During the recent years significant progress has been made in the modeling of red giant atmospheres with the aid of 3D hydrodynamical model atmosphere codes. In this contribution we provide an overview of selected results obtained in this context by utilizing 3D hydrodynamical CO5BOLD stellar model atmospheres. Hydrodynamical simulations show that convective motions lead to significant differences in the atmospheric structures of red giants with respect to those predicted by the classical 1D model atmospheres. Results of these simulations also show that in certain cases 1D models fail to reproduce even the average properties of the 3D hydrodynamical models, such as P-T profiles. Large horizontal temperature fluctuations in the 3D model atmospheres, as well as differences between the temperature profiles of the average <3D> and 1D models, lead to large discrepancies in the strengths of spectral lines predicted by the 3D and 1D model atmospheres. This is especially important in models at lowest metallicities ([M/H]<-2.0) where the 3D-1D abundance differences may reach (or even exceed) -0.6 dex for lines of neutral atoms and molecules. We also discuss several simplifications and numerical aspects involved in the present 3D hydrodynamical modeling of red giant atmospheres, and briefly address several issues where urgent progress may be needed.

We study the influence of convection on the asymmetries and Doppler shifts of Fe I spectral lines in the metal-poor red giant HD 122563. To this end, we compute theoretical Fe I line shifts and line bisectors using 3D hydrodynamical model atmosphere of HD 122563 calculated with the CO5BOLD code. We then make a detailed comparison of the theoretical line shifts and bisectors with those derived from the high quality HARPS spectrum of HD 122563 taken from the ESO Science Archive Facility (R = 115 000, average signal-to-noise ratio, S/N = 310). In general, we find a good agreement between the theoretically predicted and observed Doppler shifts of Fe I line cores, with somewhat larger discrepancies seen in the case of weaker (equivalent width W < 5 pm) and stronger lines (W > 11 pm). Both observed and theoretical coreshifts cover a range between 0 and -1 km/s, with increasingly stronger blueshifts for weaker lines and slight hints of a coreshift dependence on wavelength. Theoretical bisectors reproduce the observed ones reasonably well too, however, theoretical bisectors of the weak red (lambda > 600 nm) Fe I lines have blueshifts that are by up to 200 m/s larger than observed. The obtained results therefore suggest that the current CO5BOLD models are capable of reproducing the large-scale velocity fields in the atmosphere of HD 122563 suficiently well. Nevertheless, further efforts are needed in order to understand the physical reasons behind the discrepancies in theoretical predictions and observed properties of the weakest and strongest Fe I lines.

In this work we present first results of our current project aimed at combining the 3D hydrodynamical stellar atmosphere approach with non-LTE (NLTE) spectral line synthesis for a number of key chemical species. We carried out a full 3D-NLTE spectrum synthesis of the oxygen IR 777 nm triplet, using a modified and improved version of our NLTE3D package to calculate departure coefficients for the atomic levels of oxygen in a CO5BOLD 3D hydrodynamical solar model atmosphere. Spectral line synthesis was subsequently performed with the Linfor 3D code. In agreement with previous studies, we find that the lines of the oxygen triplet produce deeper cores under NLTE conditions, due to the diminished line source function in the line forming region. This means that the solar oxygen IR 777 nm lines should be stronger in NLTE, leading to negative 3D NLTE-LTE abundance corrections. Qualitatively this result would support previous claims for a relatively low solar oxygen abundance. Finally, we outline several further steps that need to be taken in order to improve the physical realism and numerical accuracy of our current 3D-NLTE calculations.

We investigate the role of convection in the formation of atomic and molecular lines in the atmosphere of a red giant star. For this purpose we study the formation properties of spectral lines that belong to a number of astrophysically important tracer elements, including neutral and singly ionized atoms, and molecules. We focus our investigation on a prototypical red giant located close to the red giant branch (RGB) tip (Teff=3660K, logg=1.0, [M/H]=0.0). We used two types of model atmospheres, 3D hydrodynamical and classical 1D, calculated with the CO5BOLD and LHD stellar atmosphere codes, respectively. Both codes share the same atmospheric parameters, chemical composition, equation of state, and opacities, which allowed us to make a strictly differential comparison between the line formation properties predicted in 3D and 1D. The influence of convection on the spectral line formation was assessed with the aid of 3D-1D abundance corrections, which measure the difference between the abundances of chemical species derived with the 3D hydrodynamical and 1D classical model atmospheres. We find that convection plays a significant role in the spectral line formation in this particular red giant. The derived 3D-1D abundance corrections rarely exceed \pm0.1 dex when lines of neutral atoms and molecules are considered, which is in line with the previous findings for solar-metallicity red giants located on the lower RGB. The situation is different with lines that belong to ionized atoms, or to neutral atoms with high ionization potential. In both cases, the corrections for high-excitation lines (\chi>8 eV) may amount to \Delta_3D-1D ~ -0.4 dex. The 3D--1D abundance corrections generally show a significant wavelength dependence; in most cases they are smaller in the near-infrared, at 1600-2500 nm.

We compare the abundances of various chemical species as derived with 3D hydrodynamical and classical 1D stellar atmosphere codes in a late-type giant characterized by T_eff=3640K, log g = 1.0, [M/H] = 0.0. For this particular set of atmospheric parameters the 3D-1D abundance differences are generally small for neutral atoms and molecules but they may reach up to 0.3-0.4 dex in case of ions. The 3D-1D differences generally become increasingly more negative at higher excitation potentials and are typically largest in the optical wavelength range. Their sign can be both positive and negative, and depends on the excitation potential and wavelength of a given spectral line. While our results obtained with this particular late-type giant model suggest that 1D stellar atmosphere models may be safe to use with neutral atoms and molecules, care should be taken if they are exploited with ions.