GCN Circular 286

Bradley E. Schaefer and Eric H. McKenzie (Yale University) report:

"We have obtained 139 photometric observations in B, V, and I of the late
time light curve for SN1998bw associated with GRB980425.  These
demonstrate a strikingly linear decline in magnitude versus time.  

Our data was taken with the Yale 1.0-m telescope on Cerro Tololo between
27 June and 28 October 1998.  We used standard IRAF reduction and the
comparison stars of Galama et al.
(http://www.astro.uva.nl/~titus/grb980425/grb980425chart.html).  During
the time interval of our observations, the light from the underlying
galaxy (ESO 184-G82) was insignificant in our photometry.
	
Our first and last data nights had measured photometry as follows:
  JD2450992.9   B=16.68+-0.03, V=15.79+-0.02, I=15.16+-0.03
  JD2451115.6   B=18.35+-0.06, V=18.09+-0.05, I=17.33+-0.06
Between these two nights, all our photometry is perfectly consistent with
an exactly linear decline in magnitudes (hence an exponential decline in
luminosity).  Our measured uniform decline rates are as follows:
  B   0.0141+-0.0002 mag/day for equivalent half-life of 53.4+-0.8 days
  V   0.0184+-0.0003 mag/day for equivalent half-life of 40.9+-0.7 days
  I   0.0181+-0.0003 mag/day for equivalent half-life of 41.6+-0.7 days

The observed exponential decline is in contradiction to the theoretical
predictions of Iwamoto et al. (1998, Nature, 395, 672) and Iwamoto (1999,
ApJ, astro-ph/9810400), which claims that the decline will be as a power
law.  The observed exponential decline has a rate similar to that expected
from the decay of radioactive cobalt as modified by the effects due to the
expansion of the shell.  So it is reasonable to conclude that the
late-time light curve of SN1998bw is being powered by radioactive cobalt.
This then implies that the underlying explosion mechanism must create
large masses of radioactive cobalt.

The light curve of SN1998bw is significantly different from all
previously known supernovae.  For a comparison with Type Ia events, the
decline rate of SN1998bw is the same in B but not V and I as for Type Ia
events, while SN1998bw does not show the bump in the I band light curve
from 20-50 days after peak.  A comparison with Type Ic events is difficult
since their lights curves are not well defined, yet the late time decline
rate of Type Ic events is substantially smaller than for SN1998bw.
	
Our observations of SN1998bw are continuing with the Yale 1.0-m
telescope.  However, since the source has come out from behind the Sun,
the light from the galaxy  provides an increasingly significant
obstacle to accurate photometry.  Perhaps future accurate photometry must
await the complete fading of SN1998bw to allow for subtraction of the
galaxy light."