2000ApJ...530..166H

The distribution of high-redshift Type Ia supernovae (SNe Ia) with respect to projected distance from the center of the host galaxy is studied and compared with the distribution of local SNe. The distribution of high-z SNe Ia is found to be similar to that of the local sample of SNe Ia discovered with CCDs, but different than that of the sample discovered photographically. This is shown to be due to the Shaw effect. These results have implications for the use of SNe Ia to determine cosmological parameters if the local sample of supernovae used to calibrate the light-curve decline relationships is drawn from a sample discovered photographically. A K-S test shows that the probability that the high-redshift SNe of the Supernova Cosmology Project are drawn from the same distribution as the low-redshift calibrators of Riess et al. is 0.1%. This is a potential problem because photographically discovered SNe are preferentially discovered farther away from the galaxy nucleus, where SNe show a lower scatter in absolute magnitude and are on average 0.3 mag fainter than SNe located closer to the center of their host galaxy. This raises questions about whether or not the calibration SNe sample the full range of parameters potentially present in high-redshift SNe Ia. The limited data available suggest that the calibration process is adequate; however, it would be preferable if high-redshift SNe and the low-redshift SNe used to calibrate them were drawn from the same sample, as subtle differences may be important. Data are also presented that suggest that the seeming anti-Malmquist trend noticed by Tammann et al. for SNe Ia in galaxies with Cepheid distances may be due to the location of the SNe in their host galaxies.