Weighing dark matter haloes with gravitationally lensed supernovae

TL;DR

This study uses gravitational lensing of Type Ia supernovae in the GOODS survey to measure dark matter halo properties, providing an independent check against galaxy-galaxy lensing with results consistent within uncertainties.

Contribution

It introduces a novel method of using supernova brightness magnification to estimate dark matter halo scaling laws, complementing existing galaxy-galaxy lensing techniques.

Findings

Velocity dispersion normalisation is less than 190 km/s at 95% confidence.

Results are consistent with galaxy-galaxy lensing measurements.

Systematic effects, especially light curve fitting, significantly impact the results.

Abstract

High redshift Type Ia supernovae (SNe Ia) are likely to be gravitationally lensed by dark matter haloes of galaxies in the foreground. Since SNe Ia have very small dispersion after light curve shape and colour corrections, their brightness can be used to measure properties of the dark matter haloes via gravitational magnification. We use observations of galaxies and SNe Ia within the Great Observatories Origins Deep Survey (GOODS) to measure the relation between galaxy luminosity and dark matter halo mass. The relation we investigate is a scaling law between velocity dispersion and galaxy luminosity in the B-band. The best-fitting values to this relation (velocity dispersion normalisation and exponent) are 136 km/s and 0.27. We find the velocity dispersion normalisation to be less than 190 km/s at the 95 per cent confidence level. This method provides an independent cross-check of measurements of dark matter halo properties from galaxy-galaxy lensing studies. Our results agree with the galaxy-galaxy lensing results, but have much larger uncertainties. The GOODS sample of SNe Ia is relatively small (we include 24 SNe) and the results therefore depend on individual SNe Ia. We have investigated a number of potential systematic effects. Light curve fitting, which affects the inferred brightness of the SNe Ia, appears to be the most important one. Results obtained using different light curve fitting procedures differ at the 68.3 per cent confidence level.