Selection effects, biases, and constraints in the Calán/Tololo supernova survey

We use Monte Carlo simulations of the Calán/Tololo photographic supernova survey to show that a simple model of the survey's selection effects accounts for the observed distributions of recession velocity, apparent magnitude, angular offset, and projected radial distance between the supernova and the host galaxy nucleus for this sample of Type Ia supernovae (SNe Ia). The model includes biases due to the flux-limited nature of the survey, the different light-curve morphologies displayed by different SNe Ia, and the difficulty of finding events projected near the central regions of the host galaxies. From these simulations we estimate the bias in the zero point and slope of the absolute magnitude-decline rate relation used in SNe Ia distance measurements. For an assumed intrinsic scatter of 0.15 mag about this relation, these selection effects decrease the zero point by 0.04 mag. The slope of the relation is not significantly biased. We conclude that despite selection effects in the survey, the shape and zero point of the relation determined from the Calán/Tololo sample are quite reliable. We estimate the degree of incompleteness of the survey as a function of decline rate and estimate a corrected luminosity function for SNe Ia in which the frequency of SNe appears to increase with decline rate (the fainter SNe are more common). Finally, we compute the integrated detection efficiency of the survey in order to infer the rate of SNe Ia from the 31 events found. For a value of H₀ = 65 km s^-1 Mpc^-1 we obtain a SN Ia rate of 0.21^+0.30_-0.13 SNu. This is in good agreement with the value 0.16 ± 0.05 SNe recently determined by Capellaro et al.