The plot shows all thermal and non-thermal loss processes of hydrogen and oxygen from the present to 3.5 Gyr ago (PU: ion pick up, SP: atmosphere sputtering, DR: dissociative recombination, * thermal escape). The dotted lines at the top of the plot show an area where escape rates are comparable to hydrodynamic escape.

Neutral gas temperatures inferred from topside plasma scale heights of electron density distributions and Lyman-alpha day-glow observations obtained by several Mars missions imply a much higher dependence on solar activity expressed by the 10.7 cm radio flux F_10.7 than found for Venus. This dependence does not appear to be consistent with the observed solar cycle dependence of Martian ionospheric peak plasma densities. Photochemical equilibrium applies only to altitudes below 170 km, whereas topside scale heights are usually derived for a much greater altitude range an will be modified by transport and hot ENA particle populations.

In 1969 UV spectrometers on board of NASA's Mariner 6 and 7 spacecraft observed the Lyman-alpha day-glow of atomic H atoms in an altitude range from 200 and 24,000 km. From these observations it was found that the exospheric temperature should be in the order of 350 K ± 100 K, which seems to hot and is not consistent by using comparative planetological studies between Venus and Mars. On the other hand it is known that solar wind protons and accelerated planetary H⁺ ions can be transformed into energetic H atoms and thus also contribute to the observed neutral H atoms (see Lichtenegger et al. 2004, 237 KB).

Since these hot H may rise the exosphere temperature, we study the density distribution functions of the hot H populations and investigate their temperature effect on the Martian exosphere. We discuss our results in the view of ESA's Mars Express mission and show that correlated observations of various atmospheric parameters should give the possibility to adjust exospheric models resulting in a better understanding of the loss of water from Mars.