Figure 4.3.2/1: The frequency distribution of [Fe/H] in an unbiased sample of G (continuous line) and K (dashed line) dwarfs in the solar neighbourhood.
Data from Hipparcos were used by M.A.C. Perryman, in collaboration with members of the Hipparcos scientific teams, to determine distances to the recently discovered planetary system candidates, 47 UMa, 70 Vir and 51 Peg (Perryman et al., 1996). Standard errors of the absolute trigonometric parallaxes are 0.66-0.76 mas (milliarcsec), with resulting distances accurate to around 1%. For 70 Vir, the Hipparcos parallax (55.22±0.73 mas) resolves a discrepancy of almost a factor of three in published trigonometric and photometric parallaxes. The residuals of the astrometric parameters were used in combination with the radial velocity data for these orbital systems to place upper limits on the companion masses of 7-22 and 38-65 MJupiter for 47 UMa and 70 Vir, respectively, with less stringent limits for 51 Peg. The Hipparcos results provided confirmation of the existence of sub-stellar masses significantly below the brown dwarf limit (of about 0.08 MSun ) surrounding other stars.
Reference
Perryman, M.A.C. et al. (1996). A&A 310,
L21.
The Hipparcos 30-month solution was used by M.A.C. Perryman, in collaboration with members of the Hipparcos scientific consortia, to provide powerful confirmation of the Hipparcos astrometric data quality (Perryman et al., 1995). On the basis of approximately 20 000 accurate trigonometric parallaxes, the Hipparcos results demonstrated a well-defined broad main sequence extending to M=-5 mag, a well-defined giant branch, and a distinct degenerate sequence. Solar neighbourhood 'clump giants', stars in a post-helium flash stage of evolution, are located in the HR diagram on the basis of trigonometric parallax distance estimates for the first time, and the first direct luminosity calibration of Mgiants is reported. A comparison with stars previously considered to lie within 25 pc demonstrated significant discrepancies for those distances less reliably determined from ground-based observations, and indicated that a corresponding re-evaluation of stellar masses and luminosities within the solar neighbourhood is implied.
Reference
Perryman, M.A.C. et al. (1995). A&A 304,
69.
D. Hestroffer has completed the construction of the solar system parts (astrometry and photometry) of the Hipparcos and Tycho Catalogues. This concerns the validation and reduction of the data to astrometric positions in the International Conventional Reference System (ICRS). The photometry part consists of apparent magnitudes in the Hipparcos system Hp or in the Tycho VT and BT bands. The positions gathered by Tycho on the Galilean satellites J III Ganymede and J IV Callisto confirm the discrepancy between the optical positions of Jupiter obtained with the Carlsberg instrument at La Palma and the DE403 ephemeris of the Jet Propulsion Laboratory, which are essentially fitted to VLBI and radar observations.
The very high quality Hipparcos astrometry yield a dramatic improvement in the determination of the extragalactic-dynamical frame tie from analysis of the Hipparcos minor planets positions, providing precisions competitive to the tie obtained by lunar laser ranging measurements. Minor planet ephemerides improvement, mass determinations and establishment of the dynamical reference frame from Hipparcos observations of minor planets is being undertaken in collaboration with M. Rapaport and B. Viateau (Bordeaux Observatory).
Analysis of the Hipparcos minor planet photometry in order to gather information on their pole orientation and triaxial ellipsoid shape is being undertaken with A. Cellino and V. Zapalla (Torino Observatory) and F. Mignard (CERGA, Grasse). Observations with a modulating grid enhance the theoretical resolution; thus the modulated part of photometric measurements yield information on the size and limb darkening in a way similar to observations with an interferometer. Analysis of the Hipparcos 'modulated photometry' of minor planet 1 Ceres is being undertaken with F. Mignard. In parallel, studies on the determination of diameter and limb darkening of celestial objects from interferometric observations (especially with Gaia) are being undertaken in collaboration with F. Vakili (CERGA, Grasse).
F. Favata and collaborators have continued their study of solar-type stars in the solar neighbourhood, in particular the behaviour of the elemental abundances in solar-type stars. Disc population main sequence K dwarfs are, as a group, relatively little studied, so that very little was known on their abundance distribution. K dwarfs have deep convection zones, and thus the light element abundances are expected to evolve rapidly. Lithium in particular should in principle be depleted, according to most stellar evolution models, in the pre-main sequence phase, so that no detectable lithium is expected in K dwarfs a few tens of millions of years old.
The study of lithium in the cooler dwarfs is made difficult by the presence of a large number of both atomic and molecular spectral lines in their spectra, which completely blanket the continuum and which cause severe blending. Their study thus requires the highest possible spectral resolution as well as a detailed spectral synthesis analysis. Using updated atmospheric models and line lists and spectral synthesis techniques, Favata et al. (1995) have shown that, unless there is an unidentified line with the same wavelength as the lithium line (within a few mÅ), measurable lithium abundances are often found in disc- population K dwarfs.
A large-scale study of the lithium abundance of a volume- limited sample of solar-type dwarfs in the solar neighbourhood (Favata et al., 1996a) has shown that such a small, yet measurable abundance of lithium is indeed a common feature of a large fraction of the K-type dwarfs found in the solar neighbourhood, being found in more than one third of the stars. Given that lithium is expected to be burned quickly at the bottom of the deep convection regions of such stars, one possible hypothesis is that the detected lithium is the product of spallation reactions between the heavy nuclei in the photosphere and the high energy nuclei being produced during flares in the stellar coronae. Favata et al. (1996b) have shown that such an hypothesis is not incompatible with the coronal energetic balance of small mass dwarfs.
A similar study, conducted on an X-ray selected sample (Favata et al., 1996c), has confirmed the above results, additionally showing the presence, in the solar neighbourhood, of a small group of stars with high lithium abundances and high activity levels, indicative of very young ages. These stars share common kinematics, suggesting a very young kinematic stream.
No data on the metal (and in particular Fe) abundance of unbiased samples of K-type dwarfs were available until now in the literature, yet the distribution of abundances in such samples is a fundamental observational test of the chemical evolution of the Galaxy. In fact, the distribution of Fe abundance in unbiased samples of G dwarfs has traditionally proved to be very hard to reconcile with galactic chemical evolution models, a problem so fundamental as to merit the name of 'the G star problem'. Recent progress in chemical evolution theory has alleviated the G star problem somewhat, but additional observational data are certainly necessary. To this end, Favata et al. (1996c) have determined the Fe abundance in the same unbiased sample of K dwarfs as used for lithium abundance determination. The surprising result is that the [Fe/H] frequency distribution of K-type dwarfs appears to be substantially different from the same distribution for G-type dwarfs selected with the same unbiased criteria.
In particular, there appears to be a significant deficit of metal-poor K dwarfs with respect to G dwarfs. In a sense, there is thus a harder 'K star problem' to solve. While the size of the sample investigated allows conclusions to be drawn only at a ~95% statistical uncertainty level, they will soon be verified on a much larger sample of K dwarfs, which is being studied by a group at Lick Observatory. The implications of such a deficit of metal- poor K dwarfs are potentially very wide-ranging.
A direct comparison of the [Fe/H] frequency distribution of G and K dwarfs (Fig. 4.3.2/1) implicitly assumes that the mass function is a universal function of space and time, i.e. that the same relative fraction of G and K dwarfs is always formed, independent of crucial initial conditions such as the metallicity of the parent gas cloud. A natural explanation for the observed difference in abundance distribution can be found assuming that the initial mass function is dependent on the metallicity of the parent cloud, with metal-poor clouds forming more higher mass stars. Interestingly, recent studies of star-forming regions do, in fact, show that the initial mass function is not at all universal, with different star-forming regions showing differences in the relative number of low and intermediate mass stars.
Figure 4.3.2/1: The frequency distribution of [Fe/H] in an
unbiased sample of G (continuous line) and K (dashed line) dwarfs
in the solar neighbourhood.
References
Favata, F., Micela, G. & Sciortino, S. (1995).
A&A 297, L1.
Favata, F., Micela, G. & Sciortino, S. (1996a).
A&A, sub-mitted.
Favata, F., Micela, G. & Sciortino, S. (1996b).
A&A 311, 951.
Favata, F., Micela, G. & Sciortino, S. (1996c).
A&A,in press.
Work in this area has been pursued by C.V.M. Fridlund and collaborators. The first-epoch HST images of the L1551 IRS5 jet have been obtained and reduced (Fig. 4.3.2/2). Although the goal is to image this jet in two more cycles of HST observations in order to discern changes and ultimately to estimate the mass and energetics of the jet, the morphological structure has already proved interesting. This is the first observation of an astrophysical jet or an HH object this close to the Earth (150 pc); the details that can be resolved are smaller than 7.5 AU. For the first time, the observational resolution elements are equal to those used in most hydrodynamic codes.
Figure 4.3.2/2: I-band (F814W) HST WFPC2 image of the L1551
IRS5 jet. The low mass young stellar object is located
approximately 1 arcsec towards the NE (upper left corner) of the
apex of the visible jet, and is hidden behind 150 magnitudes of
visual extinction, presumably caused by a dense accretion disc
viewed edge-on. The jet appears out of this disc and expands
rapidly (within 25 AU), after which it collimates. All these size
scales are much smaller than hitherto assumed in model
calculations of astrophysical jets.
The last of the IUE data on the Herbig-Haro object HH29 have been published (Liseau et al., 1996). Taken together with earlier imagery (using the ESA Photon Counting Detector camera), they yield strong indications of very small (~10 AU) and very dense (=106/cm3) clumps in the interstellar medium in star-forming areas. Indications of such clumps were found earlier by inference radio data, but these are the first direct observations of such entities. A proposal to image them in the UV with HST has been submitted. High resolution echelle spectra (obtained with the ESO NTT 3.5 m telescope) of this object have been reduced and interpreted. This HH object (which is associated with the L1551 IRS5 jet) shows a velocity field that cannot be interpreted in the context of the 'jet-model', which is the current paradigm for interpreting Herbig-Haro objects.
After 7 years, the useful life of the ESA PCD II photon counting detector has come to an end. The time-resolved mode was utilised in observations of EQ Pegasi, a flare star, during an eruption. Using the PCD as the detector on a Boller & Chivens spectrograph mounted on the 2.6 m Nordic Optical Telescope, a region of the spectrum around H gamma was monitored for 5 h during the event. More than 20 emission lines were seen to appear and, later, to disappear.
The effective time resolution in the data is as small as 10 s (for lines with good signal/noise ratio). Molecular line observations of the bipolar molecular out-flow in L1551 have been reduced and are being interpreted. The material, which consists of 2.6 mm CO and 13CO J=1-0 data from 14 positions within one of the outflow lobes, has an unprecedented quality in terms of signal-to-noise. It can now be more or less taken as proved that the outflow lobe is filled with (relatively) slow moving molecular gas, as well as atomic gas expanding at a much higher velocity.
Reference
Liseau, R., Huldtgren, M., Fridlund, C.V.M. & Cameron,
M. (1996). A&A 306, 255.
G. Pilbratt and C.V.M. Fridlund have launched a two-pronged observational programme to unravel the physical structure of the S88B region. This bright H alfa region, located in the vicinity of the main diffuse part of S88, features spatially displaced peaks of emission in the optical, near-IR, far-IR, carbon monoxide (CO) and radio continuum, suggestive of substantial and varying extinction, and complicated energetics possibly connected with massive star formation.
Building on previous work by White & Fridlund (1992), the goal is to map S88B in the J=1-0 transition of 12CO and 13CO (at 115 GHz and 110 GHz, respectively) over a greater extent, and with better sampling and sensitivity than has been done before, using the Onsala 20 m telescope in Sweden. We have also obtained complementary, well-calibrated, sub-arcsec resolution optical images in a number of shock diagnostic lines (Halfa,Hbeta, O[III], S[II] and O[I]) using the CCD/ESA PCD II detectors on the Nordic Optical Telescope telescope on La Palma, Spain.
Most of the data needed to produce a 12CO map covering a region of 10 10 arcmin on the sky with 30 arcsec spacing (~ one beamwidth) have been obtained. It is clear that both the line strength and shape are rapidly and greatly varying over the mapped region, and in some areas even between adjacent points. By integrating the emission in narrow velocity bins across the map, it has been shown conclusively that this source exhibits 'red' and 'blue' high velocity (relative to the cloud velocity) wings of CO emission normally associated with molecular outflows. The 'blue' wing emission is very well collimated, while the 'red' wing appears bifurcated. S88B may be argued to exhibit 'tripolar' emission.
The location of the optical nebulosity, right on the 'edge' of the peak in total (velocity) integrated CO emission, and superposed on part of the 'red' CO lobe (Fig. 4.3.2/3), indicates that the optical emission could be connected to the outflow activity. The very intense [SII] 6717, 6731 Å and spatially collocated H alfa emission indicate that the nebulosity could indeed be shock excited, rather than radiatively excited. In this picture, the nebulosity would not be a classical HII region but would instead be excited by an atomic outflow (with significantly higher velocity than the molecular gas) interacting with ambient gas. Using a model of Hartigan et al. (1987), a shock velocity of 100 km/s with a pre-ionisation density of 300/cm3 has been derived.
Figure 4.3.2/3: The 'blue' (short-dashed) and 'red' (long-
dashed) emission from CO plotted together with the H alfa (solid)
emission. Units are in arcmin.
The conclusion is that the nebulosity is essentially a Herbig- Haro object, albeit within a high mass star-forming region. The caveats are that the [OI] 6300 Å emission is relatively weak and that the [OIII] 5007 Å emission peak is displaced from those of H alfa and [SII], and in the direction of the embedded source, which is presumably the originating source of the molecular outflow. An alternative interpretation is that the optical nebulosity is a so-called 'bright-rim' structure.
Reference
White, G. J. & Fridlund, C. V. M. (1992). A&A
266, 452.
J. Tauber and collaborators continued work on the data collected during the flight of the Pirog 7 experiment in May 1994. Careful analysis of the data has given additional certainty to the preliminary results reported earlier. Observations of the Orion core in the 557 GHz ground state transition of H2O show no hint of the broad, centrally peaked line emission seen in the recently detected H218O isotopic counterpart at 548 GHz. This fact suggests that the abundance of water in the core is large and its emission optically thick. On the other hand, a weak and narrow absorption feature was detected towards Orion. This feature can be interpreted as the result of absorption by a foreground ridge-like component of bright and wide emission arising from the warm gas in the high velocity core of Orion. This interpretation is consistent with the available information on the physical characteristics of the gas in this region. The results of the analysis have been published (Tauber et al., 1996; Olofsson et al., 1996). The next flight of Pirog (planned for May 1997) will focus on observations of the O2 line at 424 GHz, requiring a new (Solid State Imaging Spectrometer) receiver, which will be provided by the DEMIRM laboratory. J. Tauber maintains an active involvement in the experiment, being responsible for the autocorrelator spectrometer.
References
Olofsson, G., Nordh, L., Tauber, J., Pilbratt, G. &
Frisk, U. (1996). IAU Symposium 170, (Ed. J. Mangum),
Kluwer, Dordrecht.
Tauber, J.A., Olofsson, G., Pilbratt, G., Nordh, L. &
Frisk, U. (1996). A&A 308, 913.
J. Tauber continued work on the analysis of high S/N observations of emission from HCO+ obtained at the Onsala Space Observatory in 1994, in an attempt to develop diagnostics of the small-scale spatial and velocity structure of molecular clouds. A new model for analysis of radiative transfer through clumpy molecular clouds was developed, which extends and includes previous simpler ideas. The model was applied to observations of Orion and M17. The model and applications were presented in preliminary fashion at IAU Symposium 170, and a paper on this subject (Tauber, 1996) appeared in A&A.
Reference
Tauber, J.A. (1996). A&A 315, 591.
J. Tauber and M. Fridlund, together with T. Liljestrom (Helsinki) and G. Olofsson (Stockholm), have used the BIMA mm- wave interferometer to observe the double star Z CMa, an FU- Orionis type object which has a dense disc aligned perpendicularly to a molecular bipolar outflow. The observations are being used to study the intermediate region between the small far-IR disc and the larger molecular envelope. Compact array observations were obtained in the autumn of 1995, and processed into 10 arcsec-resolution maps. The source is clearly detected, but it is now clear that the higher resolution of the B-array is required to resolve the details of the structure on this source, and finish the project. A new proposal has been submitted to BIMA to carry out these observations in the winter of 1996/1997.
J. Tauber, M. Fridlund, G. White (London) and M. Huldtgren (Stockholm) have embarked on a study of a number of cometary globules associated with the open cluster IC 1848. These objects are dense molecular condensations, illuminated by massive young stars, and as a consequence probably undergoing Radiation Driven Implosion, leading to star formation. Our aim is to understand the physics of these regions, using molecular line observations to study the physical properties, morphology and kinematics of the cores of the globules, and optical emission images to study the radiation- or shock-excited regions and the layers forming the transition to the surface of the globules. During 1995 and 1996 we collected observations of four of these globules in a number of molecular emission lines (the J=2-1, J=3-2 and, in some cases, J=4-3 transitions of CO and 13CO) at the James Clerk Maxwell Observatory, and of optical emission lines (H alfa, [SII], [OI] and [OIII]) at the Nordic Optical Telescope. Analysis is ongoing. Part of the data was presented at IAU Symposium 178 in April 1996 (Tauber et al, 1996).
Reference
Tauber, J. et al. (1996). IAU Symposium 178, April 1996.
White, G.J., Lefloch, B., Fridlund, C.V.M., Aspin, C.A., Minchin,
N.R. & Huldtgren, M. (1996). A&A,in press.
P. Jakobsen, with D. Tytler (UC San Diego), carried out follow-up HST spectroscopic observations of the z=3.17 quasar PKS1935-692 using the FOC objective prism. The initial HST/FOS observations of this object had led to the second detection of intervening intergalactic ionised helium towards a quasar, but seemingly less intense than seen in the first detection towards Q0302-003 at slightly higher redshift (Jakobsen et al., 1994). The FOC observations confirm, however, that the two HST detections of HeII absorption are mutually consistent. Combined with the third detection made at lower redshift (z~2.4) with the Hopkins Ultraviolet Telescope by Davidsen et al., the HeII observations suggest that complete reionisation of the intergalactic medium did not occur before z~3.1.
In anticipation of the capabilities of future X-ray missions such as XMM, K.S. O'Flaherty and P. Jakobsen carried out a theoretical study of the expected X-ray opacity of quasar absorption line systems out to large redshift. Although the cumulative photoelectric absorption in quasar absorption line gas provides a significant source of opacity at the softest (E<1 keV) energies, the imprint of this absorption is not readily observable since for most lines of sight it is masked by much stronger Galactic interstellar foreground absorption. This confirms that the 'excess' absorption detected in the ROSAT spectra of many high redshift quasars is most likely of intrinsic origin.
Work also continued on using quasar absorption to map large- scale structure at high redshift. With A. Picard, P. Jakobsen obtained exploratory spectra of ten quasars surrounding the rich absorption object Q 0122-380 using the ESO 3.6 m telescope at La Silla.
References
Jakobsen, P., Boksenberg, A., Deharveng, J.M., Greenfield,
P., Jedrzejewski, R. & Paresce, F. (1994). Nature
370, 35.
O'Flaherty, K.S. & Jakobsen, P. (1997). ApJ, in
press.
A.N. Parmar, E. Kuulkers and O.R. Williams, in collaboration with N.E. White and L. Angelini from the Goddard Space Flight Centre and M. van der Klis from the University of Amsterdam, have continued their archival investigations of the X-ray transient 4U 1630-47. This source is an optically unidentified ultra-soft X-ray transient discovered by Uhuru, with the first recorded outburst in 1969 detected by Vela 5B. The first five outbursts were observed by a combination of the Vela 5B, Uhuru, OSO 7 and Ariel V satellites and seemed to occur every 7600 days. A strict outburst periodicity was apparently ruled out by an extended 1977 outburst, which started 770 days later than predicted and may have lasted for up to 6 months. In 1984 the source underwent another outburst detected by Tenma, the decay of which was observed by EXOSAT (Parmar et al., 1986). Recently, an archival search revealed the presence of two previously unreported outbursts (Parmar et al., 1995). The first was observed by the Einstein SSS in 1979 and the second by the ROSAT PSPC in 1992. The times of these outbursts are consistent with the previously reported 7600 day outburst recurrence interval, allowing Parmar et al. (1995) to constrain the recurrence interval at 601.7±3.0 days (see Fig. 4.3.2/4).
Figure 4.3.2/4: The time residuals in days after a linear fit
of the outburst times to cycle number. The error bars reflect the
uncertainties in deriving the outburst start times. The dotted
line shows the best-fit ephemeris of Parmar et al. (1995), which
excludes the cycle 11, 12, 15 and 16 measurements reported in
Parmar et al. (1996).
The ultra-soft/ultra-hard spectral shape at low and high energies, the change in relative contributions of the two spectral components observed by EXOSAT (Parmar et al., 1986) and the lack of detected pulsations or bursts, are all indicative of a black hole nature for the compact object in 4U 1630-47. Similar behaviour is seen from other X-ray transients, such as A 0620-00, GS 1124-684 and GS 2023+338, which are believed to contain black holes on the basis of their dynamically-derived masses. There are now at least 15 similar black hole X-ray transient (BHXT) systems, of which five are known to recur. The other BHXT systems have been observed only once during the 25 years of satellite observations. This suggests a typical recurrence timescale of 10- 50 years. The more prolific outburst activity of 4U 1630-47 is therefore unusual and provides the best-known laboratory for studying outburst to outburst variations amongst these systems. In addition, for the first time, archival Einstein observations revealed clear evidence for low level inter-outburst activity (Parmar et al., 1996).
The increased interest in 4U 1630-47 inspired the authors to reanalyse the EXOSAT observations in order to characterise the rapid X-ray variability observed from this source using analysis techniques more modern than were available at the time of EXOSAT's operations. In particular, the timing properties of a number of black hole candidates have now been characterised into low, high and very high states. A reanalysis of the EXOSAT observations reveals that the timing characteristics of 4U 1630- 47 are at times inconsistent with those of other BHXT systems, except those systems that exhibit evidence for superluminal expansion, such as GRS 1915+105 and GRO J1655-40, which have different timing properties. In Kuulkers et al. (1996), we speculate that the timing behaviour of 4U 1630-47 may be connected with the presence of relativistic jets, and therefore that 4U 1630-47 is also a superluminal source. We await the next outburst from 4U 1630-47 with great interest, especially since we have guaranteed SAX observing time to follow the evolution of the outburst.
References
Kuulkers, E., van der Klis, M. & Parmar, A.N. (1996).
ApJL, in press.
Parmar, A.N., Stella, L. & White, N.E. (1986).
ApJ 304, 664.
Parmar, A.N., Angelini, L. & White, N.E. (1995).
ApJ 452, L129.
Parmar, A.N., Williams, O.R., Kuulkers, E., Angelini, L.
& White, N.E. (1996). A&A, in press.
SP1211