The impact of mass map truncation on strong lensing simulations

TL;DR

This paper investigates how different mass truncation schemes in strong lensing simulations can introduce artificial shear, potentially biasing cosmological measurements, but finds that extending the lens map reduces this effect.

Contribution

It identifies the impact of mass truncation on strong lensing simulations and quantifies how to minimize artificial shear bias.

Findings

Improper truncation introduces artificial shear.

Bias in H0 measurement can reach several percent.

Extending the lens map beyond 50 Einstein radii mitigates the effect.

Abstract

Strong gravitational lensing is a powerful tool to measure cosmological parameters and to study galaxy evolution mechanisms. However, quantitative strong lensing studies often require mock observations. To capture the full complexity of galaxies, the lensing galaxy is often drawn from high resolution, dark matter only or hydro-dynamical simulations. These have their own limitations, but the way we use them to emulate mock lensed systems may also introduce significant artefacts. In this work we identify and explore the specific impact of mass truncation on simulations of strong lenses by applying different truncation schemes to a fiducial density profile with conformal isodensity contours. Our main finding is that improper mass truncation can introduce undesired artificial shear. The amplitude of the spurious shear depends on the shape and size of the truncation area as well as on the slope and ellipticity of the lens density profile. Due to this effect, the value of H0 or the shear amplitude inferred by modelling those systems may be biased by several percents. However, we show that the effect becomes negligible provided that the lens projected map extends over at least 50 times the Einstein radius.