We report first results from a study comparing directly observable properties of Galactic planetary nebulae with synthetic models for the PN population. In this way we avoid problems such as deriving distances for individual PN.

We have developed a new procedure for the generation of a planetary nebula luminosity function (PNLF), improving on previous work (Mendez et al. 1993 A&A 275, 534). The procedure is based, as before, on an exponential central star mass distribution, on H-burning post-AGB evolutionary tracks, and on the avoidance of nebular models for the calculation of nebular fluxes. We have added new post-AGB evolutionary tracks (Blöcker 1995 A&A 299, 755) and introduced the following improvements:

We study the local population of AGB stars, and determine their densities and scale heights according to their masses, following the classification scheme proposed by Ortiz and Maciel (1994, A&A 287, 552), which discriminates the precursors of type I, II, and III planetary nebulae. The formation (or death) rates of AGB stars are determined by means of stellar evolution models, and are compared to the formation rates of planetary nebulae from the literature. A good agreement is found between these rates; however, the results are very sensitive to the adopted model, especially regarding the time spent at the thermal-pulse AGB phase.(FAPESP, CNPq)

Hb 7 (003.9 - 14.9) and Sp 3 (342.5 - 14.3) are high galactic latitude planetary nebulae. Ultraviolet spectra obtained with the IUE reveal UV stellar continuum and P - Cygni stellar wind profiles of NV and CIV indicating post-AGB mass loss from the central stars of these two PN. From an analysis of optical and ultraviolet spectra we have derived the parameters of the central stars and nebulae. The IUE ultaviolet continuum shows that the central stars of Hb 7 and Sp3 are 08 stars. The Si IV 1400A doublet in the spectra of both these stars is very weak or absent most likely due to underabundance of Si and due to also due to earlier spectral type (O 8). The Si IV lines in hot stars are sensitive to surface gravity, their weakness suggests that surface gravity the central stars of Hb7 and Sp3 is relatively higher. The terminal velocity of the stellar wind in Hb7 and Sp 3 are found to be −2460 km/sec and −3630 km/sec respectively.

Planetary nebulae have recently been shown to be useful as standard candles (Ciardullo et al. 1989, ApJ, 339, 53; Jacoby 1989, ApJ, 339, 39). Distances to many galaxies have been determined by fitting a planetary nebula luminosity function (PNLF) to observations of the OIII 5007å line of PNe. Here, the effect of the core mass distribution on the determination of the luminosity function is investigated and a technique for interpolating between model evolutionary tracks is discussed.

Planetary Nebulae (PN) are bright emission line objects, observable at large distances throughout the Galaxy. They serve as probes of abundance gradients and chemical enrichment history of the ISM.

We present the results of our major HST study of the evolution of PN in the Magellanic Clouds. This consists of imaging studies in [O III] and FOS UV spectroscopy. These data are then used in theoretical photoionisation models in conjuction with ground-based spectrophotometry, absolute flux and expansion velocity and density to derive self consistent diameters, ages, masses, and nebular abundances and to accurately place the central stars on the H-R Diagram. We find that observed sizes and ages can be reconciled with evolutionary theory provided that the He-burners outnumber the H-burners in the approximate ratio 2:1. For the LMC observed abundance patterns are qualitatively consistent with the (mass-dependent) operation of the various chemical dredge-up processes as predicted by theory. However, the observed dredge-up efficiencies do not agree with current theory. Finally, since core masses are determined with adequate precision, we are able to derive, for the first time, the metallicity age relation for of the LMC. We find that the base metallicity of the LMC rapidly increased ∼ 2 Gyr ago, consistent with the age of the burst of star formation inferred from field stars and clusters.

At the current epoch, the dynamics of galactic systems is largely controlled by the behaviour of the stellar component. There is probably also a large amount of non-stellar matter present, which probably dominates the dynamics on large scales. Nevertheless, the (centers of) galaxies are still, for better or worse, defined as those regions were there happens to be a lot of stellar light.

The key properties of elliptical galaxies are compared to present-day models calling attention on several points of contradiction. The nature of the UV fluxes and their usefulness to cast light on the age-metallicity dilemma are discussed. Finally, preliminary results of an attempt aimed at reconciling theory with observations by means of an new IMF are presented.

I review the arguments favoring the high reliability of the PNLF method. Agreement with Cepheid distances is better than 8%, and consistency for multiple PNLF distances within clusters is even better. Agreement between Cepheid distances to spirals and PNLF distances to ellipticals within the same cluster also is excellent. In order for the PNLF method to work despite the vast diversity of properties seen among PN, several factors must operate. Most importantly, the progenitors of the bright extragalactic PN probably have ages less than ∼10 Gyrs. Also, very young PN must either be absent from the bright extragalactic samples or must have lower luminosities than suggested by their predicted central star masses. The latter may be due to internal dust, nitrogen enrichment that competes with oxygen for ionizing photons, and/or a tendency for massive PN to be optically thin to ionizing radiation. In addition, models that reproduce observed PNLFs rely on theoretical evolutionary tracks of central stars, and so these tracks also must be reasonably correct. PNLF observations suggest, however, that the slope of the initial-to-final mass relation is shallower than the Weidemann (1987) relation.

The extragalactic distance scale depends heavily on one primary distance indicator: Cepheid variables. Given the conflict between stellar ages and the age of the Universe implied by H0, it is worth checking the zero point with as many independent methods as possible.

The MACHO microlensing experiment's time-sampled photometry database contains blue and red lightcurves for nearly 9 million stars in the central bar region of the Large Magellanic Cloud (LMC). We have identified known LMC Planetary Nebulae (PN) in the database and find one, Jacoby 5, to be variable. We additionally present data on the “parent populations” of LMC PN, and discuss the star formation history of the LMC bar.

As part of a broad study of stellar winds in different galaxies, to establish observational constraints on the influence of metallicity on the mass loss process, we investigate the winds of planetary nebula (PN) nuclei in the Large Magellanic Cloud. Theoretically, the radiation pressure mechanism implies a strong dependence of the wind acceleration on the element abundances. Observationally, model predictions have been tested using Pop. I stars in the Magellanic Clouds (Puls et al. 1996, A&A, 305, 171) and massive stars in the nearby galaxies M31, M33 and NGC 6822 (Bianchi et al. 1996, AJ, 111, 2303; Bianchi et al. 1996, in prep.). We now extend this investigation to low luminosity, evolved objects.

In the present work we have computed the physical parameters and chemical abundances for 45 planetary nebulae (PN) in the Large Magellanic Cloud (LMC) using the photoionization code CLOUDY, developed by Ferland (1993). CLOUDY is used as a subroutine in the code DIANA, developed by Elizalde & Steiner (1996), which minimises indices that measures the difference between the calculated and real nebula.

We use planetary nebulae major- and minor-axis kinematics (Hui et al. 1995) of the dust-lane elliptical galaxy NGC 5128 (Centaurus A) to build triaxial dynamical models.

Several weeks after the explosion of supernova (SN) SN1987A, the UV flash of the SN illuminated a ring-like structure in the circumstellar material at about 0.65 ly from the SN. The interaction between the stellar winds from the SN progenitor is considered to be the candidate for the formation of the circumstellar structure. In the case that the stellar winds are spherically symmetric, the interaction should result in a shell-like structure. However, Washimi, Shibata & Mori (1996) show that the magnetic field in the winds causes an anisotropy which leads to the formation of a ring-like structure. When the fast wind of the blue supergiant phase of the progenitor sweeps up the surrounding slow wind of the red-supergiant phase, the magnetic field as well as the wind material are piled up in the interaction region. Since the magnetic energy increases in proportion to the square of the amplitude, the magnetic field exhibits its effect prominently at the interaction region; due to the magnetic pressure force the material at lower latitudes is compressed into a ring-like structure. It is suggested that this magnetic process can also explain the newly observed pair of rings of the SN1987A nebula. We note that the idea of a magnetic field effect is consistent with the radio observation of a supernova remnant, detected by Staveley-Smith et al. (1992) at about 1200 days after the explosion. This radio emission is explained by the collision of the supernova blast wave with the shocked blue wind. This position corresponds to the averaged expansion speed of the supernova ejecta ∼ 0.08 ly which is consistent with the estimation by Shigeyama and Nomoto (1990). The estimated magnetic-energy density by the minimum-energy argument is ∼ 4 × 10–8f–4/7N m–2, where f is the fractional volume of the radiating acceleration region, suggesting a magnetic field of a few milli-Gauss or more (Chevalier 1992). This field intensity is consistent with an intensity of ≈ 2 · 10–4 Gauss obtained between the reverse shock and the contact surface shown, if one takes into account a further enhancement of the field due to the sweeping-up process by the supernova blast wave. When the SN ejector collides with the ring at the end of this century or at the beginning of the next one, we can also expect more intense radio emission at rather middle and high latitudes where the magnetic intensity is greater, rather than at the equator where the ring-like structure is located.

We consider the possibility of using bright planetary nebulae to study the chemical evolution of their parent galaxies. In particular, we would like to use planetary nebulae as abundance probes in galaxies where star formation ended long ago, such as ellipticals or the bulges of spirals, and in which no other direct probes exist. However, if we are to use planetary nebulae to study the chemical evolution of their parent galaxies, we must relate the oxygen abundance in bright planetary nebulae to the oxygen abundance in the interstellar medium when star formation stopped. Obviously, only theoretical modelling can provide this information once star formation has stopped.

We have obtained spectra of 28 planetary nebulae in the bulge of M31 using the MOS spectrograph at the Canada-France-Hawaii Telescope. Typically, we observed the [O II] λ3727 to He I λ5876 wavelength region at a resolution of approximately 1.6 å/pixel. For 19 of the 21 planetary nebulae whose [OIII]λ5007 luminosities are within 1 mag of the peak of the planetary nebula luminosity function, our oxygen abundances are based upon a measured [OIII]λ4363 intensity, so they are based upon a measured electron temperature. The oxygen abundances cover a wide range, 7.85 dex < 12 + log(O/H) < 9.09 dex, but the mean abundance is surprisingly low, 12 + log(O/H)–8.64 ± 0.32 dex, i.e., roughly half the solar value (Anders & Grevesse 1989). The distribution of oxygen abundances is shown in Figure 1, where the ordinate indicates the number of planetary nebulae with abundances within ±0.1 dex of any point on the x-axis. The dashed line indicates the mean abundance, and the dotted lines indicate the ±1 σ points. The shape of this abundance distribution seems to indicate that the bulge of M31 does not contain a large population of bright, oxygen-rich planetary nebulae. This is a surprising result, for various population synthesis studies (e.g., Bica et al. 1990) have found a mean stellar metallicity approximately 0.2 dex above solar. This 0.5 dex discrepancy leads one to question whether the mean stellar metallicity is as high as the population synthesis results indicate or if such metal-rich stars produce bright planetary nebulae at all. This could be a clue concerning the mechanism responsible for the variation in the number of bright planetary nebulae observed per unit luminosity in different galaxies (e.g., Hui et al. 1993).

The majority of planetary nebulae (PNe) at the distance of the Magellanic Clouds typically subtend 1 arcsec or less on the sky. The Planetary Camera onboard the Hubble Space Telescope has a scale of 0.044 arcsec per pixel and is therefore ideal to use to image these objects. At present, 16 LMC and 5 SMC objects have been imaged in the [O III] 5007 angstrom emission line, as part of this program. An additional 10 LMC and 5 SMC objects have been imaged in the same line under the Guaranteed Time Observer program.

Abundances in early-type galaxies are measured from the analysis of stellar spectra (e.g. colour indices, Peletier et al, 1990). The presence of many planetary nebulae (PN) in early-type galaxies provides an independent measure of abundances for the old stellar population and allows the spread in abundances to be sampled at a range of galacto-centric distances. PN are feasible for this project since the nebular O, Ne and S abundances in most PN reflect that of the progenitor star.