Robert L. Kurucz

Model atmospheres for G, F, A, B, and O stars

A grid of LTE model atmospheres is presented for effective temperatures ranging from 5500 to 50,000 K, for gravities from the main sequence down to the radiation pressure limit, for abundances solar, 1/10 solar, and 1/100 solar. The models were computed by use of a statistical distribution-function representation of the opacity of almost 10/sup 6/ atomic lines. For each model we tabulate the temperature structure, fluxes, UBV and uvby colors, bolometric correction, and Balmer line profiles. The solar abundance models are compared to narrow, intermediate (by Relyea and Kuruez), and wide (by Relyea and Kuruez and by Buser and Kuruez) band photometry and are found to be in good agreement with the observations for effective temperatures aboue 8000 K. Excellent agreement exists with the spectrophotometry and Balmer line profiles of Vega. A small systematic error in the colors of late A and F stars is probably due to an overestimate of convection in weakly convective models. This error does not seem to affect greatly the use of the predicted colors for differential studies. The solar model has approximately a 2% error in the V flux because molecular lines were not included.

Model Atmospheres for Population Synthesis

I have used my newly calculated iron group line list together with my earlier atomic and molecular line data, 58,000,000 lines total, to compute new opacities for the temperature range 2000K to 200000K. Calculations have been completed at the San Diego Supercomputer Center for 56 temperatures, for 21 pressures, for microturbulent velocities 0, 1, 2, 4, and 8 km/s, for 3,500,000 wavelength points divided into 1221 intervals from 10 to 10000 nm, for scaled solar abundances [+1.0], [+0.5], [+0.3], [+0.2], [+0.1], [+0.0], [−0.1], [−0.2], [−0.3], [−0.5], [−1.0], [−1.5], [−2.0], [−2.5], [−3.0], [−3.5], [−4.0], [−4.5], and [−5.0]. I have rewritten my model atmosphere program to use the new line opacities, additional continuous opacities, and an approximate treatment of convective overshooting. The opacity calculation was checked by computing a new theoretical solar model that matches the observed irradiance. Thus far I have completed a grid of 7000 model atmospheres at 2 km/s for all the abundances, for the temperature range 3500K to 50000K, and for log g from 0.0 to 5.0. This grid will allow a consistent theoretical treatment of photometry from K stars to B stars. Fluxes are tabulated from .09 to 160 micrometers. Preliminary results are reported for many photometric systems. Work is underway on grids for other microturbulent velocities. Microturbulent velocity strongly affects the interpretation of Cepheid and RR Lyrae photometry. The models, fluxes, and colors are available on magnetic tape and will also be distributed on CD-ROMs.

A new sunspot umbral model and its variation with the solar cycle

Semiempirical model atmospheres are presented for the darkest parts of large sunspot umbrae, regions have called umbral cores. The approach is based on general-purpose computational procedures that are applicable to different types of stellar atmospheres. It is shown that recent umbral intensity measurements of the spectral energy distribution may be accounted for by an umbral core atmospheric model that varies with time during the solar cycle; the observed center-limb variation can be accounted for by the properties of the model. Three umbral core models are presented, corresponding to the early, middle, and late phases of the solar cycle. These three models also may be regarded as having the properties of dark, average, and bright umbral cores respectively. The effects of atomic, opacity, and abundance data uncertainties on the model calculations are briefly discussed. For comparison, a new reference model for the average quiet solar photosphere is given. 94 references.

Calculation of Solar Irradiances. I. Synthesis of the Solar Spectrum

Variations in the total radiative output of the Sun as well as the detailed spectral irradiance are of interest to terrestrial and solar-stellar atmosphere studies. Recent observations provide measurements of spectral irradiance variations at wavelengths in the range 1100-8650 ? with improved accuracy, and correlative studies give procedures for estimating the spectral irradiance changes from solar activity records using indicators such as those derived from Ca II K and Mg II indices. Here we describe our approach to physical modeling of irradiance variations using seven semiempirical models to represent sunspots, plage, network, and quiet atmosphere. This paper gives methods and details, and some preliminary results of our synthesis of the variations of the entire irradiance spectrum. Our calculation uses object-oriented programming techniques that are very efficient and flexible. We compute at high spectral resolution the intensity as a function of wavelength and position on the disk for each of the structure types corresponding to our models. These calculations include three different approximations for the line source function: one suited for the very strong resonance lines where partial redistribution (PRD) is important, another for the most important nonresonance lines, and another approximation for the many narrow lines that are provided in Kuruczs listings. The image analysis and calculations of the irradiance variation as a function of time will be described in a later paper. This work provides an understanding of the sources of variability arising from solar-activity surface structures. We compute the Ly? irradiance to within 3% of the observed values. The difference between our computations and the Neckel & Labs data is 3% or less in the near-IR wavelengths at 8650 ?, and less than 1% in the red at 6080 ?. Near 4100 ? we overestimate the irradiance by 9%-19% because of opacity sources missing in our calculations. We also compute a solar cycle variability of 49% in the Ly? irradiance, which is very close to observed values. At wavelengths between 4100 ? and 1.6 ?m, we obtain spectral irradiance variations ranging from -0.06% to 0.46% in the visible?the higher values correspond to the presence of strong lines. The variability in the IR between 1.3 and 2.2 ?m is ~-0.15%.

The binary progenitor of Tycho Brahe's 1572 supernova

The brightness of type Ia supernovae, and their homogeneity as a class, makes them powerful tools in cosmology, yet little is known about the progenitor systems of these explosions. They are thought to arise when a white dwarf accretes matter from a companion star, is compressed and undergoes a thermonuclear explosion. Unless the companion star is another white dwarf (in which case it should be destroyed by the mass-transfer process itself), it should survive and show distinguishing properties. Tychos supernova is one of only two type Ia supernovae observed in our Galaxy, and so provides an opportunity to address observationally the identification of the surviving companion. Here we report a survey of the central region of its remnant, around the position of the explosion, which excludes red giants as the mass donor of the exploding white dwarf. We found a type G0–G2 star, similar to our Sun in surface temperature and luminosity (but lower surface gravity), moving at more than three times the mean velocity of the stars at that distance, which appears to be the surviving companion of the supernova.

New Opacity Calculations

I have computed new opacities for model stellar atmospheres and envelopes using a large grant of Cray computer time at the San Diego Supercomputer Center. The opacities include 58,000,000 atomic and diatomic molecular lines. Twelve-step distribution functions are tabulated for 56 temperatures in the range from 2000K to 200000K, for 21 log pressures from −2 to 8, for 1212 wavelength intervals from 10 to 10000 nm, for microturbulent velocities 0, 1, 2, 4, and 8 km/s, for scaled solar abundances [+1.0], [+0.5], [+0.3], [+0.2], [+0.1], [+0.0], [−0.1], [−0.2], [−0.3], [−0.5], [−1.0], [−1.5], [−2.0], [−2.5], [−3.0], [−3.5], [−4.0], [−4.5], [−5.0], and [+0.0, no He] (log abundance of elements heavier than helium relative to solar). Rosseland means are also tabulated for each case. The final files for each abundance require three 6250 bpi VAX backup tapes. I am now distributing tape copies. I hope to have CD-ROMs available in the near future. A solar photospheric model computed with the new opacities matches the observed energy distribution.

A major upgrade of the VALD database

Vienna atomic line database (VALD) is a collection of critically evaluated laboratory parameters for individual atomic transitions, complemented by theoretical calculations. VALD is actively used by astronomers for stellar spectroscopic studies—model atmosphere calculations, atmospheric parameter determinations, abundance analysis etc. The two first VALD releases contained parameters for atomic transitions only. In a major upgrade of VALD—VALD3, publically available from spring 2014, atomic data was complemented with parameters of molecular lines. The diatomic molecules C2, CH, CN, CO, OH, MgH, SiH, TiO are now included. For each transition VALD provides species name, wavelength, energy, quantum number J and Lande-factor of the lower and upper levels, radiative, Stark and van der Waals damping factors and a full description of electronic configurarion and term information of both levels. Compared to the previous versions we have revised and verify all of the existing data and added new measurements and calculations for transitions in the range between 20 A and 200 microns. All transitions were complemented with term designations in a consistent way and electron configurations when available. All data were checked for consistency: listed wavelength versus Ritz, selection rules etc. A new bibliographic system keeps track of literature references for each parameter in a given transition throughout the merging process so that every selected data entry can be traced to the original source. The query language and the extraction tools can now handle various units, vacuum and air wavelengths. In the upgrade process we had an intensive interaction with data producers, which was very helpful for improving the quality of the VALD content.

The nonsolar abundance ratios of Arcturus deduced from spectrum synthesis

Using opacity distribution functions based on a newly expanded atomic and molecular line list, we have calculated a model atmosphere for Arcturus that reproduces the observed flux distribution. Individual line parameters in the list were adjusted to match the solar spectrum in a preliminary way, in the regions 5000-5500 A, 6000-6500 A, and 7500-9000 A. The Arcturus model spectrum calculated with these adjustments reproduces well the profiles of all lines in the observed spectrum of the Griffin atlas for which the solar gf-values are well determined. The Arcturus model has an iron abundance [Fe/H]=−0.5±0.1, a temperature T eff =4300±30 K, gravity log g=1.5±0.15, and an overabundance of the light metals

Synthetic Infrared Spectra

The Sun is the star we can observe with the highest spectral resolution and signal-to-noise. From studying the infrared spectrum we can learn about the Sun, about stars in general, and about atomic and molecular spectroscopy. We discuss the computer programs for spectrum synthesis, the infrared flux and central intensity atlases of the solar spectrum, and the atomic and molecular line data. Considerable work is still required to improve the observations and to improve the line data.

The Progenitor of supernova 1993J revisited

From Hubble Space Telescope images with 0 05 resolution, we identify four stars brighter than V = 25 mag within 2 5 of SN 1993J in M81, which contaminated previous ground‐based brightness estimates for the supernova progenitor. Correcting for the contamination, we find that the energy distribution of the progenitor is consistent with that of an early K‐type supergiant star with MV ≈ -7.0 ± 0.4 mag and an initial mass of 13–22 M⊙. The brightnesses of the nearby stars are sufficient to account for the excess blue light seen from the ground in preexplosion observations. Therefore, the SN 1993J progenitor did not necessarily have a blue companion, although by 2001, fainter blue stars are seen in close proximity to the supernova. These observations do not strongly limit the mass of a hypothetical companion. A blue dwarf star with a mass up to 30 M⊙ could have been orbiting the progenitor without being detected in the ground‐based images. Explosion models and observations show that SN 1993J progenitor had a helium‐rich envelope. To test whether the helium abundance could influence the energy distribution of the progenitor, we calculated model supergiant atmospheres with a range of plausible helium abundances. The models show that the presupernova colors are not strongly affected by the helium abundance longward of 4000 A, and abundances ranging between solar and 90% helium (by number) are all consistent with the observations.