A New Look into K-giants' Chromospheres

Cool evolved stars contribute significantly to the replenishment of material such as dust and molecules in the universe. The understanding of mass loss from the most evolved stars on the Asymptotic Giant Branch (AGB) has recently improved, and is understood to involve a combination of pulsation-induced shock waves lifting the material high-enough to enable dust condensation, and radiation pressure on dust grains generating enough momentum to power a dust-driven stellar wind. Observational evidence of the mass-loss history in circumstellar envelope of AGB stars can be found in their detached shells (e.g., Brunner et al. 2019).

However, for less evolved, less pulsating K-type Red Giant Branch stars (K-giants), the details of the mass-loss scenario are still unknown. In their outer atmospheres, the mass loss appears to be powered by Magneto-Hydrodynamic (MHD) Alfvén waves (e.g., Airapetian et al. 2010; Carpenter et al. 2018; Rau et al. 2018, 2019a, 2019b). However, the mechanisms driving the chromospheric heating and wind acceleration in K-giants is not yet completely understood due to a lack of necessary observations and comprehensive modeling. With the increasing high-angular resolution capability of new instruments mounted at the most powerful ground-based telescopes, such as the incoming second-generation instruments at the Center for High Angular Resolution Astronomy (CHARA), this work to reveal the chromospheric structure of K-giant stars is timely.

We used the PHOENIX atmosphere code to perform preliminary modeling of the K-giant star β Cet observed with the Visible spEctroGraph and polArimeter (VEGA, Mourard et al. 2008) instrument at CHARA. PHOENIX has the capacity to do multi-line non-local thermodynamic equilibrium (non-LTE) calculations for many species, which becomes important in the modeling of hot, low density outer layers of stars, such as the chromosphere. It is also equipped with a stellar wind module (Aufdenberg & Barman 2002) necessary for the modeling of giant stars.

Figure 1 shows a schematic structure of the chromosphere and wind of a K-type giant star (left), and illustrates an example of our preliminary PHOENIX model of β Cet (right), which closely reproduces the observed infrared Ca ii triplet line profiles. This triplet is known to be chromospheric in origin and a strong chromospheric activity tracer in K-giants (e.g., Meszaros et al. 2008; Berio et al. 2011; Vieytes et al. 2011). In Figure 1, we compare the synthetic PHOENIX spectrum to the observations, rotationally broadening the lines according to the literature value (Dupree et al. 2005; Berio et al. 2011). The model takes T_eff, M_⋆, log(g), and metallicity (Z) as input parameters, along with a prescribed structure in the chromosphere and transition region. This preliminary modeling includes a modest set of non-LTE species: H, He, C, N, O, Mg, Ca, Fe, which will be extended to a more complete list for the final published set of models. Even though this preliminary model includes only the chromospheric contribution and not yet the underlying winds, it closely reproduces the observed infrared Ca ii triplet line profiles. Non-LTE computations for additional stars will provide contribution functions that indicate the relative locations where chromospheric line cores are formed, constraining the radial extent of the chromospheres.

Zoom In Zoom Out Reset image size

This preliminary work shows the capability of PHOENIX to model K-giant stars. Efforts are ongoing to extend this work to a larger set of K-giants. Our preparatory model described above includes only the chromospheric component, and represents the first step of this analysis: identifying the accurate chromospheric structures for each star and finalizing the complete non-LTE species set.

In order to investigate the importance of non-LTE metal line blanketing in the chromospheres of these cool evolved stars, we will compute a grid of non-LTE atmospheric models and synthetic spectra for a set of K-giants, including e.g., η Cet, δ Crt, ρ Boo, β Oph, 109 Her, ι Cep, β Gem, μ Leo. We will use literature values for T_eff, M_⋆, log(g), Z, and rotational velocity as model inputs, and compute several upper atmospheric profiles to create a grid of spectra with different levels of chromospheric activity, in order to identify the correct structure that reproduces the observations of each star in our target list.

Once the accurate structure is identified, we will move to the next step of adding the wind component to the model. We will add the overlying wind absorption to the fine-tuned chromospheric model spectrum, and identify the detailed chromospheric structure. In this way we will determine the complete thermal structure in the upper atmosphere.

We plan to produce a set of K-giant synthetic spectra covering wavelengths from the ultraviolet (UV) to the infrared (IR), simultaneously reproducing archival UV Hubble Space Telescope/International Ultraviolet Explorer data of the stars and new visible/IR CHARA observations, and providing a multi-wavelength view of a set of K-giants chromospheres. Since the ultraviolet emission from K giants is largely chromospheric in origin, UV data will help us measure the chromospheric activity. The second generation instrument Stellar Parameters and Images with a Cophased Array (SPICA) (Mourard et al. 2017, 2018; Pannetier et al. 2020) at CHARA is the successor of VEGA, now decommissioned. With its exceptional high-angular resolution SPICA will help us to measure chromospheric lines in the V-band (0.65–0.9 μm), for example: Hα at 660 nm, Ca ii triplet lines at 850 nm, and Fe i triplet lines at 540 nm. The combination of archival UV spectra together with high-angular resolution interferometric CHARA/SPICA data will be the perfect multi-wavelength tool to observe the chromospheric structure in a set of K-giant stars.

The material is based upon work supported by NASA under award number 80GSFC17M0002.