[ Previous | Next ]
ATEL # 2131; M. Fraser, S. J. Smartt, (Queens University Belfast), M. Crockett (Oxford), S. Mattila (University of Turku, Finland), A. Gal-Yam (Weizmann Institute, Israel) A. Stephens, K. Roth (Gemini Observatory)
on 15 Jul 2009; 12:06 UT
Password Certification: Stephen Smartt (firstname.lastname@example.org)
Subjects: Radio, Infra-Red, Nova, Supernova, Transients
We obtained a 30 minute exposure of the central region of M82, using the Near InfraRed Imager and Spectrometer (NIRI) together with the ALTAIR adaptive optics system on the Gemini North telescope. The observation was taken on June 11 2009 (54993.26 UT), using the K band filter and the f/14 camera (0.05 arcsec / pixel, over a 51 x 51 arcsec field of view).
The field of view covers two recent radio transients: the radio supernova SN 2008iz (CBET #1803) and the un-named transient described in ATEL #2073 (henceforth the "MERLIN transient"). We aligned the Gemini image to archival HST NICMOS images to search for the presence of a supernova, or the disappearance of a progenitor star in the pre-discovery images. We used images from the NIC2 camera with the F222W and F237W filters, taken on April 12 1998. We found no detection of either transient in the K-band Gemini images - inspection of the regions of the two radio transients does not reveal any difference between the post-explosion Gemini images and the pre-explosion NICMOS.
We performed image subtraction (ISIS 2.2 : http://www2.iap.fr/users/alard/package.html) between the pre and post explosion images. Due to the differences in filters, we essentially obtained a colour map of the field. The difficulties inherant in subtracting an adaptive-optics PSF were also a significant obstacles. No object was detected at either location in the K band. Assuming a distance modulus of 28.3 from the NASA Extragalactic Database (NED) we calculate a limiting absolute K band magnitude of -5 for a 3 sigma detection, and a limiting absolute magnitude of -5.7 for 5 sigma. This calculation was made based on pixel to pixel noise at the location, and given the crowded nature of the field is likely to be somewhat optimistic. Other searches for SNe in starbursts and LIRGs have used a 20 sigma limit, and hence the limiting magnitude could well be several magnitudes lower. Estimates of extinction in the core of M82 vary widely, a typical value is Av = 30 (see Mattila & Meikle, 2001), which corresponds to a K band extinction of about 4 magnitudes, althouugh it could be significantly higher.
Brunthaler et al. (CBET 1803) report that the measured expansion rate of SN2008iz indicates an explosion date in late January 2008. Our K-band image is ~500 days after this. Typical K-band absolute magnitudes for type II SNe at 500 days are diverse, ranging from -7 (normal type II) to -17 for slowly declining IIn and II-L SNe, and the latter tend to be bright radio sources. SN2008iz was 1.5 times brighter at 22Ghz than 1993J, suggesting that it is more likely to result from a IIn/II-L or IIb SN. The lack of K-band detection suggests either a much higher extinction (K-band extinction of up to 11 magnitudes), or a fainter SN but with peculiarly bright radio emission (or a combination of both)
For the "MERLIN Transient", the radio brightening is constrained to the period from the 27 April 2009 to 5 May 2009, or between 38 and 46 days prior to the Gemini image. If this radio transient is caused by prompt emission from a supernova, and comparing the radio light curve to that of SN 1993J (Benz, 1996), the appearance of the transient at 5 GHz would indicate an explosion epoch less than 150 days prior. At this epoch, SN 1993J had a K band absolute magnitude of -15. From this we conclude that the transient is unlikely to be caused by the prompt emission from a supernova, or that the extinction towards the supernova is on the order of 9 magnitudes or higher in the K band. Again, however, a high extinction combined with an intrinsically faint supernova is compatible with our observations.