Abstract
The gravitational magnification and demagnification of Type Ia supernovae (SNe) modify their positions on the Hubble diagram, shifting the distance estimates from the underlying luminosity-distance relation. This can introduce a systematic uncertainty in the dark energy equation of state (EOS) estimated from SNe, although this systematic is expected to average away for sufficiently large data sets. Using mock SN samples over the redshift range 0 < z ⩽ 1.7, we quantify the lensing bias. We find that the bias on the dark energy EOS is less than half a percent for large data sets (≳2000 SNe). However, if highly magnified events (SNe deviating by more than 2.5 σ) are systematically removed from the analysis, the bias increases to ~0.8%. Given that the EOS parameters measured from such a sample have a 1 σ uncertainty of 10%, the systematic bias related to lensing in SN data out to z ∼ 1.7 can be safely ignored in future cosmological measurements.