Lensing cluster galaxies through model fitting. – AMALGAM

To measure the small lensing signals in the weak, strong and intermediate regime of flexion, with 3-4% accuracy in the largest sample of massive clusters, we had to minutely treat large amounts of archival images in order to control spurious biasing artefacts.
Shape measurements are based on the fit of light profiles of a large number of galaxies. We have improved and tested the SExtractor and PSFEx codes by participating to the GREAT3 international challenge of accurate lensing shear recovery. We have implemented in SExtractor the measurement of flexion but have not been able to thoroughly test it for production.
We had to pay particular attention to optimally exploit the limited amount of multiband photometry in our lens modelling code sl_fit, which can deliver mass reconstructions in both the strong and weak lensing regimes. It uses SExtractor shape catalogues augmented with a redshift probability distribution we obtained with a new technics that we called poorman-z and that makes the most of the available photometry by a nearest neighbour association from a reference photometric redshift catalogue.
We also have analysed cosmological numerical simulations with ray-tracing technics to predict the statistics of multipolar moments of the shear signal around clusters.

We have built a powerful processing pipeline of wide fields ground-based inhomogeneous data, which delivers reduced, astrometrically and photometrically calibrated of weak lensing quality. Our very successful participation to the GREAT3 challenge, we found that our shape measurement are among the best ones in the community and were able to implement the yet more difficult flexion measurement, but could not test it thoroughly though.
An optimal modelling of the multiband photometry allowed us to contain possible redshift depend biases in terms of recovered masses below the 3-5% level, as required by the sizable sample of 140 massive clusters we gathered from archival data.
Hence, we can constrain the scaling relation of cluster of galaxies, their degree of concentration as well as departs from spherical symmetry for which we also derived the first numerical and analytical predictions.

In this project we propose to develop and improve model fitting software that can simultaneously address the problem of shape (and henceforth shear) measurement with a controlled low level of systematics that meets our requirements for the present scientific project but also sets the way towards the near future requirements of surveys like LSST or Euclid.

1 paper on data processing of CFHT and Subaru data for weak lensing is submitted to A&A. 8 oral presentations in international conferences were given around this subject as well as a poster.
1 paper (MNRAS) reports how we performed in the GREAT3 challenge.
2 papers (Astronomy&Astrophysics) on the theoretical prediction of the multipolar moments spectrum of weak lensing signal around clusters were published along with one poster.

We propose to carry out at IAP an original short term gravitational lensing study of ~50 galaxy clusters present in the CFHT archive in order to probe the mass properties of the galaxies experiencing the tidal stripping and, more generally, specific environmental effects as compared to field objects. This problem will be addressed on a large range of scales as we intend to explore both the strong and weak lensing regimes. Strong lensing cross-sections in dense cluster environment as opposed to the field will probe the inner 10 kpc scale (1 effective radius) whereas shear AND flexion signals will allow us to extend the mass constraints out to 1-2 Mpc. Hence we will be able to shed light on the size of dark matter sub-halos as they enter and sink into the potential well of their host cluster halo.

This will be a particular application of the development and testing of two new lensing facilities at the center of this project. Two parallel development efforts that rely on the highly versatile, fast and broadly used SExtractor software will lead us to the construction of:
- A fast model-fitting code that will measure the shape of millions of faint background sources, and extract the second and third orders of the distortions produced by weak lensing, that are the shear and flexion signals.
- A robust, fast and automated model-fitting code that will be able to identify galaxy-scale strongly lensed features into wide field data (that need not be in clusters environment!)

Although the promising and widely unexplored short term application to lensing galaxies in clusters is the main scope of the project, we keep in mind the many prospects of extending this project to the future generation of large scale cosmological surveys like LSST and Euclid. Throughout our developments we propose to build techniques that are aimed to comply with the long-terms requirements of Euclid and LSST in terms of systematics control and understanding of the purity and completeness of the recovered lens samples.

All the data and code will be made public at the end of the project.