No Arabic abstract
The space-borne missions CoRoT and Kepler have opened a new era in stellar physics, especially for evolved stars, with precise asteroseismic measurements that help determine precise stellar parameters and perform ensemble astero seismology. This paper deals with the quality of the information that we can retrieve from the oscillations. It focusses on the conditions for obtaining the most accurate measurement of the radial and non-radial oscillation patterns. This accuracy is a prerequisite for making the best with asteroseismic data. From radial modes, we derive proxies of the stellar mass and radii with an unprecedented accuracy for field stars. For dozens of subgiants and thousands of red giants, the identification of mixed modes (corresponding to gravity waves propagating in the core coupled to pressure waves propagating in the envelope) indicates unambiguously their evolutionary status. As probes of the stellar core, these mixed modes also reveal the internal differential rotation and show the spinning down of the core rotation of stars ascending the red giant branch. A toy model of the coupling of waves constructing mixed modes is exposed, for illustrating many of their features.
The detection of oscillations with a mixed character in subgiants and red giants allows us to probe the physical conditions in their cores. With these mixed modes, we aim at determining seismic markers of stellar evolution. Kepler asteroseismic data were selected to map various evolutionary stages and stellar masses. Seismic evolutionary tracks were then drawn with the combination of the frequency and period spacings. We measured the asymptotic period spacing for more than 1170 stars at various evolutionary stages. This allows us to monitor stellar evolution from the main sequence to the asymptotic giant branch and draw seismic evolutionary tracks. We present clear quantified asteroseismic definitions that characterize the change in the evolutionary stages, in particular the transition from the subgiant stage to the early red giant branch, and the end of the horizontal branch.The seismic information is so precise that clear conclusions can be drawn independently of evolution models. The quantitative seismic information can now be used for stellar modeling, especially for studying the energy transport in the helium-burning core or for specifying the inner properties of stars entering the red or asymptotic giant branches. Modeling will also allow us to study stars that are identified to be in the helium-subflash stage, high-mass stars either arriving or quitting the secondary clump, or stars that could be in the blue-loop stage.
Seismic observations have shown that a number of evolved stars exhibit low-amplitude dipole modes, which are referred to as depressed modes. Recently, these low amplitudes have been attributed to the presence of a strong magnetic field in the stellar core of those stars. We intend to study the properties of depressed modes in evolved stars, which is a necessary condition before concluding on the physical nature of the mechanism responsible for the reduction of the dipole mode amplitudes. We perform a thorough characterization of the global seismic parameters of depressed dipole modes and show that these modes have a mixed character. The observation of stars showing dipole mixed modes that are depressed is especially useful for deriving model-independent conclusions on the dipole mode damping. Observations prove that depressed dipole modes in red giants are not pure pressure modes but mixed modes. This result invalidates the hypothesis that the depressed dipole modes result from the suppression of the oscillation in the radiative core of the stars. Observations also show that, except for the visibility, the seismic properties of the stars with depressed modes are equivalent to those of normal stars. The mixed nature of the depressed modes in red giants and their unperturbed global seismic parameters carry strong constraints on the physical mechanism responsible for the damping of the oscillation in the core. This mechanism is able to damp the oscillation in the core but cannot fully suppress it. Moreover, it cannot modify the radiative cavity probed by the gravity component of the mixed modes. The recent mechanism involving high magnetic field proposed for explaining depressed modes is not compliant with the observations and cannot be used to infer the strength and the prevalence of high magnetic fields in red giants.
Due to its unique long-term coverage and high photometric precision, observations from the Kepler asteroseismic investigation will provide us with the possibility to sound stellar cycles in a number of solar-type stars with asteroseismology. By comparing these measurements with conventional ground-based chromospheric activity measurements we might be able to increase our understanding of the relation between the chromospheric changes and the changes in the eigenmodes. In parallel with the Kepler observations we have therefore started a programme at the Nordic Optical Telescope to observe and monitor chromospheric activity in the stars that are most likely to be selected for observations for the whole satellite mission. The ground-based observations presented here can be used both to guide the selection of the special Kepler targets and as the first step in a monitoring programme for stellar cycles. Also, the chromospheric activity measurements obtained from the ground-based observations can be compared with stellar parameters such as ages and rotation in order to improve stellar evolution models.
By combining ground-based spectrographic observations of variability in the chromospheric emission from Sun-like stars with the variability seen in their eigenmode frequencies, it is possible to relate the changes observed at the surfaces of these stars to the changes taking place in the interior. By further comparing this variability to changes in the relative flux from the stars, one can obtain an expression for how these activity indicators relate to the energy output from the stars. Such studies become very pertinent when the variability can be related to stellar cycles as they can then be used to improve our understanding of the solar cycle and its effect on the energy output from the Sun. Here we present observations of chromospheric emission in 20 Sun-like stars obtained over the course of the nominal 4-year Kepler mission. Even though 4 years is too short to detect stellar equivalents of the 11-year solar cycle, observations from the Kepler mission can still be used to analyse the variability of the different activity indicators thereby obtaining information of the physical mechanism generating the variability. The analysis reveals no strong correlation between the different activity indicators, except in very few cases. We suggest that this is due to the sparse sampling of our ground-based observations on the one hand and that we are likely not tracing cyclic variability on the other hand. We also discuss how to improve the situation.
Turbulent motions in the convective envelope of red giants excite a rich spectrum of solar-like oscillation modes. Observations by CoRoT and Kepler have shown that the mode amplitudes increase dramatically as the stars ascend the red giant branch, i.e., as the frequency of maximum power, $ u_mathrm{max}$, decreases. Most studies nonetheless assume that the modes are well described by the linearized fluid equations. We investigate to what extent the linear approximation is justified as a function of stellar mass $M$ and $ u_mathrm{max}$, focusing on dipole mixed modes with frequency near $ u_mathrm{max}$. A useful measure of a modes nonlinearity is the product of its radial wavenumber and its radial displacement, $k_r xi_r$ (i.e., its shear). We show that $k_r xi_r propto u_mathrm{max}^{-9/2}$, implying that the nonlinearity of mixed modes increases significantly as a star evolves. The modes are weakly nonlinear ($k_r xi_r > 10^{-3}$) for $ u_mathrm{max} lesssim 150 , mumathrm{Hz}$ and strongly nonlinear ($k_r xi_r > 1$) for $ u_mathrm{max} lesssim 30 , mumathrm{Hz}$, with only a mild dependence on $M$ over the range we consider ($1.0 - 2.0 M_odot$). A weakly nonlinear mixed mode can excite secondary waves in the stellar core through the parametric instability, resulting in enhanced, but partial, damping of the mode. By contrast, a strongly nonlinear mode breaks as it propagates through the core and is fully damped there. Evaluating the impact of nonlinear effects on observables such as mode amplitudes and linewidths requires large mode network simulations. We plan to carry out such calculations in the future and investigate whether nonlinear damping can explain why some red giants exhibit dipole modes with unusually small amplitudes, known as depressed modes.