is presented in Figure 19. We note that regardless of assumption on λ0, as pSO2 increases from 2 × 10−7 − 2 × 10−5 bar, the eustressor pathway is favored over the stressor pathways, by as much as 2 orders of magnitude (dependent on λ0). We conclude that it

– 42 –
seems plausible that high-SO2 planetary atmospheres have a UV throughput more clement for abiogensis compared to low-SO2 atmospheres under the assumption that the stressor and eustressor pathways we have identiﬁed were important. However, better measurements of the spectral QY of these photoprocesses is required to conﬁrm and quantify the magnitude of this eﬀect.

– 43 –

UMP-193/CuCN3-254 UMP-230/CuCN3-254 UMP-254/CuCN3-254

UMP-193/CuCN3-300 UMP-230/CuCN3-300 UMP-254/CuCN3-300

101 pCO2=2 bar, τcloud = 0

100

¯DUMP − X/ ¯DCuCN3 − Y

10-1

10-2

10-3
101 pCO2=0.02 bar, τcloud = 1000 (unscaled)

100

¯DUMP − X/ ¯DCuCN3 − Y

10-1

10-2

10-130-9 10-8 10-7 10-6 10-5 10-4 pSO2

Fig. 19.— Ratio of stressor dose rates UMP-X divided by eustressor dose rates UMP-Y for CO2-H2O-SO2 atmospheres, as a function of pSO2 (T0 = 250K, SZA=0, A corresponding to desert). The atmospheres are highly scattering, either because of high pCO2 or thick CO2 clouds. Lower ratios imply a more favorable environment for abiogenesis as measured by these two photoprocesses.

While Halevy and Head (2014) focused on the abundance of SO2 in the Martian atmosphere, H2S is emitted in equal proportion by the more reduced Martian mantle Halevy et al. (2007). Hu et al. (2013) model the atmospheric composition as a function of sulfur

– 44 –
emission rate for a 1-bar CO2 atmosphere assuming equipartition of the outgassed sulfur between SO2 and H2S, irradiated by a G2V star at a distance of 1.3 AU. They ﬁnd H2S concentrations to be even higher than SO2 concentrations, by over an order of magnitude at high S emission rates. H2S is also a stronger, broader UV absorber than CO2 or H2O. Consequently, we sought to explore the impact of elevated levels of H2S on the UV surface environment and prebiotic photochemistry.
Figure 20 presents the dose rates calculated for a clear-sky atmosphere with varying pH2S and pCO2. Figure 21 presents the dose rates calculated for an atmosphere with pCO2=0.02 bar (optically thin at scattering wavelengths), but varying levels of CO2 clouds. In both cases, SZA=0 and A corresponding to desert. As with SO2 and dust, highly scattering atmospheres can amplify H2S absorption. For pCO2 ≤ 0.2 bar, Di < 0.1 for pH2S≥ 2 × 10−5 bar, but for pCO2 ≥ 2 bar, Di < 0.1 for pH2S≥ 2 × 10−6 bar. Similarly, for τcloud ≤ 10, Di < 0.1 for pH2S≥ 2 × 10−5 bar, but for τcloud ≥ 100, Di 0.1 for pH2S≥ 2 × 10−6 bar. For pH2S≥ 2 × 10−4 bar, surface photochemistry is strongly quenched regardless of atmospheric state.

– 45 –

Relative Dose Rate ¯Di

Relative Dose Rate ¯Di

1111111111000000010000-------3222111---09630741852 1111111111000000010000-------3222111---09630741852 11111111111110000000000000----------3332222111---185285296307410-9

UMP UMP UMP

Bond Bond Bond

Cleavage Cleavage Cleavage

(((λλλ000

= = =

193) 230) 254)

CCuuCCNN33

Photoionization Photoionization

((λλ00

= =

254) 300)

pCO2=0.02 bar

pCO2=0.2 bar

pCO2=2 bar
10-8 10-7 10-6 10-5 10-4 10-3 pH2S (bar)

Relative Dose Rate ¯Di

Fig. 20.— UV dose rates Di for CO2-H2O-H2S atmospheres as a function of pCO2 and pH2S (T0 = 250K, SZA=0, A corresponding to desert).

– 46 –

Relative Dose Rate ¯Di Relative Dose Rate ¯Di Relative Dose Rate ¯Di Relative Dose Rate ¯Di

1111111111000000010000-------3222111---09630741852 1111111111100000000000--------32221111---74118529630 1111111111100000000000---------433322111--73395173951 1111111111111100000000000000------------554443322211--1846284062840620-9

UMP UMP UMP

Bond Bond Bond

Cleavage Cleavage Cleavage

(((λλλ000

= = =

193) 230) 254)

CCuuCCNN33

Photoionization Photoionization

((λλ00

= =

254) 300)

τcloud = 1(unscaled)

τcloud = 10 (unscaled)

τcloud = 100 (unscaled)

τcloud = 1000 (unscaled)
10-8 10-7 10-6 10-5 10-4 10-3 pH2S (unscaled)

Fig. 21.— UV dose rates Di for CO2-H2O-H2S atmospheres with a CO2 cloud deck emplaced, as a function of pH2S and τcloud (T0 = 250K, pCO2=0.02 bar, SZA=0, A corresponding to desert).

We again considered the hypothesis that H2S attenuation might have a diﬀerential impact on the stressor and eustressor dose rates. We calculated DUMP −X/DCuCN3−Y for pCO2 = 2 bar (no clouds) and τcloud = 1000 (pCO2=0.02 bar) for a broad range of pH2S. This calculation is presented in Figure 22. As in the case of dust in a dense CO2 atmosphere, the ratios diverge from 1, but in opposite directions depending on the value assumed for λ0 for

– 47 –

photoionization of tricyanocuprate. We conclude that attenuation from H2S may well have a diﬀerential impact on the stressor and eustressor dose rates, but that assessing which it favors requires better constraints on the QYs of the photoprocesses, especially the photoionization of tricyanocuprate.

UMP-193/CuCN3-254 UMP-230/CuCN3-254 UMP-254/CuCN3-254

UMP-193/CuCN3-300 UMP-230/CuCN3-300 UMP-254/CuCN3-300

¯DUMP − X/ ¯DCuCN3 − Y

1023 1021 1019 1017 1015 1013 1011 109 107 105 103 101 10-1 10-3 1035 1032 1029 1026 1023 1020 1017 1014 1011 108 105 102 10-1 10-140-9

pCO2=2 bar, τcloud = 0 pCO2=0.02 bar, τcloud = 1000 (unscaled)
10-8 10-7 10-6 10-5 10-4 pH2S

10-3

¯DUMP − X/ ¯DCuCN3 − Y

Fig. 22.— Ratio of stressor dose rates UMP-X divided by eustressor dose rates UMP-Y for CO2-H2O-H2S atmospheres, as a function of pH2S (T0 = 250K, SZA=0, A corresponding to desert). The atmospheres are highly scattering, either because of high pCO2 or thick CO2 clouds. Lower ratios imply a more favorable environment for abiogenesis as measured by these two photoprocesses.

– 48 –
6. Conclusion
We have used a two-stream multi-layer radiative transfer model to estimate the UV surface environment on the surface of 3.9 Ga Mars as a function of atmospheric composition, and explored the implications for prebiotic chemistry. Prebiotic photoreaction rates are within an order of magnitude of the terrestrial values for normative clear-sky CO2-H2O atmospheres, in agreement with past work (e.g., Cockell 2000) suggesting early Martian and terrestrial atmospheres featured comparable UV environments. In agreement with prior work, we ﬁnd shortwave radiation to be eﬀectively attenuated by CO2 absorption, with ﬂuence 185 nm removed for pCO2 ≥ 2 × 10−5 bar and ﬂuence < 204 nm removed for pCO2 ≥ 0.2 bar. Fluence drops oﬀ more slowly in the > 204 nm regime, where neither CO2 nor H2O absorb. This is a consequence of random walk statistics in highly scattering regimes, and stands in contrast to prior studies(Ronto´ et al. 2003; Cnossen et al. 2007) which did not account for multiple scattering. The impact of CO2 and H2O clouds on their own is similarly muted because they too are pure scatterers in the > 204 nm regime; τcloud ≥ 100 is required to signiﬁcantly aﬀect surﬁcial reaction rates, comparable to but in excess of the highest patchy cloud optical depths predicted by some 3D GCM studies (e.g., Wordsworth et al. 2013).
While dense atmospheres and cloud decks only modest reduce surface ﬂuence on their own, in concert with other absorbers (dust, SO2, H2S) they can have a dramatic eﬀect on surface ﬂuence and reaction rates, though ampliﬁcation of the eﬀects of these absorbers. Dust levels of τd = 1, only a factor of a few higher than that sustained in the modern atmosphere, could suppress prebiotic reaction rates by orders of magnitude for pCO2 ≥ 2bar or τcloud ≥ 1000, and dust levels of τd = 10 would sharply reduce reaction rates independent of the atmospheric state. Similarly, pSO2 ≥ 10−7 bar or pH2S≥ 2 × 10−6 bars are required to signiﬁcantly reduce reaction rates for pCO2 ≥ 2bar or τcloud ≥ 100, but for less scattering atmospheres, pSO2 ≥ 10−6 bar or pH2S≥ 2 × 10−5 bar is required. pSO2 ≥ 10−5 bars or pH2S≥ 2 × 10−4 bars quenches UV-sensitive photochemistry by many orders of magnitude regardless of other atmospheric conditions.
The absorbers described above have spectrally variable absorption, and prebiotic photochemistry is wavelength-dependent, leading us to speculate whether high abundances of these absorbers, despite suppressing UV-sensitive chemistry generally, might not favor or disfavor eustressor photoprocesses conducive to the origin of life compared to stressor processes that impede life’s formation. We compare the relative impact of absorption from dust, SO2, and H2S attenuation on two such stressor (cleavage of the N-glycosidic bond of UMP) and eustressor photoprocesses (production of aquated electrons from CuCN23−. We ﬁnd that it is possible for high levels of these absorbers to disproportionately favor one or the other

– 49 –
of these photoprocesses. In particular, high SO2 levels may create an especially favorable environment for abiogenesis, under the assumption that these photoprocesses were important to abiogenesis (so long as enough ﬂuence reaches the ground to power these reactions). However, the magnitude and direction of this eﬀect is sensitive to assumptions about the QY of these processes. Better characterization of the spectral quantum yields of these processes are required to rule deﬁnitively on this question.
7. Acknowledgements
We thank C. Magnani and S. Rugheimer for sharing their data with us, and for many insightful conversations. We thank J. Sutherland and J. Szostak for sharing their insights into prebiotic chemistry. We thank two anonymous referees, whose comments improved this manuscript.
This research has made use of NASA’s Astrophysics Data System Bibliographic Services, and the MPI-Mainz UV-VIS Spectral Atlas of Gaseous Molecules.
S. R. and D. D. S. gratefully acknowledge support from the Simons Foundation, grant no. 290360.
8. Author Disclosure Statement
The authors declare no competing ﬁnancial interests.

– 50 –
REFERENCES
Adcock, C. T., E. M. Hausrath, and P. M. Forster (2013). Readily available phosphate from minerals in early aqueous environments on Mars. Nature Geoscience 6 (10), 824–827.
Au, J. W., G. Cooper, G. R. Burton, T. N. Olney, and C. Brion (1993). The valence shell photoabsorption of the linear alkanes, c n h 2n+ 2 (n= 1–8): absolute oscillator strengths (7–220 ev). Chemical physics 173 (2), 209–239.
Backx, C., G. R. Wight, and M. J. Van der Wiel (1976, February). Oscillator strengths (1070 eV) for absorption, ionization and dissociation in H2, HD and D2, obtained by an electron-ion coincidence method. Journal of Physics B Atomic Molecular Physics 9, 315–331.
Batalha, N., S. D. Domagal-Goldman, R. Ramirez, and J. F. Kasting (2015, September). Testing the early Mars H2-CO2 greenhouse hypothesis with a 1-D photochemical model. Icarus 258, 337–349.
Benner, S. A. (2013). Planets, Minerals and Life’s Origin. Mineralogical Magazine 77 (5), 686.
Benner, S. A., H.-j. K. Foundation, and A. Fl (2015). The Case for a Martian Origin for Earth Life. In R. B. Hoover, G. V. Levin, A. Y. Rozanov, and N. C. Wickramasinghe (Eds.), Instruments, Methods and Missions for Astrobiology XVII, Volume 9606, pp. 1–16.
Benner, S. A., H.-J. Kim, M.-J. Kim, and A. Ricardo (2010). Planetary organic chemistry and the origins of biomolecules. Cold Spring Harbor perspectives in biology 2 (7), a003467.
Bohren, C. F. (1987, June). Multiple scattering of light and some of its observable consequences. American Journal of Physics 55, 524–533.
Bristow, T. F., D. L. Bish, D. T. Vaniman, R. V. Morris, D. F. Blake, J. P. Grotzinger, E. B. Rampe, J. A. Crisp, C. N. Achilles, D. W. Ming, B. L. Ehlmann, P. L. King, J. C. Bridges, J. L. Eigenbrode, D. Y. Sumner, S. J. Chipera, J. M. Moorokian, A. H. Treiman, S. M. Morrison, R. T. Downs, J. D. Farmer, D. Des Marais, P. Sarrazin, M. M. Floyd, M. A. Mischna, and A. C. McAdam (2015). The origin and implications of clay minerals from Yellowknife Bay, Gale crater, Mars. American Mineralogist 100 (4), 824–836.

– 51 –
Chen, F. and C. Wu (2004, May). Temperature-dependent photoabsorption cross sections in the VUV-UV region. I. Methane and ethane. Journal of Quantitative Spectroscopy and Radiative Transfer 85 (2), 195–209.
Claire, M. W., J. Sheets, M. Cohen, I. Ribas, V. S. Meadows, and D. C. Catling (2012, September). The Evolution of Solar Flux from 0.1 nm to 160 µm: Quantitative Estimates for Planetary Studies. ApJ 757, 95.
Cnossen, I., J. Sanz-Forcada, F. Favata, O. Witasse, T. Zegers, and N. F. Arnold (2007, February). Habitat of early life: Solar X-ray and UV radiation at Earth’s surface 4-3.5 billion years ago. Journal of Geophysical Research (Planets) 112, 2008.
Cockell, C. (2000). The ultraviolet history of the terrestrial planets: implications for biological evolution. Planetary and Space Science 48.
Cockell, C. S. (1999). Carbon biochemistry and the ultraviolet radiation environments of F, G, and K main sequence stars. Icarus 141 (2), 399–407.
Cockell, C. S. (2002). Photobiological uncertainties in the archaean and post-archaean world. International Journal of Astrobiology 1 (01), 31–38.
Colaprete, A. and O. B. Toon (2003, April). Carbon dioxide clouds in an early dense Martian atmosphere. Journal of Geophysical Research (Planets) 108, 6–1.
DeWitt, H. L., M. G. Trainer, A. A. Pavlov, C. A. Hasenkopf, A. C. Aiken, J. L. Jimenez, C. P. McKay, O. B. Toon, and M. A. Tolbert (2009, June). Reduction in Haze Formation Rate on Prebiotic Earth in the Presence of Hydrogen. Astrobiology 9, 447–453.
Fanale, F. P., J. R. Salvail, W. B. Banerdt, and R. S. Saunders (1982, June). Mars - The regolith-atmosphere-cap system and climate change. Icarus 50, 381–407.
Forget, F., F. Hourdin, R. Fournier, C. Hourdin, O. Talagrand, M. Collins, S. R. Lewis, P. L. Read, and J.-P. Huot (1999). Improved general circulation models of the martian atmosphere from the surface to above 80 km. Journal of Geophysical Research: Planets 104 (E10), 24155–24175.
Forget, F. and R. T. Pierrehumbert (1997). Warming early mars with carbon dioxide clouds that scatter infrared radiation. Science 278 (5341), 1273–1276.
Forget, F., R. Wordsworth, E. Millour, J.-B. Madeleine, L. Kerber, J. Leconte, E. Marcq, and R. M. Haberle (2013, January). 3D modelling of the early martian climate under a denser CO2 atmosphere: Temperatures and CO2 ice clouds. Icarus 222, 81–99.

– 52 –
Gollihar, J., M. Levy, and A. D. Ellington (2014). Many paths to the origin of life. Science 343 (6168), 259–260.
Grotzinger, J. P., S. Gupta, M. C. Malin, D. M. Rubin, J. Schieber, K. Siebach, D. Y. Sumner, K. M. Stack, a. R. Vasavada, R. E. Arvidson, F. Calef, L. Edgar, W. F. Fischer, J. a. Grant, J. Griﬀes, L. C. Kah, M. P. Lamb, K. W. Lewis, N. Mangold, M. E. Minitti, M. Palucis, M. Rice, R. M. E. Williams, R. a. Yingst, D. Blake, D. Blaney, P. Conrad, J. Crisp, W. E. Dietrich, G. Dromart, K. S. Edgett, R. C. Ewing, R. Gellert, J. a. Hurowitz, G. Kocurek, P. Mahaﬀy, M. J. McBride, S. M. McLennan, M. Mischna, D. Ming, R. Milliken, H. Newsom, D. Oehler, T. J. Parker, D. Vaniman, R. C. Wiens, and S. a. Wilson (2015). Deposition, exhumation, and paleoclimate of an ancient lake deposit, Gale crater, Mars. Science (New York, N.Y.) 350 (6257), aac7575.
Gurzadyan, G. G. and H. Go¨rner (1994). Damage to uracil-and adenine-containing bases, nucleosides, nucleotides and polynucleotides: Quantum yields on irradiation at 193 and 254 nm. Photochemistry and photobiology 60 (4), 323–332.
Halevy, I. and J. W. Head, III (2014, December). Episodic warming of early Mars by punctuated volcanism. Nature Geoscience 7, 865–868.
Halevy, I., M. T. Zuber, and D. P. Schrag (2007, dec). A sulfur dioxide climate feedback on early Mars. Science (New York, N.Y.) 318 (5858), 1903–7.
Hansen, G. B. (1997, September). The infrared absorption spectrum of carbon dioxide ice from 1.8 to 333 µm. J. Geophys. Res. 102, 21569–21588.
Hansen, G. B. (2005, November). Ultraviolet to near-infrared absorption spectrum of carbon dioxide ice from 0.174 to 1.8 µm. Journal of Geophysical Research (Planets) 110, E11003.
Hansen, J. E. and L. D. Travis (1974, October). Light scattering in planetary atmospheres. Space Sci. Rev. 16, 527–610.
Herd, C. D. K., L. E. Borg, J. H. Jones, and J. J. Papike (2002, June). Oxygen fugacity and geochemical variations in the martian basalts: implications for martian basalt petrogenesis and the oxidation state of the upper mantle of Mars. Geochim. Cosmochim. Acta 66, 2025–2036.
Hoekzema, N. M., M. Garcia-Comas, O. J. Stenzel, B. Grieger, W. J. Markiewicz, K. Gwinner, and H. U. Keller (2010, June). Optical depth and its scale-height in Valles Marineris from HRSC stereo images. Earth and Planetary Science Letters 294, 534– 540.

– 53 –
Horva´th, A., S. Papp, and Z. D´ecsy (1984). Formation of aquated electrons and the individual quantum yields for photoactive species in the Cu (I)KCNH2O system. Journal of photochemistry 24 (4), 331–339.
Hu, R., S. Seager, and W. Bains (2012, December). Photochemistry in Terrestrial Exoplanet Atmospheres. I. Photochemistry Model and Benchmark Cases. ApJ 761, 166.
Hu, R., S. Seager, and W. Bains (2013, May). Photochemistry in Terrestrial Exoplanet Atmospheres. II. H2S and SO2 Photochemistry in Anoxic Atmospheres. ApJ 769, 6.
Joseph, J. H., W. J. Wiscombe, and J. A. Weinman (1976, December). The delta-Eddington approximation for radiative ﬂux transfer. Journal of Atmospheric Sciences 33, 2452– 2459.
Kasting, J. F. (1991, November). CO2 condensation and the climate of early Mars. Icarus 94, 1–13.
Kerber, L., F. Forget, and R. Wordsworth (2015, November). Sulfur in the early martian atmosphere revisited: Experiments with a 3-D Global Climate Model. Icarus 261, 133–148.
Kirschvink, J. L. and B. P. Weiss (2002). Mars, panspermia, and the origin of life: where did it all begin. Palaeontologia electronica 4 (2), 8–15.
Kylling, A., K. Stamnes, and S. Tsay (1995). A reliable and eﬃcient two-stream algorithm for spherical radiative transfer: Documentation of accuracy in realistic layered media. Journal of Atmospheric Chemistry 21 (2), 115–150.
Lemmon, M. T., M. J. Wolﬀ, J. F. Bell, III, M. D. Smith, B. A. Cantor, and P. H. Smith (2015, May). Dust aerosol, clouds, and the atmospheric optical depth record over 5 Mars years of the Mars Exploration Rover mission. Icarus 251, 96–111.
Madronich, S. (1987). Photodissociation in the atmosphere: 1. actinic ﬂux and the eﬀects of ground reﬂections and clouds. Journal of Geophysical Research: Atmospheres 92 (D8), 9740–9752.
Magnani, C. J. (2015). Life’s ﬁrst steps: Exploring the feasibility of prebiotic chemical reactions through ultraviolet dependence. A.B. Thesis, Harvard University.
Maher, K. A. and D. J. Stevenson (1988, February). Impact frustration of the origin of life. Nature 331, 612–614.

– 54 –
Mayer, B., A. Kylling, S. Madronich, and G. Seckmeyer (1998, December). Enhanced absorption of UV radiation due to multiple scattering in clouds: Experimental evidence and theoretical explanation. J. Geophys. Res. 103, 31.
Miller, T. M. (2009). Atomic and molecular polarizabilities. In D. R. Lide (Ed.), CRC Handbook of Chemistry and Physics (90 ed.). Boca Raton, FL: CRC Press.
Mishra, M. K., P. Chauhan, R. Singh, S. M. Moorthi, and S. S. Sarkar (2016, February). Estimation of dust variability and scale height of atmospheric optical depth (AOD) in the Valles Marineris on Mars by Indian Mars Orbiter Mission (MOM) data. Icarus 265, 84–94.
Mulkidjanian, A. Y., D. A. Cherepanov, and M. Y. Galperin (2003). Survival of the ﬁttest before the beginning of life: selection of the ﬁrst oligonucleotide-like polymers by UV light. BMC Evolutionary Biology 3, 12.
Niles, P. B., D. C. Catling, G. Berger, E. Chasseﬁ`ere, B. L. Ehlmann, J. R. Michalski, R. Morris, S. W. Ruﬀ, and B. Sutter (2013). Geochemistry of carbonates on mars: implications for climate history and nature of aqueous environments. Space Science Reviews 174 (1-4), 301–328.
Pang, K. and J. M. Ajello (1977, January). Complex refractive index of Martian dust Wavelength dependence and composition. Icarus 30, 63–74.
Parkinson, W. H., J. Rufus, and K. Yoshino (2003, May). Absolute absorption cross section measurements of CO2 in the wavelength region 163-200 nm and the temperature dependence. Chemical Physics 290, 251–256.
Patel, B. H., C. Percivalle, D. J. Ritson, C. D. Duﬀy, and J. D. Sutherland (2015). Common origins of RNA, protein and lipid precursors in a cyanosulﬁdic protometabolism. Nature Chemistry (March), 1–7.
Pierrehumbert, R. T. (2010). Principles of planetary climate. Cambridge University Press.
Powner, M. W., B. Gerland, and J. D. Sutherland (2009, may). Synthesis of activated pyrimidine ribonucleotides in prebiotically plausible conditions. Nature 459 (7244), 239–42.
Ramirez, R. M., R. Kopparapu, M. E. Zugger, T. D. Robinson, R. Freedman, and J. F. Kasting (2014). Warming early mars with co2 and h2. Nature Geoscience 7 (1), 59–63.

– 55 –
Ranjan, S. and D. D. Sasselov (2016, January). Inﬂuence of the UV Environment on the Synthesis of Prebiotic Molecules. Astrobiology 16, 68–88.
Ranjan, S. and D. D. Sasselov (2017, March). Constraints on the Early Terrestrial Surface UV Environment Relevant to Prebiotic Chemistry. Astrobiology 17, 169–204.
Rapf, R. J. and V. Vaida (2016). Sunlight as an energetic driver in the synthesis of molecules necessary for life. Phys. Chem. Chem. Phys. 18, 20067–20084.
Ritson, D. and J. D. Sutherland (2012, November). Prebiotic synthesis of simple sugars by photoredox systems chemistry. Nature chemistry 4 (11), 895–9.
Robinson, T. D. and D. C. Catling (2014, January). Common 0.1bar tropopause in thick atmospheres set by pressure-dependent infrared transparency. Nature Geoscience 7, 12–15.
Ronto´, G., A. B´erces, H. Lammer, C. S. Cockell, G. J. Molina-Cuberos, M. R. Patel, F. Selsis, and A. Be (2003). Solar UV irradiation conditions on the surface of Mars. Photochemistry and photobiology 77 (1), 34–40.
Rosenberg, R., M. Abu Haija, and P. Ryan (2008, October). Chiral-Selective Chemistry Induced by Spin-Polarized Secondary Electrons from a Magnetic Substrate. Physical Review Letters 101 (17), 178301.
Rugheimer, S., A. Segura, L. Kaltenegger, and D. Sasselov (2015, June). UV Surface Environment of Earth-like Planets Orbiting FGKM Stars through Geological Evolution. ApJ 806, 137.
Sagan, C. (1973, April). Ultraviolet selection pressure on the earliest organisms. Journal of theoretical biology 39 (1), 195–200.
Sarker, P. K., J.-i. Takahashi, Y. Obayashi, T. Kaneko, and K. Kobayashi (2013, June). Photo-alteration of hydantoins against UV light and its relevance to prebiotic chemistry. Advances in Space Research 51 (12), 2235–2240.
Segura, A., V. S. Meadows, J. F. Kasting, D. Crisp, and M. Cohen (2007, September). Abiotic formation of O2 and O3 in high-CO2 terrestrial atmospheres. A&A 472, 665–679.
Sinsheimer, R. L. and R. Hastings (1949). A reversible photochemical alteration of uracil and uridine. Science 110 (2864), 525–526.
Sleep, N. H., K. J. Zahnle, J. F. Kasting, and H. J. Morowitz (1989, November). Annihilation of ecosystems by large asteroid impacts on the early earth. Nature 342, 139–142.

– 56 –
Smith, M. D., B. J. Conrath, J. C. Pearl, and P. R. Christensen (2002, May). NOTE: Thermal Emission Spectrometer Observations of Martian Planet-Encircling Dust Storm 2001A. Icarus 157, 259–263.
Stark, G., K. Yoshino, P. L. Smith, and K. Ito (2007, January). Photoabsorption cross section measurements of CO2 between 106.1 and 118.7 nm at 295 and 195 K. J. Quant. Spec. Radiat. Transf. 103, 67–73.
Stephenson, J. D., L. J. Hallis, K. Nagashima, and S. J. Freeland (2013). Boron Enrichment in Martian Clay. PLoS ONE 8 (6), 6–9.
Tian, F., J. F. Kasting, and S. C. Solomon (2009, January). Thermal escape of carbon from the early Martian atmosphere. Geophys. Res. Lett. 36, L02205.
Toon, O. B., C. P. McKay, T. P. Ackerman, and K. Santhanam (1989, November). Rapid calculation of radiative heating rates and photodissociation rates in inhomogeneous multiple scattering atmospheres. J. Geophys. Res. 94, 16287–16301.
Victor, G. A. and A. Dalgarno (1969, March). Dipole Properties of Molecular Hydrogen. J. Chem. Phys. 50, 2535–2539.
Vincendon, M., C. Pilorget, B. Gondet, S. Murchie, and J.-P. Bibring (2011, November). New near-IR observations of mesospheric CO2 and H2O clouds on Mars. Journal of Geophysical Research (Planets) 116, E00J02.
Voet, D., W. B. Gratzer, R. a. Cox, and P. Doty (1963, June). Absorption spectra of nucleotides, polynucleotides, and nucleic acids in the far ultraviolet. Biopolymers 1 (3), 193–208.
Wagner, W. and A. Pruß (2002, June). The IAPWS Formulation 1995 for the Thermodynamic Properties of Ordinary Water Substance for General and Scientiﬁc Use. Journal of Physical and Chemical Reference Data 31, 387–535.
Wagner, W., A. Saul, and A. Pruss (1994, May). International Equations for the Pressure Along the Melting and Along the Sublimation Curve of Ordinary Water Substance. Journal of Physical and Chemical Reference Data 23, 515–527.
Warren, S. G. and R. E. Brandt (2008, July). Optical constants of ice from the ultraviolet to the microwave: A revised compilation. Journal of Geophysical Research (Atmospheres) 113, D14220.

– 57 –
Wolﬀ, M. J., M. D. Smith, R. T. Clancy, R. Arvidson, M. Kahre, F. Seelos, S. Murchie, and H. Savija¨rvi (2009, June). Wavelength dependence of dust aerosol single scattering albedo as observed by the Compact Reconnaissance Imaging Spectrometer. Journal of Geophysical Research (Planets) 114, E00D04.
Wordsworth, R., F. Forget, E. Millour, J. Head, J.-B. Madeleine, and B. Charnay (2013). Global modelling of the early martian climate under a denser CO2 atmosphere: Water cycle and ice evolution. Icarus 222 (1), 1–19.
Wordsworth, R., F. Forget, E. Millour, J. W. Head, J.-B. Madeleine, and B. Charnay (2013, January). Global modelling of the early martian climate under a denser CO2 atmosphere: Water cycle and ice evolution. Icarus 222, 1–19.
Wordsworth, R., Y. Kalugina, S. Lokshtanov, A. Vigasin, B. Ehlmann, J. Head, C. Sanders, and H. Wang (2017, January). Transient reducing greenhouse warming on early Mars. Geophys. Res. Lett. 44, 665–671.
Wordsworth, R. D. (2016, June). The Climate of Early Mars. Annual Review of Earth and Planetary Sciences 44, 381–408.
Wordsworth, R. D., L. Kerber, R. T. Pierrehumbert, F. Forget, and J. W. Head (2015). Comparison of ”warm and wet” and ”cold and icy” scenarios for early Mars in a 3-D climate model. Journal of Geophysical Research E: Planets 120 (6), 1201–1219.
Yoshino, K., J. R. Esmond, Y. Sun, W. H. Parkinson, K. Ito, and T. Matsui (1996, January). Absorption cross section measurements of carbon dioxide in the wavelength region 118.7-175.5 nm and the temperature dependence. J. Quant. Spec. Radiat. Transf. 55, 53–60.
Zurek, R. W. (1978, August). Solar heating of the Martian dusty atmosphere. Icarus 35, 196–208.
This preprint was prepared with the AAS LATEX macros v5.2.

– 58 –
A. Supplementary Figures
This appendix presents sample atmospheric proﬁles calculated using the methods in Section 3.2, to illustrate the application of these methods. Figure 23 shows the T/P proﬁles associated with a sample atmosphere with varying (pCO2, T0), where pCO2 is the surface partial pressure of CO2. Figure 24 shows the pressure, temperature, and H2O molar concentration altitude provides associated with an atmosphere with T0 = 250 K and varying pCO2.

– 59 –

Pressure (bar)

Pressure (bar)

pCO2=0.02 (bar)
10-5

10-4

10-3

10-2
10-5 pCO2=0.2 (bar)

10-4

10-3

10-2

10-1

10-4 10-3

pCO2=2 (bar)

T0 T0 T0

= = =

210K 250K 300K

10-2

10-1

100 180 200 220 240 260 280 300 Temperature (K)

Pressure (bar)

Fig. 23.— Sample T/P proﬁles for CO2-dominated CO2-H2O atmospheres using the methodology in Section 3.2 for T0 = 210, 250, 300K and pCO2 = 0.02, 0.2, 2 bar. The pressure is the total atmospheric pressure (CO2 and H2O).

– 60 –

Altitude (km)

Altitude (km)

60

50

40

30

20 10 0

pppCCCOOO222

= = =

002..b022abrba,arT,r0,T=T0 0=2=52025K500KK

10-4 10-3

10-2

10-1

Pressure (bar)

100

60 50 40 30 20 10 0 180

200 220 Temperature (K)

240

60 50 40 30 20 10 0 10-7 10-6 10-5 10-4 10-3 10-2 10-1 100
[H2O]

Altitude (km)

Fig. 24.— Temperature, pressure, and molar concentration of H2O for atmospheres with pCO2 = 0.02, 0.2, 2 bar and T0 = 250K.