Aims. We investigate the impact of realistic three-dimensional (3D) hydrodynamical model atmospheres of red giant stars at different metallicities on the formation of spectral lines of a number of ions and molecules. Methods. We carry out realistic, ab initio, 3D, hydrodynamical simulations of surface convection at the surface of red giant stars with varying effective temperatures and metallicities. We use the convection simulations as time-dependent hydrodynamical model stellar atmospheres to calculate spectral lines of a number of ions (Li i, O i, Na i, Mg i, Ca i, Fe i, and Fe ii) and molecules (CH, NH, and OH) under the assumption of local thermodynamic equilibrium (LTE). We carry out a differential comparison of the line strengths computed in 3D with the results of analogous line formation calculations for classical, 1D, hydrostatic, plane-parallel marcs model atmospheres in order to estimate the impact of 3D models on the derivation of elemental abundances. Results. The temperature and density inhomogeneities and correlated velocity fields in 3D models, as well as the differences between the mean 3D stratifications and corresponding 1D model atmospheres significantly affect the predicted strengths of spectral lines. Under the assumption of LTE, the low atmospheric temperatures encountered in 3D model atmospheres of very metal-poor giant stars cause spectral lines from neutral species and molecules to appear stronger than within the framework of 1D models. As a consequence, elemental abundances derived from these lines using 3D models are significantly lower than according to 1D analyses. In particular, the differences between 3D and 1D abundances of C, N, and O derived from CH, NH, and OH weak low-excitation lines are found to be in the range −0.5 dex to −1.0 dex for the the red giant stars at [Fe/H] = −3 considered here. At this metallicity, large negative corrections (about −0.8 dex) are also found, in LTE, for weak low-excitation Fe i lines. We caution, however, that the neglected departures from LTE might be significant for these and other elements and comparable to the effects due to stellar granulation.

Aims. We present the Stagger-grid, a comprehensive grid of time-dependent, three-dimensional (3D), hydrodynamic model atmospheres for late-type stars with realistic treatment of radiative transfer, covering a wide range in stellar parameters. This grid of 3D models is intended for various applications besides studies of stellar convection and atmospheres per se, including stellar parameter determination, stellar spectroscopy and abundance analysis, asteroseismology, calibration of stellar evolution models, interferometry, and extrasolar planet search. In this introductory paper, we describe the methods we applied for the computation of the grid and discuss the general properties of the 3D models as well as of their temporal and spatial averages (here denoted 3D models). Methods. All our models were generated with the Stagger-code, using realistic input physics for the equation of state (EOS) and for continuous and line opacities. Our ∼220 grid models range in effective temperature, T eff , from 4000 to 7000 K in steps of 500 K, in surface gravity, log g, from 1.5 to 5.0 in steps of 0.5 dex, and metallicity, [Fe/H], from −4.0 to +0.5 in steps of 0.5 and 1.0 dex. Results. We find a tight scaling relation between the vertical velocity and the surface entropy jump, which itself correlates with the constant entropy value of the adiabatic convection zone. The range in intensity contrast is enhanced at lower metallicity. The granule size correlates closely with the pressure scale height sampled at the depth of maximum velocity. We compare the 3D models with currently widely applied one-dimensional (1D) atmosphere models, as well as with theoretical 1D hydrostatic models generated with the same EOS and opacity tables as the 3D models, in order to isolate the effects of using self-consistent and hydrodynamic modeling of convection, rather than the classical mixing length theory approach. For the first time, we are able to quantify systematically over a broad range of stellar parameters the uncertainties of 1D models arising from the simplified treatment of physics, in particular convective energy transport. In agreement with previous findings, we find that the differences can be rather significant, especially for metal-poor stars.

The granulation pattern that we observe on the surface of the Sun is due to hot plasma from the interior rising to the photosphere where it cools down, and descends back into the interior at the edges of granules. This is the visible manifestation of convection taking place in the outer part of the solar convection zone. Because red giants have deeper convection zones and more extended atmospheres than the Sun, we cannot a priori assume that granulation in red giants is a scaled version of solar granulation. Until now, neither observations nor 1D analytical convection models could put constraints on granulation in red giants.However, thanks to asteroseismology, this study can now be performed. The resulting parameters yield physical information about the granulation. We analyze ∼ 1000 red giants that have been observed by Kepler during 13 months. We fit the power spectra with Harvey-like profiles to retrieve the characteristics of the granulation (time scale τ gran and power P gran ). We also introduce a new time scale, τ eff , which takes into account that different slopes are used in the Harvey functions. We search for a correlation between these parameters and the global acoustic-mode parameter (the position of maximum power, ν max ) as well as with stellar parameters (mass, radius, surface gravity (log g) and effective temperature (T eff )). We show that τ eff ∝ ν −0.89 max and P gran ∝ ν −1.90 max , which is consistent with the theoretical predictions. We find that the granulation time scales of stars that belong to the red clump have similar values while the time scales of stars in the red-giant branch are spread in a wider range. Finally, we show that realistic 3D simulations of the surface convection in stars, spanning the (T eff , log g)-range of our sample of red giants, match the Kepler observations well in terms of trends.

Asteroseismology with the Kepler space telescope is providing not only an improved characterization of exoplanets and their host stars, but also a new window on stellar structure and evolution for the large sample of solar-type stars in the field. We perform a uniform analysis of 22 of the brightest asteroseismic targets with the highest signal-to-noise ratio observed for 1 month each during the first year of the mission, and we quantify the precision and relative accuracy of asteroseismic determinations of the stellar radius, mass, and age that are possible using various methods. We present the properties of each star in the sample derived from an automated analysis of the individual oscillation frequencies and other observational constraints using the Asteroseismic Modeling Portal (AMP), and we compare them to the results of model-grid-based methods that fit the global oscillation properties. We find that fitting the individual frequencies typically yields asteroseismic radii and masses to ∼1% precision, and ages to ∼2.5% precision (respectively 2, 5, and 8 times better than fitting the global oscillation properties). The absolute level of agreement between the results from different approaches is also encouraging, with model-grid-based methods yielding slightly smaller estimates of the radius and mass and slightly older values for the stellar age relative to AMP, which computes a large number of dedicated models for each star. The sample of targets for which this type of analysis is possible will grow as longer data sets are obtained during the remainder of the mission. 15 Remaining affiliations removed due to arXiv error 16 Kepler data are collected by quarters that lasted three months except for the first quarter, which lasted one month (referred as Q1). One month of the other quarters are denoted as Q2.1 for example to refer to the first month of the second quarter.

Present grids of stellar atmosphere models are the workhorses in interpreting stellar observations, and determining their fundamental parameters. These models rely on greatly simplified models of convection, however, lending less predictive power to such models of late type stars. We present a grid of improved and more reliable stellar atmosphere models of late type stars, based on deep, 3D, convective, stellar atmosphere simulations. This grid is to be used in general for interpreting observations, and improve stellar and asteroseismic modeling. We solve the Navier Stokes equations in 3D and concurrent with the radiative transfer equation, for a range of atmospheric parameters, covering most of stellar evolution with convection at the surface. We emphasize use of the best available atomic physics for quantitative predictions and comparisons with observations. We present granulation size, convective expansion of the acoustic cavity, asymptotic adiabat, as function of atmospheric parameters. These and other results are also available in electronic form.

We perform a calibration of the mixing length of convection in stellar structure models against realistic 3D radiation-coupled hydrodynamics (RHD) simulations of convection in stellar surface layers, determining the adiabat deep in convective stellar envelopes.The mixing-length parameter α is calibrated by matching averages of the 3D simulations to 1D stellar envelope models, ensuring identical atomic physics in the two cases. This is done for a previously published grid of solar-metallicity convection simulations, covering from 4 200 K to 6 900 K on the main sequence, and 4 300-5 000 K for giants with log g = 2.2.Our calibration results in an α varying from 1.6 for the warmest dwarf, which is just cool enough to admit a convective envelope, and up to 2.05 for the coolest dwarfs in our grid. In between these is a triangular plateau of α ∼ 1.76. The Sun is located on this plateau and has seen little change during its evolution so far. When stars ascend the giant branch, they largely do so along tracks of constant α, with α decreasing with increasing mass.

Aims. We present the implementation of a radiative transfer solver with coherent scattering in the new BIFROST code for radiative magneto-hydrodynamical (MHD) simulations of stellar surface convection. The code is fully parallelized using MPI domain decomposition, which allows for large grid sizes and improved resolution of hydrodynamical structures. We apply the code to simulate the surface granulation in a solar-type star, ignoring magnetic fields, and investigate the importance of coherent scattering for the atmospheric structure. Methods. A scattering term is added to the radiative transfer equation, requiring an iterative computation of the radiation field. We use a short-characteristics-based Gauss-Seidel acceleration scheme to compute radiative flux divergences for the energy equation. The effects of coherent scattering are tested by comparing the temperature stratification of three 3D time-dependent hydrodynamical atmosphere models of a solar-type star: without scattering, with continuum scattering only, and with both continuum and line scattering. Results. We show that continuum scattering does not have a significant impact on the photospheric temperature structure for a star like the Sun. Including scattering in line-blanketing, however, leads to a decrease of temperatures by about 350 K below log 10 τ 5000 < ∼ −4. The effect is opposite to that of 1D hydrostatic models in radiative equilibrium, where scattering reduces the cooling effect of strong LTE lines in the higher layers of the photosphere. Coherent line scattering also changes the temperature distribution in the high atmosphere, where we observe stronger fluctuations compared to a treatment of lines as true absorbers.

Aims. We extend semi-analytical computations of excitation rates for solar oscillation modes to those of other solar-like oscillating stars to compare them with recent observations Methods. Numerical 3D simulations of surface convective zones of several solar-type oscillating stars are used to characterize the turbulent spectra as well as to constrain the convective velocities and turbulent entropy fluctuations in the uppermost part of the convective zone of such stars. These constraints, coupled with a theoretical model for stochastic excitation, provide the rate P at which energy is injected into the p-modes by turbulent convection. These energy rates are compared with those derived directly from the 3D simulations.Results. The excitation rates obtained from the 3D simulations are systematically lower than those computed from the semi-analytical excitation model. We find that P max , the P maximum, scales as (L/M) s where s is the slope of the power law and L and M are the mass and luminosity of the 1D stellar model built consistently with the associated 3D simulation. The slope is found to depend significantly on the adopted form of χ k , the eddy time-correlation; using a Lorentzian, χ L k , results in s = 2.6, whereas a Gaussian, χ G k , gives s = 3.1. Finally, values of V max , the maximum in the mode velocity, are estimated from the computed power laws for P max and we find that V max increases as (L/M) sv . Comparisons with the currently available ground-based observations show that the computations assuming a Lorentzian χ k yield a slope, sv, closer to the observed one than the slope obtained when assuming a Gaussian. We show that the spatial resolution of the 3D simulations must be high enough to obtain accurate computed energy rates.