Where is the matter in the Merging Cluster Abell 2218?

TL;DR

This paper models the mass distribution of galaxy cluster Abell 2218 using strong lensing data, revealing a bimodal dark matter structure and emphasizing the importance of large-scale halos over galaxy-sized substructures.

Contribution

The study provides a detailed parametric lensing model of Abell 2218, incorporating spectroscopic data and proposing new multiply imaged systems, highlighting the dominance of large-scale dark matter halos.

Findings

Mass distribution is bimodal, supported by light and velocity data.

Large-scale halos account for about 85% of the mass within 100".

Dark galaxy-sized substructures are constrained only when they perturb multiple images.

Abstract

We present a parametric strong lensing model of the cluster Abell 2218 based on HST ACS data. We constrain the lens model using 8 previously known multiply imaged systems, of which 7 have spectroscopically confirmed redshifts. In addition, we propose five candidate multiply imaged systems and estimate their redshifts using our mass model. The model parameters are optimized in the source plane by a bayesian Monte Carlo Markov Chain as implemented in the the publicly available software Lenstool. We find rms=0."12 for the scatter of the sources in the source plane, which translates into rms=1."49 between the predicted and measured image positions in the image plane. We find that the projected mass distribution of Abell 2218 is bimodal, which is supported by an analysis of the light distribution. We also find evidence for two structures in velocity space, separated by ~1000 km/s, corresponding to the two large scale dark matter clumps. We find that the lensing constraints can not be well reproduced using only dark matter halos associated with the cluster galaxies, but that the dark matter is required to be smoothly distributed in large scale halos. At 100" the enclosed projected mass is 3.8e14 solar masses. At that radius, the large scale halos contribute ~85% of the mass. We find that the model is not very sensitive to the fainter galaxy sized halos, unless they locally perturb a given multiply imaged system. Therefore, dark galaxy sized substructure can be reliably constrained only if it locally perturbs one of the systems. In an appendix we give a self-contained description of the parametric profile we use, the dual pseudo isothermal elliptical mass distribution (dPIE). This profile is a two component pseudo isothermal mass distribution (PIEMD) with both a core radii and a scale radii. (Abridged)