GMOS optical long-slit spectroscopy at the Gemini-North telescope was used to classify targets from the Supernova Legacy Survey (SNLS) from July 2005 and May 2006 - May 2008. During this time, 95 objects were observed. Where possible the objects' redshifts (z) were measured from narrow emission or absorption features in the host galaxy spectrum, otherwise they were measured from the broader supernova features. We present spectra of 68 confirmed or probable SNe Ia from SNLS with redshifts in the range 0.17 \leq z \leq 1.02. In combination with earlier SNLS Gemini and VLT spectra, we used these new observations to measure pseudo-equivalent widths (EWs) of three spectral features - CaII H&K, SiII and MgII - in 144 objects and compared them to the EWs of low-redshift SNe Ia from a sample drawn from the literature. No signs of changes with z are seen for the CaII H&K and MgII features. Systematically lower EW SiII is seen at high redshift, but this can be explained by a change in demographics of the SNe Ia population within a two-component model combined with an observed correlation between EW SiII and photometric lightcurve stretch.

Aims: We present a quantitative study of a new data set of high redshift Type Ia supernovae spectra, observed at the Gemini telescopes during the first 34 months of the Supernova Legacy Survey. During this time 123 supernovae candidates were observed, of which 87 have been identified as SNe Ia at a median redshift of z=0.720. Spectra from the entire second year of the survey and part of the third year (59 total SNe candidates with 46 confirmed SNe Ia) are published here for the first time. The spectroscopic measurements made on this data set are used determine if these distant SNe comprise a population similar to those observed locally. Methods: Rest-frame equivalent width and ejection velocity measurements are made on four spectroscopic features. Corresponding measurements are presented for a set of 167 spectra from 24 low-z SNe Ia from the literature. Results: We show that there exists a sample at high redshift with properties similar to nearby SNe. No significant difference was found between the distributions of measurements at low and high redsift for three of the features. The fourth feature displays a possible difference that should be investigated further. Correlations between Type Ia SNe properties and host galaxy morphology were also found to be similar at low and high z, and within each host galaxy class we see no evidence for redshift-evolution in SN properties. A new correlation between SNe Ia peak magnitude and the equivalent width of SiII absorption is presented. We demonstrate that this correlation reduces the scatter in SNe Ia luminosity distances in a manner consistent with the lightcurve shape-luminosity corrections that are used for Type Ia SNe cosmology. Conclusions: We show that this new sample of SNLS SNe Ia has spectroscopic properties similar to nearby objects. (Abridged)

We present new techiques for improving the efficiency of supernova (SN) classification at high redshift using 64 candidates observed at Gemini North and South during the first year of the Supernova Legacy Survey (SNLS). The SNLS is an ongoing five year project with the goal of measuring the equation of state of Dark Energy by discovering and following over 700 high-redshift SNe Ia using data from the Canada-France-Hawaii Telescope Legacy Survey. We achieve an improvement in the SN Ia spectroscopic confirmation rate: at Gemini 71% of candidates are now confirmed as SNe Ia, compared to 54% using the methods of previous surveys. This is despite the comparatively high redshift of this sample, where the median SN Ia redshift is z=0.81 (0.155 <= z <= 1.01). These improvements were realized because we use the unprecedented color coverage and lightcurve sampling of the SNLS to predict whether a candidate is an SN Ia and estimate its redshift, before obtaining a spectrum, using a new technique called the "SN photo-z." In addition, we have improved techniques for galaxy subtraction and SN template chi^2 fitting, allowing us to identify candidates even when they are only 15% as bright as the host galaxy. The largest impediment to SN identification is found to be host galaxy contamination of the spectrum -- when the SN was at least as bright as the underlying host galaxy the target was identified more than 90% of the time. However, even SNe on bright host galaxies can be easily identified in good seeing conditions. When the image quality was better than 0.55 arcsec the candidate was identified 88% of the time. Over the five-year course of the survey, using the selection techniques presented here we will be able to add approximately 170 more confirmed SNe Ia than would be possible using previous methods.