Mars atmospheric loss and isotopic fractionation by solar-wind-induced sputtering and photochemical escape

We examine the effects of the loss of Mars atmospheric constituents by solar-wind-induced sputtering and by photochemical escape during the past 3.8 billion years. Sputtering is capable of efficiently removing species from the upper atmosphere, including the light noble gases; nitrogen and oxygen are removed by photochemical processes as well. Due to diffusive separation (by mass) above the homopause, removal from the top of the atmosphere will fractionate the isotopes of each species, with the lighter mass being preferentially lost. For carbon and oxygen, this allows us to determine the size of nonatmospheric reservoirs which mix with the atmosphere; these reservoirs can be CO2 adsorbed in the regolith and H2O in the polar ice caps. We have constructed both simple analytical models and time-dependent models of the loss of volatiles from and supply to the martian atmosphere. Both argon and neon require continued replenishment from outgassing over geologic time. For argon, sputtering loss explains the fractionation of (Ar-36)/(Ar-38) without requiring a distinct epoch of hydrodynamic escape (although fractionation of Xe isotopes still requires very early hydrodynamic loss). For neon, the current (Ne-22)/(Ne-20) ratio represents a balance between loss to space and continued resupply from the interior; the similarity of the ratio to the terrestrial value is coincidental. For nitrogen, the loss by both sputtering and photochemical escape would produce a fractionation of (N-15)/(N-14) larger than observed; an early, thicker carbon dioxide atmosphere could mitigate the nitrogen loss and produce the observed fractionation, as could continued outgassing of juvenile nitorgen. Based on the isotopic constraints, the total amount of carbon dioxide lost over geologic time is probably on the order of tens of millibars rather than a substantial fraction of a bar. The total loss from solar-wind-induced sputtering and photochemical escape, therefore, does not seem able to explain the loss of a putative thick, early atmosphere withput requiring formation of extensive surface carbonate deposits or other nonatmospheric reservoirs for CO2.