Stellar spectral models compared with empirical data

Abstract

The empirical MILES stellar library is used to test the accuracy of three different, state-of-theart, theoretical model libraries of stellar spectra. These models are widely used in the literature for stellar population analysis. A differential approach is used so that responses to elemental abundance changes are tested rather than absolute levels of the theoretical spectra. First we directly compare model line strengths and spectra to empirical data to investigate trends. Then we test how well line strengths match when element response functions are used to account for changes in [α/Fe] abundances. The aim is to find out where models best represent real star spectra, in a differential way, and hence identify good choices of models to use in stellar population analysis involving abundance patterns. We find that most spectral line strengths are well represented by these models, particularly iron-and sodium-sensitive indices. Exceptions include the higher order Balmer lines (Hδ, Hγ), in which the models show more variation than the data, particularly at low temperatures. C 2 4668 is systematically underestimated by the models compared to observations. We find that differences between these models are generally less significant than the ways in which models vary from the data. Corrections to C 2 line lists for one set of models are identified, improving them for future use.

To use or not to use synthetic stellar spectra in population synthesis models?

“…7, 8 and 11 show a complex and index-dependent behaviour. One may argue that the best performance of the sps-s models occurs for the indices H10, B4000, Hδ F , CN 2 , G4300, Hγ F , Fe4383, Hβ, Mg b and Fe5782, whereas the indices that deserve most attention from modellers are Fe5270, Fe5015, CaHK, Fe5709, Fe4383 and Fe4668 (the latter had already been pointed out by Knowles et al 2019).…”

“…(iii) Synthetic stellar spectra were computed with the SYNTHE code (Kurucz & Avrett 1981;Sbordone et al 2004), based on the ATLAS9 and MARCS models. The opacities are as in C14 with one update, the inclusion of the molecular transition C 2 D-A from Brooke et al (2013) 6 to correct for the problem identified by Knowles et al (2019). For stars with T eff ≥ 3500 K the synthetic spectrum was computed based on the ATLAS9 models.…”

“…The dependence with the age of the population is apparent: as expected, the largest deviations from the 1-to-1 line occur for the old population in the case of the synthetic effect, and for the young population in the case of the coverage effect. In these plots we can locate opacity related problems, e.g., Fe4668 (see discussion in Knowles et al 2019), or the high sensitivity to HRD coverage of the Balmer line indices. The effect of randomising the stellar parameters is to increase the dispersion of the data points but they do not deviate significantly from the 1-to-1 line except for young populations: most likely this is induced by the coverage effect due to the scarcity of hot stars in the HRD.…”

sMILES: a library of semi-empirical MILES stellar spectra with variable [α/Fe] abundances

“…This approach, making use of both models and observations, builds upon the work of La Barbera et al (2017), implementing both the accuracy of empirical spectra with the differential abundance pattern predictions of theoretical spectra. Using only differential predictions from theoretical spectra has been shown to reproduce observations of abundance pattern effects more accurately than fully theoretical spectra, particularly for wavelengths below Mg b (e.g see figure 11 of Knowles et al 2019or Martins & Coelho 2007Bertone et al 2008;Coelho 2014;Villaume et al 2017;Allende Prieto et al 2018).…”

Spectra-GANs: A New Automated Denoising Method for Low-S/N Stellar Spectra

“…In this subsection, we compare the performance of Spectra-GANs, PCA, DnCNN, RBM, and Wavelet on recovering the Lick index. The Lick/IDS index system is a set of absorption-line indices, defined by the Lick group for low-resolution (R∼8) spectra with wavelength coverage between the range of 4000Å and 6400Å [34]- [39], which is often used as an approach to measure the atmospheric parameters in stars and galaxies. Barbuy et al explored In this experiment, we first denoised the spectra with Spectra-GANs, PCA, DnCNN, RBM, and Wavelet, and then computed the Lick indices of the denoised spectra.…”