Prominence detection and chromosphere feature on the prototype RS CVn of active binary systems

A definition of RS CVn-type binary systems was first performed by Hall (1976) and subsequently refined by Fekel, Moffett, & Henry (1986): a binary system which has at least one cool component showing strong magnetic activity in several forms. This activity manifests itself as strong photometric variability caused by large starspots, chromospheric emission, transition region emission, and coronal radiation. It is commonly accepted that all of these active phenomena arise from a powerful magnetic dynamo generated by the interplay between turbulent motions in the convection zone and the stellar differential rotation. At Yunnan Observatories, we are performing a long-term high-resolution spectroscopic monitoring project that covers several RS CVn-type systems to study their magnetic activity phenomena (e.g. Gu et al., 2002; Cao & Gu, 2015; Cao et al., 2019). In the present paper, we focus on the prototype system RS CVn itself.
RS CVn (= HD 114519 = BD+36$\degr$2344) is a double-lined, totally eclipsing binary system consisting of a K2 IV primary and a F5 V secondary (Reglero, Gimenez, & Estela, 1990; Rodonò, Lanza, & Becciani, 2001). The system has a period of about 4.798 days in an almost circular orbit with an inclination of $i$ = 85$\degr$ (Catalano & Rodonò, 1974; Eaton et al., 1993). Table 1 summarizes RS CVn’s properties, compiled from Eaton et al. (1993) and Rodonò, Lanza, & Becciani (2001).
RS CVn shows a significant distortion wave in its photometric light curve, caused by starspots on the surface of the K2 IV primary star (e.g. Eaton et al., 1993; Heckert & Ordway, 1995; Rodonò, Lanza, & Catalano, 1995; Rodonò, Lanza, & Becciani, 2001). Based on long-term light curves obtained between 1949 and 1993, Rodonò, Lanza, & Catalano (1995) analyzed its starspot evolution, possible activity cycles and orbital period variations. Rodonò, Lanza, & Becciani (2001) further determined accurate photometric parameters using a long-term sequence of light curves of RS CVn, taking into account the light curve distortions caused by starspots. More recently, using Doppler imaging, Xiang et al. (2020) reconstructed the starspot distribution on the primary star and estimated its surface differential rotation based on images derived from two consecutive rotational cycles.
Chromospheric activity of RS CVn has been studied by several authors, e.g. Popper (1988), Reglero, Gimenez, & Estela (1990), Fernández-Figueroa et al. (1986, 1994), and Montes et al. (1996a). RS CVn shows pronounced chromospheric emission in Ca ii H & K and other activity lines, the emission features are mainly associated with the K2 IV primary star (Reglero, Gimenez, & Estela, 1990; Montes et al., 1996a).
So far, we know only little about the magnetic activity of RS CVn’s outer atmosphere, in particular in relation to its chromosphere. Here we present the detection of prominence-like structures on RS CVn and the results of our chromospheric activity study based on the Ca ii IRT, H_α, Na i D₁, D₂ doublet, He i D₃, and H_β lines. Our study makes use of a large set of high-resolution spectra obtained during several observing runs between 1998 and 2017. This includes data from a joint observation campaign using telescopes located in China and Mexico in 2017 April, designed to get a dense short-term phase coverage. In Section 2, we provide details of our observations and data reduction. Our analysis of chromospheric activity indicators is described in Section 3. In Section 4, we describe in detail the different activity phenomena of RS CVn during our observations, including prominence-like events, optical flares, and chromospheric activity variations. Finally, we state our conclusions in Section 5.

We simultaneously analyze different chromospheric activity indicators, namely Ca ii IRT, H_α, Na i D₁, D₂ doublet, He i D₃, and H_β lines, formed in a wide range of atmospheric heights. To separate the contribution from the photosphere absorption profile in these lines, we apply a spectral subtraction technique, making use of the STARMOD program (Barden, 1985; Montes et al., 1997, 2000). STARMOD synthesizes a spectrum, later subtracted, by rotationally broadening, RV-shifting, and adaptively weighting spectra of two suitable template stars. These template stars are chosen as inactive, but having the same spectral-type and luminosity class as the components of the system. Thus, the synthesized spectrum approximates the non-active state of the binary, and the subtraction between the observed and synthesized spectra produces the activity contribution as excess emission and/or absorption, relative to the presumed inactive state.
For the observations of Jan. 2016 and Nov–Dec. 2017, taken at Xinglong, we use spectra of HR 8088 (K2 IV) and HR 3262 (F6 V) as templates for the primary and secondary star, respectively. The $vsini$ of each binary component is determined by STARMOD using these template spectra, resulting in an average $vsini$ of 45 km s^-1 for the primary star and 14 km s^-1 for the secondary one, based on a high signal-to-noise ratio (SNR) spectral window spanning many photospheric absorption lines, observed at phases where the two components of the system are well separated. The obtained $vsini$ values are in good agreement with the results of 42 $\pm$ 3 km s^-1 & 11 $\pm$ 2 km s^-1 estimated by Strassmeier & Fekel (1990), and 44.9 $\pm$ 1 km s^-1 & 12.4 $\pm$ 0.5 km s^-1 by Xiang et al. (2020). The adopted intensity ratios (primary/secondary) used by STARMOD are 0.57/0.43 for the Ca ii $\lambda$8662 spectral region, 0.56/0.44 for the Ca ii $\lambda$8542 spectral region, 0.55/0.45 for the Ca ii $\lambda$8498 spectral region, 0.5/0.5 for the H_α spectral region, 0.47/0.53 for the Na i D₁ and D₂ doublet, He i D₃ spectral region, finally 0.43/0.57 for the H_β spectral region.
The same template stars, HR 8088 and HR 3262, are also used for the observations of Apr. 2017, obtained at Weihai, as well as the observations of Mar. 1998, Feb. 2000, and Feb. 2004 obtained at Xinglong. We use the same $vsini$ values and intensity weight ratios as given above, with the exception of the H_β line region for the Weihai observations, because of very low SNR, and the Xinglong observations, where the line is not covered. Also, during the 1998 observing run, the Ca ii $\lambda$8498 line was not covered by the spectra due to an echelle setup change.
Because the template stars used above were not observed by the TIGRE telescope, we use spectra of two other inactive stars HR 5227 (K2 IV) and HR 8697 (F6 V) as templates instead. The $vsini$ values of 45 & 14 km s^-1 are used, and the resulting and adopted intensity weight ratios are 0.59/0.41 for the Ca ii IRT spectral region, 0.51/0.49 for the H_α spectral region, and 0.44/0.56 for the H_β spectral region. Also for some of the TIGRE observations, H_β could not be analyzed because of too low SNRs.
During eclipses of RS CVn’s components, the intensity weights of the two components change with orbital phase, and the line profiles are distorted – a situation that STARMOD can not model. Since some of our observations were obtained during eclipse, the resulting spectra have been excluded from our analysis, they are not listed in Table 1. This applies with the exception of spectra observed during total primary eclipse, i.e. when the secondary is completely occulted by the primary, such as observations on 2017 Dec 11. In these cases, we do use the template spectrum of the K2 IV alone to perform our analysis.
As a typical example, we show the chromospherically sensitive lines of one spectrum and illustrate the above described processing in Fig. 1. As expected, RS CVn shows pronounced chromospheric emission in all analyzed lines, mainly associated with the primary star. Moreover, for several observations of 2017 November–December and some observations of other observing runs, there are obvious excess emission features associated with the secondary star of RS CVn in the subtracted spectra (especially in the H_α line), which indicates that the secondary star is also active in the system, although much weaker.
The equivalent widths (EWs) of the subtracted Ca ii IRT, H_α, and H_β line profiles are measured with the SPLOT task in the IRAF package, as described in our previous papers (Cao & Gu, 2015, 2017), they are summarized in Table 5 together with their estimated uncertainties, where we also provide the ratios of the EW($\lambda$8542)/EW($\lambda$8498) and the E_Hα/E_Hβ for some of our observations. The E_Hα/E_Hβ ratios are calculated from the EW(H_α)/EW(H_β) values with the correction:

given by Hall & Ramsey (1992), which takes into account the absolute flux density in H_α and H_β lines, and the color index; we use $B-R$ = 0.81 for the calculation here.
The EW₈₅₄₂/EW₈₄₉₈ ratios thus obtained lie in the range of 1.0–2.5, consistent with the ratios found in solar plages ($\sim$1.5–3; Chester 1991) as well as several other chromospherically activity stars (e.g. Montes et al., 2000; Gu et al., 2002; Cao & Gu, 2015; Cao et al., 2020), thereby suggesting that the Ca ii IRT line emission arises predominantly from plage-like regions.

Based on the above analysis, our main results are:

1. 1.

   RS CVn shows excess emission features in the Ca ii IRT, H_α, and H_β lines, and the chromospheric emission stems mainly from the K2 IV primary of the system, which agrees to previous results reported by other authors. In addition, the F5 V secdonary shows weak chromospheric emission in some of our observations, implying an active, albeit much less active, chromosphere on this star, too.

2. 2.

   There are some unusual excess absorption features in the subtracted spectra taken near primary eclipse, which appear to be caused by two prominences located on the primary star. These tentative prominences absorb radiation from the secondary star at phases near primary eclipse. We have investigated the physical properties of the prominence structure based on the absorption features and the geometry of RS CVn. Two prominences are estimated to have heights of 1.12 $R_{pri}$ – 1.17 $R_{pri}$ and 0.45 $R_{pri}$ – 1.31 $R_{pri}$ above the surface of the primary star, respectively. Moreover, we characterize the stronger to get a mass of $2.7~{}\times~{}10^{15}$ kg for hydrogen.

3. 3.

   An optical flare was detected on 2016 January 26, most strongly indicated by the He i D₃ line emission feature. The energy released in the H_α line during flare maximum, $1.66\times 10^{31}$ erg s^-1, has a similar order of magnitude as strong flares of other very active RS CVn-type stars.

4. 4.

   RS CVn shows rotational modulation of chromospheric activity, which means the presence of the chromospheric activity longitudes over the surface of the primary star. The active longitudes most frequently appear around the two quadratures of the binary system.

We would like to thank the staff of the telescopes for their help and support during our all observations. We are grateful to the referee Prof. Jeffery Linsky for his valuable suggestions which result in a large improvement to our manuscript. This study make use of the data obtained with the TIGRE telescope, which located at La Luz observatory, Mexico. TIGRE is a collaboration of the Hamburger Sternwarte, the Universities of Hamburg, Guanajuato and Liège. This work is partially supported by the Open Project Program of the Key Laboratory of Optical Astronomy, National Astronomical Observatories, Chinese Academy of Sciences. The joint research project between Yunnan Observatories and Hamburg Observatory is funded by Sino-German Center for Research Promotion (GZ1419). The present study is also financially supported by the National Natural Science Foundation of China (NSFC) under grants Nos. 10373023, 10773027, 11333006, U1531121, and 11903074, and supported by the Natural Science Foundation of Yunnan Province of China (Grants Nos. 202201AT070186 and 202305AS350009). We acknowledge the science research grant from the China Manned Space Project with NO. CMS-CSST-2021-B07.

The data underlying this article are available from the authors upon request.