Johannes Reetz - Publications

Refereed journals

Bedding, T.R., Kjeldsen, H., Reetz, J., Barbuy, B.: 1996,
Measuring stellar oscillations using equivalent widths of absorption lines, Mon. Not. R. Astron. Soc., 280, 1155
(including figures, 115 kB)

Fuhrmann, K., Pfeiffer, M., Frank, C., Reetz, J., Gehren, T.: 1997,
The surface gravities of cool dwarf stars revisited, Astron. Astrophys., 323, 909
(including figures, 261 kB)

Mao S., Reetz J., Lennon, D.J.: 1998,
Detecting Luminous Gravitational Microlenses Using Spectroscopy, Astrophys. Astron., 338, 56

Gehren T., Ottmann R., Reetz J.: 1999
Photospheric metal abundances of AR Lacertae, Astron. Astrophys., 344, 221
(including figures, 121 kB)

Kudritzki, R.-P., Puls, J., Lennon, D.J., Venn, K.A., Reetz, J., Najarro, F., McCarthy, J.K., Herrero, A.: 1999
The Wind Momentum - Luminosity Relationship of galactic A- and B-supergiants, Astron. Astrophys., 350, 970

Gehren T., Butler K., Mashonkina L., Reetz J., Shi J.: 2001
Kinetic equilibrium of iron in the atmospheres of cool dwarf stars. I. The solar iron spectrum, Astron. Astrophys., 366, 981

PhD thesis

Reetz, J. : November 1998 (defended February 1999),
"Oxygen Abundances in Cool Stars and the Chemical Evolution of the Galaxy" (381 pages , in german, 12.3MB pdf), PhD thesis, Ludwig-Maximilians-Universität München

The revision of the solar oxygen abundance (after 20 years where the "higher" value of 8.92 (Lambert 1978) was widely adopted by many researchers as a "standard", I determined a solar oxygen abundance of 8.80 which is 0.12dex lower) was just a byproduct of the scientific endeavour to determine the oxygen abundances of metal-poor stars, the tracers of the early history of our Galaxy. From the beginning of my research on that topic it became clear, that a differential abundance analysis of the different spectral lines in other stars (OI7773, [OI6300], OH3100, are all different concerning lineformation), and that an accurate comparison of these results with theoretical yields of SNII explosion calculations require also an accurate absolute oxygen (and iron) abundance of the Sun. I decided that it is not acceptable if I could not explain the reason why the indicator- and model-specific discrepancies of the solar oxygen abundances were discussed in the literatur in the early 90ies and earlier.
Therefore, my thesis comprises also a detailed investigation of the solar oxygen abundance indicators. The semi-empirical photospheric temperature structure of Holweger and Müller (1974) turned out to be most suitable - at that time (but I think it still is) - for reproducing the solar center-to-limb behaviour also of the OI7773 intensity profiles.
It was beyond the scope of my thesis to develop a comprehensive photospheric model. Of course, it is evident that the oxygen indicators, mainly the infrared ro-vibrational and rotational OH lines, but also the OI-triplet are very sensitive to the underlying temperature stratification. But it is also clear that these spectroscopic indicators have to lead to a consistent oxygen abundance, provided effects such as blends with other spectral lines (the [OI6300] is significantly blended - differing from the results of the thorough investigation of Lambert 1978) and deviations from LTE lineformation (Kiselman 1993) are carefully taken into account. In addition, the analysis results of the infrared OH bands are sensitively depending on the adopted iron abundance owing to the contribution of FeI to the electron pressure which determines the continuum in the infrared of the solar spectrum (the revision of the adopted iron abundance from 7.67 - based on Blackwell's FeI absorption oscillator strengths - to the meteoritic value of 7.51 was the reason for the oxygen abundance update of Grevesse et al. 1996).
I am not against discrepancies. Discrepancies offer the chance to gain new insights. However, following Popper, every scientific statement that wants to be taken for serious - if it is not a postulat or definition - should be falsifiable. Therefore any temperature structure - whether it is based on a 1D or a 3D hydrodynamical model - should only try be directly or indirectly in conflict with relevant observations such as the center-to-limb variation of the solar intensity spectrum or intensity profiles, or helioseismologic measurements, or CI chondritic iron abundances, if strong arguments can be made.
In the context of my thesis, the limited availability of high-quality observed intensity profiles (e.g. OI7773 triplet observed at various positions of the solar disk using different instrumental apertures) might constrain the accuracy of the outcome. But I suggest that uncertainties will not exceed 0.1dex.
So finally, as one of my results, the solar oxygen abundance has been significantly - but not dramatically - reduced compared to previous investigations in the time span between 1978 and 1998. Results were presented in condensed form at the Meudon and ESO/Garching conferences in Sept. and Oct. 1998, but details have been published via the thesis work that was made electronically available at the Ludwig-Maximilian-University Library since July 1999 (preprints of the work have been distributed to colleagues already at Nov./Dec. 1998). Some important results are:
(1)The Ni-Blend at [OI6300] is significant and must be taken into account (reduces the determined solar oxygen abundance by ~0.13dex) and
(2) a significant - but not overwhelming large - NLTE-effect of ~0.13dex must be taken into account for the analysis of the solar OI7774-triplet.
The revised solar oxygen abundance is log(eps) = 8.80 (+- 0.06).