Generalised model-independent characterisation of strong gravitational lenses VII: impact of source properties and higher-order lens properties on the local lens reconstruction

Abstract

We investigate the impact of higher-order gravitational lens properties and properties of the background source on our approach to directly infer local lens properties from observables in multiple images of strong gravitationally lensed extended, static background sources developed in papers I to VI. As the degeneracy between local lens and source properties only allows to determine relative local lens properties between the multiple image positions, we cannot distinguish common scalings and distortions caused by lensing from intrinsic source characteristics. The consequences of this degeneracy for lens modelling and our approach and ways to break it are detailed here. We also set up quantitative measures around the critical curve to find clear limits on the validity of the approximation that source properties are negligible to infer local lens properties at critical points. The impact of the source on the local lens properties depends on the reduced shear at the image position and the amplitude and orientation of the source ellipticity, as we derive in this paper. Similarly, we investigate the role of third-order lens properties (flexion), in two galaxy-cluster simulations and in the Lenstoolreconstruction of the galaxy-cluster lens CL0024. In all three cases, we find that flexion is negligible in over 90% of all pixels of the lensing region for our current imprecision of local lens properties of about 10%. Decreasing the imprecision to 2%, higher-order terms start to play a role, especially in regions with shear components close to zero.

Generalised model-independent characterisation of strong gravitational lenses VIII. automated multi-band feature detection to constrain local lens properties

“…Most importantly, we found that, for brightness features within multiple images or for featureless multiple images generated by a galaxy-cluster scale strong gravitational lens, the fact that higher-order lensing effects like flexion are negligible in most cases (Wagner (2022)), puts the maximum intensity position and the centre of light only 1-2 pixel apart from each other. As a consequence, basing lens reconstructions on either of them has a small impact on the lens reconstruction.…”

“…Fourth, we would expect star-forming regions to be of approximately circular shape which are distorted into ellipses to first order by the gravitational lensing effect. Assuming that these features are far enough away from the critical curve not to be extremely distorted in the shape of arcs (see Wagner (2022) for a detailed analysis in galaxy-cluster scale gravitational lenses), the semi-major and semi-minor axes of the second order moment of an object's brightness distribu-tion also contribute to evaluate the usefulness of an object. Highly elongated arcs can be excluded to be star-forming regions and only included in the feature set if such arcs are clearly identified across several multiple images as a more complex feature.…”

“…Caminha et al (2016), for instance, clearly use the peak position and thus assume all noise was eliminated. On the scale of star-forming regions as features, Wagner (2022) showed that any physically caused offset between the peak and the centre of light, for instance due to higher-order lensing effects like flexion, is unlikely to occur and be resolved. This corroborates our third selection criterion for robust features of Section 3.2.…”