Empirical prediction of plasma emission measure distributions and X-EUV spectra of late-type stars

TL;DR

The study addresses the difficulty of obtaining EUV/X-ray spectra for late-type stars by proposing an empirical, temperature-resolved emission measure distribution $EM_{\rm h}(T)$ parameterized by the stellar surface X-ray flux $F_{\rm x}$. By compiling high-resolution FUV and X-ray data into two stellar groups and converting EM$v$ to hemisphere-normalized EM$_{\rm h}$, the authors build FGK- and M-type-specific grids of $EM_{\rm h}(T)$ across $T$ from $10^4$ to $10^{7.5}$ K. They fit $EM_{\rm h}(T)$ as a function of $F_{\rm x}$, generate synthetic EMD grids for a broad reference flux range, and validate against observed X-ray and EUV fluxes, exploring the impact of chemical abundances. The approach enables efficient XUV spectrum synthesis for large samples of faint stars, aiding irradiation studies of exoplanetary atmospheres and disks, while underscoring the need for expanded data on low-activity stars and M dwarfs.

Abstract

High-energy emission spectra from the outer atmospheres of late-type stars represent an important feature of the stellar activity in several contexts, such as the photoevaporation and photochemistry of planetary atmospheres or the modeling of irradiated circumstellar disks in young objects. An accurate determination of these spectra in the EUV and soft X-ray (XUV) band requires high-resolution spectroscopy, that is rarely feasible with current instrumentation. We employed a relatively large set of plasma emission measure distributions (EMDs) as a function of temperature, derived from FUV and X-ray emission line spectra acquired with the Hubble Space Telescope and Chandra or XMM-Newton, in order to devise a relatively simple recipe for predicting EMDs and XUV spectra of stars of different spectral types, activity levels, and plasma metallicity. We show that the EMDs in the range of temperatures between 10^4 K and 10^7.5 K can be described using the stellar surface X-ray flux as a control parameter, but this parameterization also depends on the spectral type. In particular, we find that M-type stars show slightly lower emission measure at temperatures below ~10^5 K and higher emission measures for T > ~10^6.5 K with respect to FGK stars with similar surface X-ray fluxes. We evaluated the uncertainties in the broad-band EUV and X-ray fluxes derived from synthetic EMDs and spectra, considering in the error budget also the limited knowledge of the chemical abundances in stellar outer atmospheres.

Figures (9)

• Figure 1: Emission measure distributions of the sample stars. Truncated curves refer to Group 1 (Table \ref{['emduv']}), while complete curves refer to Group 2 (Table \ref{['emdxuv']}). Dashed lines indicate $\alpha$ Cen A and B (see text).
• Figure 2: Differential Emission Measure at selected values of the plasma temperature vs. surface X-ray flux, for all stars in the sample. Blue symbols for F--G-type stars, green symbols for K-type stars, and red symbols for M-type stars. Empty and filled symbols identify stars in Group 1 or Group 2, respectively. Best-fit polynomials for FGK stars and M-type stars are shown with solid lines, the standard deviations are indicated on the bottom right of each panel, and shadowed bands represent $1\sigma$ confidence ranges. For temperatures $\log T \geq 5.5$, circumscribed symbols indicate $\alpha Cen$ A and B from Wood+2018, and dashed lines show their best-fit polynomial results.
• Figure 3: Synthetic emission measure distributions at selected reference values of the stellar surface X-ray flux (values on a log-scale indicated on the right side of each panel). The $EM_{\rm h}$ grid for FGK-type stars is on the left panel, and on the right panel for M-type stars. A 6th degree polynomial smoothing was applied, and 1$\sigma$ error bars are shown.
• Figure 4: Reconstructed and synthetic emission measure distributions for selected stars in the original sample. G--K-type stars (on the left) are compared with M-type stars (on the right) having similar surface X-ray flux. The color coding of the original EMDs (solid lines) is the same as in Fig. \ref{['fig:demd']}. Dotted lines and gray-shaded $1 \sigma$ uncertainty regions were obtained by interpolating the curves in Fig. \ref{['fig:grids']}.
• Figure 5: Comparison of measured X-ray and EUV surface fluxes with values based on synthetic EMDs. This is the case for stars in Group 2 and iron abundance fixed to the best-fit values for each star. Top row: scatter plot of X-ray fluxes (left), and cumulative distributions of $\Sigma$--deviations (center) and flux ratios (right). Vertical dotted lines mark the median values and the 25% and 75% quantiles. Bottom row: similar plots, but for the EUV surface fluxes.