Astron. Astrophys. 325, 745-754 (1997)

1. Introduction to problem and motivation

For several decades now attempts have been made to deduce the relevant thermodynamical parameters of the nearby local interstellar medium (LISM) by analyzing interplanetary resonance glow intensities of hydrogen and helium (see e.g. reviews by Fahr 1974; Thomas 1978; Holzer 1977; Bertaux 1984). Up to more recent times these derivations were solely based on frequency-integrated glow intensity data which give added-up contributions from local radiation sources along the line of sight. Thus the interpretation clearly suffers from the fact that hydrogen and helium producing the resonance emission are strongly inhomogeneous along the line of sight that concerns their dynamics and thermodynamics. As one of the consequences one always had to face the puzzling outcome that disjunctive thermodynamical parameters were derived for LISM helium and LISM hydrogen, respectively. Though this difference became more and more accepted as real and as a mere consequence of the interface-filtration suffered exclusively by hydrogen, it was nevertheless hoped all the time for better glow information that could either remove the discrepancies or confirm the interface idea.

This better information is now available in the form of spectrally resolved interplanetary glow data, at least in the case of H-Ly . The first spectra of the diffuse interplanetary Ly glow were obtained with the COPERNICUS spectrograph (Adams & Frisch 1977) and with the International Astronomical Explorer (IUE)(Clarke et al. 1984). More recently also the GHRS spectrograph (Goddard High Resolution Spectrograph) of the Hubble Space Telescope (HST) has been used to obtain well resolved interplanetary Ly emission spectra (Lallement et al. 1993; Clarke et al. 1995; Clarke et al. in prep.). For the analysis of the observed spectra these latter authors have used fits of the spectral data by theoretical spectra generated on the basis of the so-called "classical glow modelling". Within this modelling an optically thin approach is applied to the case of a hydrogen flow over the solar system with homogeneous temperature and a bulk velocity what is allowed to change with position due to net solar potential. Using this approach Lallement et al. (1993) come to the conclusion that the HST-GHRS spectrum, after subtraction of the geocoronal contribution, is best fitted by a model spectrum calculated for hydrogen with an inflow velocity of = 20 km/s. As they can show, at least their fit adopting = 20 km/s is significantly better than their fit taking an inflow of = 26 km/s that is indicated by results of the ULYSSES GAS experiment (Witte et al. 1993), though with an adopted temperature of = 8000 K their modelled profile turns out to be too narrow. With this result the authors conclude that they are seeing the deceleration of the hydrogen flow by about 6 km/s consistent with the charge-exchange-coupling to the interface plasma predicted by Fahr et al. (1986), Fahr (1990), Osterbart & Fahr (1992), Baranov & Malama (1993), or Fahr & Osterbart (1993).

It is the purpose of this paper to show that this conclusion is strongly biased by the "classical spectral modelling" and needs some revision when a more refined spectral analysis is carried out. How much the above conclusion may be biased by the adoption of a homogeneous flow can probably already be deduced from the result of an analogous spectral analysis carried out by Clarke et al. (1984) in which an effective solar gravity field had been admitted operating on hydrogen at its flow over the solar system. If an effective gravity operates on hydrogen, connected with an effective solar mass = (1- µ) , then the hydrogen bulk velocity locally varies, i.e. is inhomogeneous. Then a part of the spectral spread cannot be interpreted as due to thermal motions, but is caused by the bulk velocity spread along the line of sight. An interesting hint to the effect of this bulk velocity spread is the fact that Clarke et al. (1984) using µ = 0.8 (instead of µ = 1, as taken by Lallement et al. 1993) derive an LISM hydrogen inflow velocity of = 25.6 km/s, a value which is fairly close to that obtained by Witte et al. (1993) for LISM helium. In addition in the analysis of Lallement et al. (1993) and Clarke et al. (1984, 1995) no interface effect was consistently and implicitly taken into account in the modelling of the interplanetary hydrogen distribution. If this effect is taken into account in a kinetic form (Osterbart & Fahr 1992; Baranov & Malama 1993), then in fact a deceleration of the hydrogen bulk flow occurs in the upwind region at large solar distances of about 80 to 100 AU. Closer to the sun, where the radiation sources for the HST-GHRS spectrum are located, no deceleration is left, in contrast here even enhanced bulk velocities arise. Thus one may doubt whether from a spectral analysis using an inconsistent inclusion of the interface, the flow deceleration by the interface can be identified. We thus feel that a refined analysis of GHRS-HST interplanetary Ly spectra is needed in order to arrive at more solid LISM parameter and interface derivations. In the analysis which we present in this paper we shall start out from interface-modulated hydrogen distribution functions and shall include the case of effective solar gravity (i.e. ). Furthermore we use a newly developed radiation transport code to calculate the Ly spectra in which we take into account the actual solar emission profile, self-absorption of the interplanetary hydrogen, and angle-dependent partial frequency redistribution.

© European Southern Observatory (ESO) 1997

Online publication: April 28, 1998

helpdesk.link@springer.de