We aim to present 70 spectra of 68 new high-redshift type Ia supernovae (SNeIa) measured at ESO's VLT during the final two years of operation (2006-2008) of the Supernova Legacy Survey (SNLS). We use the full five year SNLS VLT spectral set to investigate a possible spectral evolution of SNeIa populations with redshift and study spectral properties as a function of lightcurve fit parameters and the mass of the host-galaxy. Reduction and extraction are based on both IRAF standard tasks and our own reduction pipeline. Redshifts are estimated from host-galaxy lines whenever possible or alternatively from supernova features. We used the spectrophotometric SNIa model SALT2 combined with a set of galaxy templates that model the host-galaxy contamination to assess the type Ia nature of the candidates. We identify 68 new SNeIa with redshift ranging from z=0.207 to z=0.98 (<z>=0.62). Each spectrum is presented individually along with its best-fit SALT2 model. The five year dataset contains 209 spectra corresponding to 192 SNeIa identified at the VLT. We also publish the redshifts of other candidates (host galaxies or other transients) whose spectra were obtained at the same time as the spectra of live SNe Ia. Using the full VLT SNeIa sample, we build composite spectra around maximum light with cuts in color, lightcurve shape parameter ('stretch'), host-galaxy mass and redshift. We find that high-z SNeIa are bluer, brighter and have weaker intermediate mass element absorption lines than their low-z counterparts at a level consistent with what is expected from selection effects. We also find a flux excess in the range [3000-3400] A for SNeIa in low mass host-galaxies or with locally blue U-V colors, and suggest that the UV flux (or local color) may be used in future cosmological studies as a third standardization parameter in addition to stretch and color.

We use simulated SN Ia samples, including both photometry and spectra, to perform the first direct validation of cosmology analysis using the SALT-II light curve model. This validation includes residuals from the light curve training process, systematic biases in SN Ia distance measurements, and the bias on the dark energy equation of state parameter w. Using the SN-analysis package SNANA, we simulate and analyze realistic samples corresponding to the data samples used in the SNLS3 analysis: 120 low-redshift (z < 0.1) SNe Ia, 255 SDSS SNe Ia (z < 0.4), and 290 SNLS SNe Ia (z <= 1). To probe systematic uncertainties in detail, we vary the input spectral model, the model of intrinsic scatter, and the smoothing (i.e., regularization) parameters used during the SALT-II model training. Using realistic intrinsic scatter models results in a slight bias in the ultraviolet portion of the trained SALT-II model, and w biases (winput - wrecovered) ranging from -0.005 +/- 0.012 to -0.024 +/- 0.010. These biases are indistinguishable from each other within uncertainty; the average bias on w is -0.014 +/- 0.007.

We present cosmological constraints from a joint analysis of type Ia supernova (SN Ia) observations obtained by the SDSS-II and SNLS collaborations. The data set includes several low-redshift samples (z<0.1), all 3 seasons from the SDSS-II (0.05 < z < 0.4), and 3 years from SNLS (0.2 <z < 1) and totals \ntotc spectroscopically confirmed type Ia supernovae with high quality light curves. We have followed the methods and assumptions of the SNLS 3-year data analysis except for the following important improvements: 1) the addition of the full SDSS-II spectroscopically-confirmed SN Ia sample in both the training of the SALT2 light curve model and in the Hubble diagram analysis (\nsdssc SNe), 2) inter-calibration of the SNLS and SDSS surveys and reduced systematic uncertainties in the photometric calibration, performed blindly with respect to the cosmology analysis, and 3) a thorough investigation of systematic errors associated with the SALT2 modeling of SN Ia light-curves. We produce recalibrated SN Ia light-curves and associated distances for the SDSS-II and SNLS samples. The large SDSS-II sample provides an effective, independent, low-z anchor for the Hubble diagram and reduces the systematic error from calibration systematics in the low-z SN sample. For a flat LCDM cosmology we find Omega_m=0.295+-0.034 (stat+sys), a value consistent with the most recent CMB measurement from the Planck and WMAP experiments. Our result is 1.8sigma (stat+sys) different than the previously published result of SNLS 3-year data. The change is due primarily to improvements in the SNLS photometric calibration. When combined with CMB constraints, we measure a constant dark-energy equation of state parameter w=-1.018+-0.057 (stat+sys) for a flat universe. Adding BAO distance measurements gives similar constraints: w=-1.027+-0.055.

For spectroscopically confirmed Type Ia supernovae we evaluate models of intrinsic brightness variations with detailed data/Monte Carlo comparisons of the dispersion in the following quantities: Hubble-diagram scatter, color difference (B-V-c) between the true B-V color and the fitted color (c) from the SALT-II light curve model, and photometric redshift residual. The data sample includes 251 ugriz light curves from the 3-season Sloan Digital Sky Survey-II, and 191 griz light curves from the Supernova Legacy Survey 3-year data release. We find that the simplest model of a wavelength-independent (coherent) scatter is not adequate, and that to describe the data the intrinsic scatter model must have wavelength-dependent variations. We use Monte Carlo simulations to examine the standard approach of adding a coherent scatter term in quadrature to the distance-modulus uncertainty in order to bring the reduced chi2 to unity when fitting a Hubble diagram. If the light curve fits include model uncertainties with the correct wavelength dependence of the scatter, we find that the bias on the dark energy equation of state parameter $w$ is negligible. However, incorrect model uncertainties can lead to a significant bias on the distance moduli, with up to ~0.05 mag redshift-dependent variation. For the recent SNLS3 cosmology results we estimate that this effect introduces an additional systematic uncertainty on $w$ of ~0.02, well below the total uncertainty. However, this uncertainty depends on the samples used, and thus this small $w$-uncertainty is not guaranteed in future cosmology results.