From: Greg Aldering (aldering@panisse.lbl.gov)
Date: Sat Mar 15 2003 - 17:43:29 PST
Hi Rob,
I haven't located a description of what you did in the following
areas:
1) How did you set the relative B and V stretches?
For P99 we performed independent B and V fits to the Hamuy
data and then plotted s_V versus s_B and chose a ratio
for s_V/s_B from that. This was a matter of some discussion,
and not entirely trivial. You must have determined an initial
ratio and then renormalized the templates to have s_V/s_B = 1,
but how did you determine the ratio in the first place?
We know that there is real scatter in this ratio, so the value
used must be based on an ensemble of well-measured SNe in order to
get the best (albeit approximate) value.
Since you perform joint BV fits, this is probably even more
important (esp. for low-z SNe) than it was for P99. Moreover,
it would be preferable to eliminate this as a difference in
your reanalysis if at all possible - that is, if you don't have
any better choice than in P99 we should try to use the P99
choice.
2) How did you establish t_Bmax - t_Vmax with your new templates?
Again, we know that this value is not really constant between
SNe, but we treat it as such in SNminuit. Thus, it can't be determined
from just one well-measure SN.
These same questions apply to U-band and R-band as well, although the
choice affects fewer SNe and has less impact of the comparison between
your current fits and those from P99.
Thanks,
Greg
This archive was generated by hypermail 2.1.4 : Sat Mar 15 2003 - 17:43:29 PST