We have performed a comparative analysis of three recent and reliable SnIa datasets available in the literature: the Full Gold (FG) dataset (157 data points $0<z<1.75$), a Truncated Gold (TG) dataset (140 data points $0<z<1$) and the most recent Supernova Legacy Survey (SNLS) dataset (115 data points $0<z<1$). We have analyzed and compared the likelihood of cosmological constant and dynamical dark energy parametrizations allowing for crossing of the phantom divide line (PDL). We find that, even though the constraints obtained using the three datasets are consistent with each other at the 95% confidence level, the latest (SNLS) dataset shows distinct trends which are not shared by the Gold datasets. We find that the best fit dynamical $w(z)$ obtained from the SNLS dataset does not cross the PDL $w=-1$ and remains above and close to the $w=-1$ line for the whole redshift range $0<z<1$ showing no evidence for phantom behavior. The $\Lambda +\mathrm{\text{Cold}}\mathrm{\text{Dark}}\mathrm{\text{Matter}}$ (LCDM) parameter values ($w_0=-1$, $w_1=0$) almost coincide with the best fit parameters of the dynamical $w(z)$ parametrizations. In contrast, the best fit dynamical $w(z)$ obtained from the Gold datasets (FG and TG) clearly crosses the PDL and departs significantly from the PDL $w=-1$ line while the LCDM parameter values are about $2\sigma$ away from the best fit $w(z)$. In addition, the $({\Omega }_{0m},{\Omega }_{\Lambda })$ parameters in a LCDM parametrization without a flat prior, fit by the SNLS dataset, favor the minimal flat LCDM concordance model. The corresponding fit with the Gold datasets mildly favors a closed universe and the flat LCDM parameter values are $1\sigma –2\sigma$ away from the best fit $({\Omega }_{0m},{\Omega }_{\Lambda })$.

• Received 1 November 2005

DOI:https://doi.org/10.1103/PhysRevD.72.123519