[Previous] | [Session 104] | [Next]
M.C. Novicki, J. Tonry (Institute for Astronomy)
The traditional method for getting the light curve of a type Ia supernova involves subtracting a final template image after the supernova has faded from all of the other images to get a light curve. This method has some disadvantages: The light curve may only be constructed after the final image has been taken - which can be a year or more after the discovery of the supernova - and the signal to noise is then dependent on the quality of the template. We demonstrate a better method in which we subtract all N(N-1)/2 pairs of images and then measure the residual flux, which can be positive or negative. We do a least squares fit of the antisymmetric difference matrix to get a light curve, letting the zero point be a free parameter. There are several advantages to this method - there need be no wait to construct the light curve, so observations can be tailored to the fading supernova, and the light curve has lower noise because all observations contribute to eventual light curve points according to their quality. It is possible to get a better estimate of the true errors, and there is better use of all the available information. We will show the details of our fitting procedure, some examples of observed light curves, and a detailed monte-carlo comparison with the traditional method.
The author(s) of this abstract have provided an email address for comments about the abstract: mnovicki@ifa.hawaii.edu