On the basis of the above simulations we can draw the following conclusions:
1. Interstellar oxygen atoms do penetrate the heliosphere and the Solar system while significantly interacting with the plasma component. The degree of penetration, is found to be of the same order, although larger than the neutral hydrogen filtration. It is due to the charge-exchange cross sections being of the same order for these elements. However, the comparison between models with zero and substantial neutral H density in the unperturbed LISM shows a very strong influence of the O ion - H atom charge exchange (the reverse charge-exchange) on the oxygen atom distribution. As a result, the inner heliosphere is filled with a very large fraction of secondary atoms resulting from the this reverse charge-echange. This has as a consequence to favor oxygen entrance in the heliosphere.
The fraction of neutral oxygen penetrating the heliosphere is rather high, of about 70% on the upwind side for an interstellar plasma density of 0.1 cm-3 (or about 20 to 30% losses).
2. One of the features coming out from the present two shocks model is the presence of an oxygen atom density maximum in the region between the bow shock and the heliopause (an "oxygen-wall"), with a density value higher than the density at infinity. This effect was already discovered for neutral hydrogen (Baranov and Malama, 1993). It is mainly due here to the reverse charge-exchange between compressed neutral H and O ions. However, it is less pronounced than the neutral H "wall".
4. A spectacular result of these simulations is the increase of the oxygen to hydrogen density ratio near the Sun and on the downwind side, which is mainly due to the differences between the oxygen and hydrogen masses. This effect is already present close to the Sun in the absence of heliospheric interface perturbations.
5. Recent nearby stars spectroscopic observations with the HST have provided the neutral oxygen to neutral hydrogen ratio in the Local Cloud (data from Linsky & al (1993)), allowing an estimate of the oxygen ions fluxes in the Sun vicinity in the frame of the present model. The comparison with the Ulysses measurements (Gloeckler & al, 1993) favours oxygen losses of the order of 50 (about 30% more than predicted by the present model), and subsequently interstellar plasma densities larger than 0.1 cm-3. When compared with the most recent interstellar abundances measurements, the helium to oxygen ratio derived from the Ulysses observations (Geiss & al, 1994) also implies a significant depletion of oxygen in the inner heliosphere (by a factor of at least 2), if one considers the most probable range for the ionization rate, and the best fit to the data. However, further modeling is required, with, in particular, the inclusion of electron impact ionization processes, and improved statistics close to the Sun.
© European Southern Observatory (ESO) 1997