A.M. Lagrange, H. Beust, D. Mouillet
Laboratoire d'Astrophysique de Grenoble,BP53X,F-38041 Grenoble, France
R. Ferlet, A. Lecavelier, A. Vidal-Madjar
Institut d'Astrophysique de Paris, 98bis bd Arago, F-75014 Paris,
France
M. Deleuil
Laboratoire d'Astronomie Spatiale Marseille, BP 8, F-13376
Marseille Cedex 12, France
P.D. Feldman, J.B. McPhate, H.W. Moos
Department of Physics and Astronomy, Johns Hopkins University,
Baltimore, MD 21218, USA
J.J. Lissauer
Astronomy Program, Department of Earth and Space Sciences, State
University of New York, Stony Brook, NY 11794, USA
L.M. Hobbs
Yerkes Observatory, University of Chicago, Williams Bay, WI
53191-0258, USA
M.A. McGrath
Space Telescope Science Institute, 3700 San Martin Drive, Baltimore,
MD 21218, USA
Pictoris
Keywords: Stars, Planetary systems, beta Pictoris
The IRAS excess Main Sequence star 
Pictoris is
probably the leading candidate for an extrasolar planetary system around a main
sequence star (see, for instance, the proceedings of the ``CS dust disks and
planet formation'' colloquium, 1994, Ed Frontieres). Direct images
in the optical to near infrared region of both the extremely faint, scattered
starlight and the thermal emission from the dust have shown a flatten
edge-on disk (Smith & Terrile 1984, Lagage & Pantin 1993, Mouillet,
Lagrange & Beuzit 1996), and high resolution spectroscopy has revealed the
presence of a large amount of circumstellar neutral and singly ionized gas
(Sletteback & Carpenter 1983, Vidal-Madjar et al. 1986) much denser than
the interplanetary gas of our solar system. The singly ionized stable gas
is located close to the star (
1AU; Hobbs et al. 1988) but further
than 0.3 AU, as recently deduced from ultra high resolution data obtained
at the AAT. From the latter
data, it could be also deduced that the temperature of the neutral gas is
probably around 2000 K. The origin of the stable gas, and its possible
connection with the grains are still to be understood.
Repeated high resolution spectroscopic observations both in the visible,
especially with the ESO/CAT and in the UV
domain with IUE have evidenced a strong variability in the circumstellar
lines of ionized
elements, which has been attributed to gas evaporated from solid cometary
like bodies falling on to the star (Lagrange et al. 1987). This Evaporating
Infalling Bodies (hereafter EIB) scenario
has been tested through observations and simulations over the years. The
observations, either long term surveys, from ESO, or multi-site campaigns
(Lagrange et al. 1992,1996 and ref. therein) enabled to show that the
infalling gas has a clumpy structure (Lagrange et al. 1989) and that
over-ionized species (i.e., that cannot be produced by stellar
photo-ionization, Lagrange et al. 1989, Deleuil et al. 1994) are present in
the variable gas. The observations also showed that the temperature and
the electronic densities within the infalling gas are high : T
15000 K, N
10
--10
cm
(Mouillet & Lagrange 1995).
The simulations of the evaporation of star grazing comets enabled to
reproduce very satisfactorily the observed variability (Beust et al.
1989,1991), and could also account for the presence of over-ionized species
in the variable gas (Beust & Tagger 1993). Finally the observations also
showed that the frequency of such events is rather high : 
200/year
(Lagrange et al. 1992, Ferlet et al. 1993), possibly due to infall of a
family of smaller bodies on the
single orbit (Beust et al. 1996). The triggering mechanism for such infalls
could be planets (Levison et al. 1994, Beust & Morbidelli 1995).
The EIB scenario explained then quite well the so far
observed variability. It has
to be noticed that independent observations
have also brought arguments in favor of the presence of a large amount of
cometary material around the star, possibly
responsible for part of the dust (Knacke et al. 1989).
However, this scenario has still to be further tested, and some important questions still need answers; among them, the origin of the stable gas, the coupling of gas and dust, the age of the system, which matter is still very controversial (Paresce 1991, Lanz, Heap & Hubeny 1995), and the nature of the disk, is it a remnant of a protoplanetary disk, or is it a protoplanetary disk?
In order to further test the EIB scenario, and to understand the origin of the stable gas, we tried to perform its chemical analysis. The visible domain reveals a very limited number of lines: CaII, FeII, NaI. The UV domain is potentially richer, as many transitions from ground- or low-excited levels are present. IUE already enabled to detect other metallic elements: FeII, MgII, AlIII, CIV. However, its performances in terms of spectral resolution and S/N did not enable it to perform abundance measurements for the stable and variable gas---indeed, these components are often blended---nor to detect weak lines of other elements. The performances of GHRS enabled to do so. Two HST campaigns have been done with HST in 1992 and 1994, coupled with simultaneous ground-based multi-site campaigns, including sites in Chile, Australia, South Africa, New Zealand, Brazil, so as to monitor the overall variability of the lines, through the observations of the CaII lines (Lagrange et al. 1996). This was absolutely mandatory as the variations may be as short as hours, and as the observations with HST in the various wavelengths domains were spread over typically three days.
Observations with GHRS gratings, Echelle B and Echelle A after refurbishment, enabled to detect several new circumstellar lines MnII, SiII, SI, FeI, FeII, ZnII, CrII, AlII, AlIII, ZnII, CI, highly excited levels of FeII (up to 3eV) (Figure 1), and to analyze them in details( Lagrange et al. 1995).
Figure: Neutral and singly ionized species in the gas around 
Pictoris, observed with Echelle A and Echelle B (except CaII, observed
with the ESO/CAT and CES)
CIV, marginally detected with IUE could also be confirmed (Vidal-Madjar et al. 1994). The stable and variable components could be separated, enabling thus to perform quantitative measurements.
The results of HST observations are now summarized both for the stable and variable gas.
Figure: log(N(X))-log(N(X)/N(H))
for the stable gas around
Pictoris (see text)
Figure 2 gives the value log(N(X))-log(N(X)/N(H))
for the stable
gas around 
Pictoris.
It clearly shows that the refractory elements have
standard abundances, very different from interstellar medium
ones. This suggests that the
origin of the gas is either evaporation from solid material (comets or
grains) or stellar. A definite way to distinguish between both
possibilities is to observe volatiles in the gas. The observations also show
that the amount of neutral hydrogen is less than 10
cm
, in
agreement with the upper limit given by radio measurements (Freudlin et
al. 1995).
The observations also enabled us to progress in the knowledge of the physical
conditions in the stable gas. In particular, limits for the electronic
density and temperature could be derived : 10
-10
cm
(Lagrange et al. 1995, Boggess et al. 1991).
The study of the chemical composition of the variable gas appeared to be extremely difficult due to the strong short time-scale variability during the observations. It seems nevertheless that the chemical composition of the refractory elements in the variable gas is again standard, which is in agreement with the EIB scenario. Simultaneous observations over the whole UV domain with the next generation of HST instrument are strongly needed to confirm this result and to further progress in the analysis.
Figure: HST observations of the CO molecule
A very important discovery with HST is the one of the CO molecule (Figure 3;
Vidal-Madjar et al. 1994). Noticeably this is the very first molecule
detected so far in the circumstellar gas around 
Pictoris
and radio observations
have failed to detect molecules (Liseau & Artymowicz 1996). The CO
lifetime around 
Pictoris being very short (hundreds of years), the gas needs to
be replenished in CO. Comets can account for this replenishment
(Lecavelier et al. 1996). Moreover, the recent observations seem to
indicate that 
CO is detected. The determination of the isotopic
ratio 
CO/
CO is under work. The temperature
deduced from the data is about 20K, which suggests that the gas is produced
far enough from the star.
Figure: Variability in the MgII lines : a) superposition of two spectra
observed some hours apart; b) division of these two spectra to visualize
the variability; c) simulation of the variability: the agreement with b)
is remarkably good
HST has enabled us to study in details the structure and the evolution of the UV variable lines. The ones of singly ionized species seem to be well correlated with the CaII ones observed in the visible. Moreover, the structure of some variable lines is well reproduced by the EIB scenario (Figure 4).
HST observations enabled to favorably test the EIB scenario; the detection
of molecular CO is very important in this context. The forthcoming Echelle
spectrograph will enable to further progress in the chemical and
dynamical study of the infalling gas. The observations also permitted to
progress in the characterization of the circumstellar stable gas around
Pictoris, in particular its chemical composition
Beust, H. et al. 1989, A&A, 223, 304
Beust, H. et al. 1991, A&A, 241, 488
Beust, H. & Tagger, M. 1993, Icarus 106, 42
Beust, H. & Morbidelli 1995, Icarus, in press
Beust, H. et al. 1996, A&A, in press
Boggess, A. et al. 1991, ApJ, 377, L52
Deleuil, M. et al. 1993, A&A, 267, 193
Ferlet, R. et al. 1993, A&A, 267, 137
Freudling, et al. 1995, A&A, 301, 231
Hobbs, L.M., et al. 1988, ApJ, 334, L41
Knacke, R.F., Fajardo-Acosta, S.B., & Telesco, C.M. 1993, ApJ, 418, 440
Lagage, P.O., & Pantin, E. 1994, Nature, 369, 628
Lagrange, A.-M., Ferlet, R., & Vidal-Madjar, A. 1987, A&A, 173, 289
Lagrange-Henri A.-M. et al. 1989, A&A, 215, L5
Lagrange-Henri A.-M. et al. 1992, A&A, 264, 637
Lagrange A.-M. et al. 1995, A&A, 296, 499
Lagrange A.-M. et al. 1996, A&A, in press
Lanz, T., Heap, S.R., & Hubeny, I. 1995, ApJ, 447, L41
Lecavelier, A., Vidal-Madjar, A., & Ferlet, R. 1996, A&A, in press
Levison, H.F. et al. 1994, Nature, 372, 441
Liseau, R. & Artymovicz, P. 1996, A&A, in press
Mouillet, D. & Lagrange, A.M. 1995, A&A, 297, 175
Mouillet, D., Lagrange, A.-M., & Beuzit, J.-L. 1996, A&A, in press
Slettebak, A. & Carpenter, K.G. 1983, ApJS, 53, 869
Smith, B.A. & Terrile, R.J. 1984, Science, 226, 1421
Paresce, F. 1991, A&A, 247, L25
Vidal-Madjar, A. et al. 1986, A&A, 167, 325
Vidal-Madjar, A. et al. 1996 A&A, 290, 245