The possibility of reconstruction of Lagrangian for the scalar field dark energy with constant effective sound speed $c_s$ is analyzed. It is found that such reconstruction can be made with accuracy up to an arbitrary constant. The value of $c_s$ is
estimated together with other dark energy parameters ($Omega_{de}$, $w_0$, $c_a^2$) and main cosmological ones on the basis of data including Planck-2013 results on CMB anisotropy, BAO distance ratios from recent galaxy surveys, galaxy power spectrum from WiggleZ, magnitude-redshift relations for distant SNe Ia from SNLS3 and Union2.1 compilations, the HST determination of the Hubble constant. It is shown that no value of $c_s$ from the range [0,1] is preferred by the used data because of very weak influence of dark energy perturbations on the large scale structure formation and CMB temperature fluctuations.
We constrain the parameters of dynamical dark energy in the form of a classical scalar field with barotropic equation of state jointly with other cosmological parameters using various combined datasets including the CMB power spectra from WMAP7, the
baryon acoustic oscillations in the space distribution of galaxies from SDSS DR7 and WiggleZ, the light curves of SN Ia from 3 different compilations: SDSS (SALT2 and MLCS2k2 light curve fittings), SNLS3 and Union2.1. The considered class of models involves both quintessential and phantom subclasses. The analysis has shown that the phantom models are generally preferred by the observational data. We discuss the effect of allowing for non-zero masses of active neutrinos, non-zero curvature or non-zero contribution from the tensor mode of perturbations on the precision of dark energy parameters estimation. We also perform a forecast for the Planck mock data.
We constrain the parameters of dynamical dark energy in the form of a classical or tachyonic scalar field with barotropic equation of state jointly with other cosmological ones using the combined datasets which include the CMB power spectra from WMAP
7, the baryon acoustic oscillations in the space distribution of galaxies from SDSS DR7, the power spectrum of luminous red galaxies from SDSS DR7 and the light curves of SN Ia from 2 different compilations: Union2 (SALT2 light curve fitting) and SDSS (SALT2 and MLCS2k2 light curve fittings). It has been found that the initial value of dark energy equation of state parameter is constrained very weakly by most of the data while the rest of main cosmological parameters are well constrained: their likelihoods and posteriors are similar, have the forms close to Gaussian (or half-Gaussian) and their confidential ranges are narrow. The most reliable determinations of the best fitting value and $1sigma$ confidence range for the initial value of dark energy equation of state parameter were obtained from the combined datasets including SN Ia data from the full SDSS compilation with MLCS2k2 fitting of light curves. In all such cases the best fitting value of this parameter is lower than the value of corresponding parameter for current epoch. Such dark energy loses its repulsive properties and in future the expansion of the Universe will change into contraction. We also perform an error forecast for the Planck mock data and show that they narrow essentially the confidential ranges of cosmological parameters values, moreover, their combination with SN SDSS compilation with MLCS2k2 light curve fitting may exclude the fields with initial equation of state parameter $>-0.1$ at 2$sigma$ confidential level.
