Using Shell Models to Investigate Clumping in the Wind of the O7Iaf+ Supergiant AzV83

TL;DR

This study introduces a shell-based clumping model for stellar winds in an O7Iaf+ supergiant, providing more realistic spectral simulations and insights into wind properties compared to traditional methods.

Contribution

The paper presents a novel shell-based approach to model wind clumping in massive stars, improving spectral fitting and physical understanding over the volume-filling factor method.

Findings

Shell model yields similar wind parameters to traditional methods.

Higher mass-loss rate needed in shell model to fit Hα.

Shell model better reproduces high ionisation resonance lines.

Abstract

Hot massive stars exhibit strong stellar winds that enrich the surrounding interstellar medium and affect the stars' evolution. However, the winds are inhomogeneous (clumped) and are difficult to model in radiative transfer codes. To produce more realistic spectra many codes use a volume-filling factor approach to incorporate the effects of clumping. While this approach is convenient it is simplistic. We introduce an alternative approach to incorporate clumping by assuming the wind is composed of dense spherical shells. Using this approach in the radiative transfer code CMFGEN we produce synthetic spectra for AzV83, an O7Iaf+ supergiant located in the Small Magellanic Cloud. The spectrum of AzV83 is rich in both photospheric and wind features, making it an ideal candidate with which to investigate the physical characteristics of stellar winds. Synthetic spectra are compared to the star's observed spectrum to better characterize the influence of clumped winds on spectral features, and to better understand the limitations of the volume-filling factor approach. The approach using spherical shells yields similar wind parameters to those obtained using the volume-filling factor approach although a slightly higher mass-loss rate is required to fit H$\alpha$. As expected, the interclump medium in the model with shells allows the high ionisation resonance transitions of N V and O VI to be fitted using $L_{\rm X-ray}/L_{\rm Bol} \approx 10^{-7}$ which is typically observed for O stars, and which is a factor of ten lower than needed with the volume-filling factor approach.