Chris (preliminary): Type Ia at z=1.08? Seems too bright for such a high redshift. An alternative is z=0.85, but this does not fit the broad bump at 8500 Angstroms.
Andy (preliminary): These data have been run through my program which subtracts off host galaxy and does a chi-sq fit to a library of templates. See results above. The best match is indeed at z=1.08 as Chris found. While this solution is best, it is far from unique. Several SNe match in the z=0.57 range (esp. Ib/c's), and several match in the z=0.83 range. (See above links for some representative examples). Furthermore, my program does not cover all of parameter space -- I have no hypernovae spectra for example.
Chris (preliminary): Gaston and I had a closer look at S02-002 and the z=1.08 looks more likely. There is some evidence of of H, K and OII at z=1.086, or H and K at z=1.067. The presence of the H and K lines, if real, indicates that there is significant host galaxy contamination in the spectrum of the supernova.
Lifan (preliminary): I have made a figure to illustrate possible host contaminations of S02-002. The fits suggest significant contamination redward of 9500 A but less so blueward. The best fit (pages 2 & 3) is for SN 90N @ -07 days, z = 1.09.
Andy (preliminary): Now that there are galaxy lines I am much more confident in the Ia at z=1.08 diagnosis.
Chris' Web Page