Reply to comment “On the hydrogen escape: Comment to variability of the hydrogen in the Martian upper atmosphere as simulated by a 3D atmosphere-exosphere coupling by J.-Y. Chaufray et al.” by V. Krasnopolsky, Icarus, 281, 262

Krasnopolsky (2017) makes a careful review of our recent results about the Martian hydrogen content of the Martian upper atmosphere (Chaufray et al., 2015). We comment here on his two major points. First, he suggests that the non-thermal escape of H₂, and particularly collisions with hot oxygen, not taken into account in our general circulation model (GCM), should modify our reported H₂ and H density profiles. This is an important issue; we acknowledge that future effective coupling of our GCM with comprehensive models of the Martian solar wind interaction, ideally after being validated with the latest plasma observations of H₂ ⁺, would allow for better estimations of the relative importance of the H₂ non-thermal and thermal escape processes. For the time being we need assumptions in the GCM, with proper and regular updates. According to a recent and detailed study of the anisotropic elastic and inelastic collision cross sections between O and H₂ (Gacesa et al., 2012), the escape rates used by Krasnopolsky (2010) for this process might be overestimated. We therefore do not include non thermal escape of H₂ in the model. And secondly, in response to Krasnopolsky's comment on the H escape variability with the solar cycle, we revised our calculations and found a small bug in the computation of the Jeans effusion velocity. Our revised computed H escape rates are included here. They have a small impact on our key conclusions: similar seasonal variations, a reduced variation with the solar cycle but still larger than Krasnopolsky (2017), and again a hydrogen scape systematically lower than the diffusion-limited flux. This bug does not affect the latest Mars Climate Database v5.2.