The temperature of Central Stars of Planetary Nebulae (CSPN) is one of the most important parameters for a better understanding of their evolution, their properties and those of the surrounding photoionized nebula. It is an elusive quantity to derive though, because CSPN are generally faint objects and their spectrum is often heavily contaminated by nebular continuum emission.

A calibration of the nebular intensity of [O III] 5007 (relative to Hβ) in terms of stellar bolometric corrections is presented. This calibration (restricted to low- and medium-excitation PN) is based exclusively on the results of non-LTE model atmosphere analyses of absorption-line profiles in the spectra of bright central stars of PN (Méndez et al. 1988, A&A 190, 113 and subsequent papers). Knowing the bolometric corrections, the distance to the Galactic center, the amount of interstellar extinction and the visual apparent magnitudes (Tylenda et al. 1989, A&AS 77, 39) of central stars of PN in the direction of the Galactic center, luminosities for 20 of these central stars are obtained without any assumptions about nebular properties. The luminosities obtained with this method are substantially higher than those obtained for the same objects by Pottasch and Acker (1989, A&A 221, 123) and Tylenda et al. (1991, A&A 246, 221). The most probable explanation for this discrepancy is that luminosity estimates based on assumptions about nebular properties are too low, because many PN are not completely optically thick in the H Lyman continuum and/or because many ionizing photons in these PN are being absorbed by nebular dust. Which of the two effects is more important cannot be decided without solving the problem of the discrepancy between extinction determinations based on the Balmer decrement versus extinction determinations based on radio and Hβ fluxes (the radio-Hβ extinctions are too low).

A first step in the accurate quantitative spectroscopic analysis of central stars of PN has been based on fitting the results of NLTE, hydrostatic, plane-parallel model atmosphere calculations to the observed H and He absorption-line profiles in high-resolution spectra of bright central stars (Méndez et al. 1988, A&A 190, 113 and subsequent papers). Such analyses have provided very useful determinations of the basic atmospheric parameters: T eff, log g and He abundance.

Absorption lines of highly ionized iron group elements dominate the UV spectra of many hot stars. They were identified in several central stars as well as in sdO stars, in PG 1159 stars and in the hottest DO and DA white dwarfs. Due to the high effective temperatures atmospheric modelling including metal line blanketing has to be done under NLTE conditions. Adequate models have become available only very recently (Dreizler & Werner 1992). We present new NLTE model atmospheres blanketed by some 120 000 lines from iron group elements (Sc through Ni). We adopted Anderson's (1991) statistical approach along with an opacity sampling technique using our Accelerated Lambda Iteration (ALI) code. We generally found:

• — Many strong metal lines cause substantial blocking at the flux maximum.

• — The temperature in the continuum forming layers is increased by backwarming due to iron group lines; outer layers, cooled by CNO lines, are unaffected.

• — The line profiles of H and He become deeper and broader, indicating that the neglect of NLTE metal line blanketing could cause the discrepancies encountered when fitting He II and H I lines in hot subdwarfs.

UV spectra of CSPN are dominated by lines from ions of iron group elements. Recently we have developed a non-LTE model of the Fe V spectrum putting particular emphasis on this spectral region (Becker and Butler, 1992a). Lines of Ni IV and Ni V are major contributors to blends with the dominating Fe V lines. Meanwhile, non-LTE models for Ni IV, Ni V and Ni VI have been developed (Becker and Butler, 1992b) and the non-LTE model spectra composed of lines from all four ions can be compared to the observations.

We present IUE observations of some nuclei of Planetary Nebulae. From these data we derive the stellar photospheric parameters (Teff Lbol , log g), and the wind characteristics (velocity, mass loss rate). Teff , R∗, Lbol are derived from UV low resolution spectra, combining optical and radio data, from Bianchi (1988) or from new IUE data, with the same method (fit of the UV continuum with model atmospheres for high gravity stars, after correcting for reddening and for the contribution of continuum emission by the nebular gas). P Cygni profiles from IUE high resolution spectra are fitted with the SEI method and V∞ is derived. The non-LTE ionisation in the wind and the mass loss rate are computed as in Bianchi et al. (1986). Details are given in a forthcoming paper. The results for a first group of objects are given in the Table below.

We continue the trend of finding discrepancies in the values of central star masses derived from photospheric line profile fitting and other methods in our analyses of ground-based and UV spectra of the nucleus of NGC 6826.

Models have been developed in Kiel for massive (Pop. I) WR stars which account for multi-level non-LTE radiation transfer in spherically expanding atmospheres. The published (Koesterke et al. 1992) grid of models for WC composition (40% helium, 60% carbon by mass) can be applied to low-mass stars as well by means of the scaling properties of WR spectra (Hamann et al. 1992) and allow a rough guess of the parameters, while individual calculations are necessary for a detailed analysis and the determination of the chemical composition.

During a program for the observation of central stars of old planetary nebulae 29 stars are classified until now. Most of them (22) are belonging to the hydrogen-rich sequence and resemble either high-gravity sdO stars or white dwarfs (14 DAOs, 3 DAs). 3 are hydrogen-deficient PG 1159 stars, and also 3 are hybrid type stars Further 3 CPN are close binaries, whose evolution has been certainly influenced by this circumstance. A complete list of so far observed objects is given in Napiwotzki (1992).

Observations of near infrared emission-lines are becoming available and can be a powerful tool to improve our knowledge on planetary nebulae properties. For wavelengths in the range 1 to 5 μm, the emission-lines correspond to atomic transitions of high ionized species of heavy elements. In particular, the [Si VI] 1.96μm and [Si VII] 2.48μm lines have already been detected (Ashley and Hyland, 1988).

Optical and UV spectrophotometric data are analyzed for the central stars of the known Population II planetary nebulae. From this, we derive visual magnitudes, spectral classification, color temperatures, luminosities and masses of the objects. It is found that all the stars show absorption type spectrum and most of them have normal H and He photospheric abundances, the only possible exceptions are M 2–29 and GJJC-1 which seem to be H-deficient stars.

A high-resolution IUE spectral atlas of central stars of planetary nebulae and hot white dwarfs has been produced (part of Tweedy, 1991, PhD thesis from the University of Leicester, UK), and examples from it are shown here. It has been sorted into an approximate evolutionary sequence, based on published spectroscopic analyses, from the cool 28,000K young central star He 2–138, through the hot objects like NGC 7293 and NGC 246 at 90,000K and 130,000K respectively, down to 40,000K DA white dwarfs like GD 2, which is the chosen cutoff for this selection. Copies of a revised version of this atlas, which will include more recent spectroscopic information and also white dwarfs down to 35,000K – to include the Si III object GD 394 – will be sent to anyone who requests one.

We have analysed absolutely calibrated 3500–8900Å AAT spectrophotometry of the WC10 central star CPD–56°8032. After dereddening by E(B–V) = 0.68, the few line-free portions of the stellar continuum are well fitted by the 18000K blackbody which corresponds to the nebular H I Zanstra temperature. The Wolf-Rayet wind terminal velocity is sufficiently low (210–230 km s−1) that it is possible to resolve virtually every fine structure component of every multiplet of C II, C III and O I listed by the NBS Multiplet Tables for this wavelength region.

The high resolution IUE images SWP 16742 and SWP 42675 have been studied to derive the fast wind properties of the central stars of the two low temperature planetary nebulae IC 2149 and Tc 1. The first image is from the IUE archive, while the second one has been taken by us.

In order to complete available C IV broadening data needed for stellar plasma research, we have calculated Stark broadening parameters for 69 C IV multiplets of large principal quantum number. The results along with a discussion of the Stark broadening parameter regularities within spectral series will be published elsewhere (Dimitrijevic and Sahal–Brechot, 1992). As an example in Figs 1 and 2 the case of C IV np 2 P 0 − 9s 2 S transitions, is presented. We can see that particularly for shifts the changes of Stark broadening parameters are relatively small, permitting the interpolation of new data or critical evaluation of mutual consistency of existing data.

Forbidden electric quadrupole (E2) and magnetic dipole (M1) transitions are of extreme importance in astrophysics. Up to now the most extensive calculations for the nitrogen isoelectronic sequence have been done using the method proposed by C.J. Zeippen [1] or in MCHF approximation [2]. To account for electron correlations both these methods use a large list of configurations. We have chosen the stationary many-body perturbation theory (MBPT) [3] for the inclusion of the electron correlations. The calculations have been perfomed in the second order in the complete model space 1s22s2p3+ 1s22p5. Relativistic corrections have been accounted for in the Breit-Pauli approximation. In the Table we present probabilities for electric quadrupole W(E2) and magnetic dipole W(M1) transitions (in s−1), wavelengths λ (in A). The comparision of the results shows that our second order calculation data in the most cases are closer to term-energy corrected ones from [1].

To identify spectra of Planetary Nebulae which usually have many atomic lines one needs very accurate theoretical atomic data.

The past decade has seen significant progress in our understanding of spectral energy distribution of planetary nebulae over the entire wavelength range from UV to radio. In this review we show the detailed breakdown of the energy budget for a planetary nebula as a system of the three components, i.e., the central star, the gaseous nebula and the dust shell. This picture of the energy distribution is further discussed in the context of planetary nebula evolution.

One of the most fundamental physical parameter in astronomy is the distance to the objects we detect in the universe. For many classes of astronomical objects, accurate and proven methods have been developed to determine their distances. Such classes of objects include stars within ∼100 pc from the sun, binary stellar systems, variable stars, stellar clusters, main sequence stars, and other galaxies. It has been, however, more difficult to develop satisfactory methods to determine accurate distances to the more than 1000 planetary nebulae that have been discovered in our galaxy.