Strong lensing in Abell 1703:: constraints on the slope of the inner dark matter distribution

Properties of dark matter haloes can be probed observationally and numerically, and comparing both approaches provides ways to constrain cosmological models. When it comes to the inner part of galaxy cluster scale haloes, interaction between the baryonic and the dark matter component is an important issue that is far from being fully understood. With this work, we aim to initiate a program coupling observational and numerical studies to probe the inner part of galaxy clusters. In this article, we apply strong lensing techniques on Abell 1703, a massive X-ray luminous galaxy cluster at z = 0.28. Our analysis is based on imaging data from both the space and ground in 8 bands, complemented by a spectroscopic survey. Abell 1703 is rather circular from the general shape of its multiply imaged systems and is dominated by a giant elliptical cD galaxy in its centre. This cluster exhibits a remarkable bright central ring formed by 4 images at z(spec) = 0.888 only 5-13 '' away from the cD centre. This unique feature offers a rare lensing constrain for probing the central mass distribution. The stellar contribution from the cD galaxy (similar to 1.25 x 10(12) M-circle dot within 30 kpc) is accounted for in our parametric mass modelling, and the underlying smooth dark matter component distribution is described using a generalized NFW profile parametrized with a central logarithmic slope alpha. The rms of our mass model in the image plane is equal to 1.4 ''. We find that within the range where observational constraints are present (from similar to 20 kpc to similar to 210 kpc), a is equal to 1.09(-0.11)(+0.05) (3 sigma confidence level). The concentration parameter is equal to c(200) similar to 3.5, and the scale radius is constrained to be larger than the region where observational constraints are available (r(s) = 730(-75)(+15) kpc). The 2D mass is equal to M(210 kpc) = 2.4 x 10(14) M-circle dot. However, we cannot draw any conclusions on cosmological models at this point since we lack results from realistic numerical simulations containing baryons to make a proper comparison. We advocate the need for a large sample of well observed (and well constrained) and simulated unimodal relaxed galaxy clusters in order to make reliable comparisons and to potentially provide a test of cosmological models.