Astron. Astrophys. 342, L13-L16 (1999)

1. Introduction

There is a long-standing debate on the properties of the hot ( K) gas which fills the so-called Local Bubble, the soft X-ray emitting region around the Sun (see Breitschwerdt et al. 1997 for updated information). The cold and warm diffuse clouds embedded in the hot gas, like the group of cloudlets in the solar vicinity, are supposed to be surrounded by a shell of semi-hot gas at an intermediate temperature ( K), and one way to detect this gas is through the neutral hydrogen absorption at Ly-. It is worthwhile here to recall that a small quantity of hot gas ( K) can produce an absorption equivalent width in Ly- comparable to that of a 3 or 4 orders of magnitude larger quantity of warm gas ( K).

Indeed, spectra obtained with the Goddard High Resolution Spectrometer (GHRS) on board the Hubble Space Telescope (HST) show hot neutral H absorption along the line-of-sight to a few nearby stars, including Sirius A and CMa (Bertin et al. 1995, hereafter BVL95; Gry et al. 1995). The origin of this absorption, however, has been a matter of debate. For the Sirius line-of-sight, BVL95 proposed that a hot conductive interface is responsible for excess Ly- seen on the red side of the absorption profile, and Bertin et al. (1995b) proposed that absorption from Sirius's wind accounted for excess Ly- absorption on the blue side. In other cases, detected hot H components are undoubtedly linked to the neutral gas formed either around our own heliosphere, due to the interaction of the solar wind with the ambient interstellar medium of the Local Interstellar Cloud (LIC) (Linsky & Wood 1996, hereafter LW96), or to neutral gas around other astrospheres due to the corresponding interaction between the stellar winds and the ambient neutral interstellar gas (Wood et al. 1996). But for Sirius A and CMa, the combination of conductive interface and stellar wind absorption has remained the most likely source.

There are 3 different types of heliospheric H atoms, in addition to the unperturbed interstellar neutral H called primary interstellar atoms or PIA's: i) the compressed, decelerated, and heated interstellar atoms (HIA's) formed by charge exchange with heated interstellar protons outside the heliopause, ii) the neutralized, decelerated, and heated solar wind atoms (HSWA's) formed in the heliosheath by charge exchange between the neutral interstellar gas and the hot protons of the decelerated and compressed solar wind, and iii) the neutralized supersonic solar wind atoms (SSWA's). Only the HIA's and HSWA's are of interest here since the SSWA component is flowing radially at very large velocities and will not produce absorption in the central part of the Ly- lines, and the PIA's are indistinguishable from the normal interstellar gas. The HIA's on the upwind side of the heliosphere (i.e. the direction from which the interstellar wind flows) collectively make up the so-called "H-wall", which is the gas that has been detected towards Cen (LW96).

The properties of the HSWA's are the most difficult to calculate, since this hot gas has a very large mean free path and its characteristics at one location in the heliosphere depend on the properties of all the source regions everywhere in the heliosheath. Indeed, significant differences between multi-fluid models and kinetic models have been found by Williams et al. (1997). These authors have also suggested that the mixing between the hot and warm populations in the heliospheric tail through H-H collisions could be the origin of the hot gas absorption observed towards Sirius. While recent computations show that H-H collisions are negligible compared to charge-exchange processes (Izmodenov et al. 1999b), our conclusions below will ultimately be similar to their original idea.

The goal of this letter is to show that when one uses updated parameters of the circumsolar interstellar medium and a very precise kinetic/gasdynamic self-consistent model of the heliosphere, HSWA's produce a non-negligible absorption in almost all directions, with a maximum effect on the downwind side. We reconsider the Sirius A HST Ly- spectrum and show that the red wing of the absorption is very well fitted using our model. Then we show using simple analogies that the additional absorption on the blue wing could be produced by HSWA's and HIA's around Sirius itself, if the star is embedded in the neighboring cloud detected towards the star by Lallement et al. (1994) and if the star produces a wind, which is likely.

1.1. Heliospheric absorption towards Sirius

A description of our self-consistent heliospheric model of the solar wind-interstellar gas interaction can be found in Baranov & Malama (1993), Baranov et al. (1998), and Izmodenov et al. (1999a). We have updated the interstellar parameters to take into account recent advances in the field, such as the velocity and temperature determinations of the LIC from in situ helium measurements and stellar spectroscopy (Witte et al. 1993; Lallement & Bertin 1992; Bertin et al. 1993), as well as estimates of the neutral H and electron density in the circumsolar interstellar medium (Lallement et al. 1996; Izmodenov et al. 1999a). In what follows, the assumed interstellar parameters are then: K, km/s, cm^-3, cm^-3. The upwind direction is taken as , (ecliptic coordinates), which translates into , (galactic coordinates). The assumed solar wind parameters at 1 AU are: cm^-3, km/s. The model does not include an interstellar magnetic field, but our estimates should not be significantly changed in the presence of a moderate field. The boundary of the model grid is at a distance of about 2000 AU in the direction of Sirius.

Fig. 1a is a sketch of the heliosphere and shows the direction of Sirius on the downwind side. The predicted absorption by HSWA's and HIA's in the direction of Sirius at an angle of from the upwind direction is displayed in Fig. 1b. The absorption is shown in a heliocentric rest frame. It can be seen that the HSWA's are the main absorbers, and that their absorption is far from negligible.

Fig. 1. a Schematic view of the Sun-Sirius line-of-sight. b  Transmission as a function of Doppler shift in the solar rest frame through heliospheric HIA's and HSWA's in the direction of Sirius, and through the siriospheric H atoms in conditions described in the text. c  The GHRS spectrum, the simulated profile after ISM absorption, and the profile after ISM + heliospheric absorption. d  Simulated profile after ISM + Heliospheric + Siriospheric absorption, where a good fit to the data is obtained for a column density of HIA and HSWA atoms two times the heliospheric column for the same orientation.

1.2. Heliospheric and interstellar absorption towards Sirius

The 2.7 pc long line-of-sight to Sirius has been shown to cross two clouds: i) the LIC, which in this direction is seen at a positive redshift of 19 km s^-1, and ii) a second cloud at a Doppler shift of 13 km s^-1 (Lallement et al. 1994), which is probably of the same type as our Local Cloudlet. Using the angularly close star CMa, Gry & Dupin (1998) have argued that the LIC extent in that direction is not longer than 0.6 pc.

Fig. 1c shows the Sirius spectrum around the Ly- line, and a simple polynomial fit to the continuum surrounding the D and H absorption. Superimposed on the data is the expected profile after absorption by the two clouds at and 19 km s^-1, respectively, with an assumed temperature of K. The column densities of the two clouds are both cm^-2, in agreement with the D absorption and a D/H ratio of (Linsky et al, 1995, BVL95). Our conclusions are not sensitive to either the exact value of D/H, or to the exact interstellar temperature. The absorption has been convolved by the instrumental profile corresponding to the G160M/SSA settings of the GHRS spectrograph.

It is clearly seen that with warm interstellar gas only, absorption is missing on both sides of the line, as already noticed by BVL95. After adding the modeled absorption by the heliosphere, the resulting profile is substantially modified on the red part of the line. To make the heliospheric effect clear in Fig. 1c, the additional absorption is shown as a hatched area. It can be seen that the red part of the observed spectrum is well fitted by the model. Thus, while the heliosphere cannot been made responsible for the absorption in the blue wing, there is no need to propose additional absorption from interstellar hot gas along the line-of-sight to fit the red side of the absorption line.

© European Southern Observatory (ESO) 1999

Online publication: December 22, 1998
helpdesk.link@springer.de