Analysis of luminosity distributions and shape parameters of strong gravitational lensing elliptical galaxies


Abstract in English

Luminosity profiles of galaxies acting as strong gravitational lenses can be tricky to study. Indeed, strong gravitational lensing images display several lensed components, both point-like and diffuse, around the lensing galaxy. Those objects limit the study of the galaxy luminosity to its inner parts. Therefore, the usual fitting methods perform rather badly on such images. Previous studies of strong lenses luminosity profiles using software such as GALFIT or IMFITFITS and various PSF-determining methods have resulted in discrepant results. The present work aims at investigating the causes of those discrepancies, as well as at designing more robust techniques for studying the morphology of early-type lensing galaxies with the ability to subtract a lensed signal from their luminosity profiles. We design a new method to independently measure each shape parameter, namely, the position angle, ellipticity, and half-light radius of the galaxy. Our half-light radius measurement method is based on an innovative scheme for computing isophotes that is well suited to measuring the morphological properties of gravitational lensing galaxies. Its robustness regarding various specific aspects of gravitational lensing image processing is analysed and tested against GALFIT. It is applied to a sample of systems from the CASTLES database. Simulations show that when restricted to small, inner parts of the lensing galaxy, the technique presented here is more trustworthy than GALFIT. It gives more robust results than GALFIT, which shows instabilities regarding the fitting region, the value of the Sersic index, and the signal-to-noise ratio. It is therefore better suited than GALFIT for gravitational lensing galaxies. It is also able to study lensing galaxies that are not much larger than the PSF. New values for the half-light radius of the objects in our sample are presented and compared to previous works.

Download