<p>Mars likely had a denser atmosphere during the Noachian eon about 3.6 to 4.0 billion years ago (Ga). How dense this atmosphere might have been, and which escape mechanisms dominated its loss are yet not entirely clear. However, non-thermal escape processes and potential sequestration into the ground are believed to be the main drivers for atmospheric loss from the present to about 4.1 Ga.</p>
<p>To evaluate non-thermal escape over the last ~4.1 billion years, we simulated the ion escape of Mars' CO<sub>2</sub> atmosphere caused by its dissociation products C and O atoms with numerical models of the upper atmosphere and its interaction with the solar wind (see Lichtenegger et al. 2021; https://arxiv.org/abs/2105.09789). We use the planetward-scattered pick-up ions for sputtering estimates of exospheric particles including <sup>36</sup>Ar and <sup>38</sup>Ar isotopes, and compare ion escape, with sputtering and photochemical escape rates. For solar EUV fluxes &#8805;3 times the present-day Sun (earlier than ~2.6 Ga) ion escape becomes the dominant atmospheric non-thermal loss process until thermal escape takes over during the pre-Noachian eon (earlier than ~4.0 - 4.1 Ga). If we extrapolate the total escape of CO<sub>2</sub>-related dissociation products back in time until ~4.1 Ga, we obtain a theoretical equivalent to CO<sub>2</sub> partial pressure of more than ~3 bar, but this amount did not necessarily have to be present and represents a maximum that could have been lost to space within the last ~4.1 Ga.</p>
<p>Argon isotopes can give an additional insight into the evolution of the Martian atmosphere. The fractionation of <sup>36</sup>Ar/<sup>38</sup>Ar isotopes through sputtering and volcanic outgassing from its initial chondritic value of 5.3, as measured in the 4.1 billion years old Mars meteorite ALH 84001, until the present day can be reproduced for assumed CO<sub>2</sub> partial pressures between ~0.2-3.0 bar, depending on the cessation time of the Martian dynamo (assumed between 3.6-4.0 Ga) - if atmospheric sputtering of Ar started afterwards. The later the dynamo ceased away, the lower the pressure could have been to reproduce <sup>36</sup>Ar/<sup>38</sup>Ar.</p>
<p>Prior to ~4.1 Ga (i.e., during the pre-Noachian eon), thermal escape should have been the most important driver of atmospheric escape at Mars, and together with non-thermal losses, might have prevented a stable and dense CO<sub>2</sub> atmosphere during the first ~400 million years. Our results indicate that, while Mars could have been warm and wet at least sporadically between ~3.6-4.1 Ga, it likely has been cold and dry during the pre-Noachian eon (see also Scherf and Lammer 2021; https://arxiv.org/abs/2102.05976).</p>
THERMAL ESCAPE IN THE HYDRODYNAMIC REGIME: RECONSIDERATION OF PARKER's ISENTROPIC THEORY BASED ON RESULTS OF KINETIC SIMULATIONS