Ghosh, Agniva2023-02-162023-02-162022-12https://hdl.handle.net/11299/252545University of Minnesota Ph.D. dissertation. December 2022. Major: Physics. Advisor: Liliya Williams. 1 computer file (PDF); xx, 140 pages.Over the last three decades, multiple imaged strong gravitational lensing systems have become an indispensable tool for determining the mass and mass-related properties of cosmic objects such as galaxies and clusters of galaxies. Lens reconstructions techniques play a crucial role in exploring the spatial organization of dark matter and its dynamics within galaxy clusters: Deviations from mass-follows-light, signatures of dark matter self-interactions, lensing effects of line-of-sight mass features and dark matter subhalos within the clusters using their lensing effects. In this work we have studied the free-form lens reconstruction method, Grale, that relies exclusively on the strongly lensed multiple image data. We used synthetic galaxy clusters to test Grale's efficiency and compatibility with future observations using 150-1000 strongly lensed multiple images. We found that with an increasing number of input images, Grale produces improved reconstructed mass distributions, with the fraction of the lens plane recovered at better than 10% accuracy increasing from 40-50% for ~150 images to 65% for ~1000 images.The goal of lensing reconstructions is to produce maps with properly quantified accuracy and precision. One of the ways to accomplish that is to compare reconstructions that use different lens inversion methods with differing modeling philosophies and assumptions leading to different reconstruction results, even if they use similar lensing data. The dispersion between various methods' reconstructions is probably the best estimation of systematic uncertainties in the cluster mass maps. We used Grale to obtain reconstructions of the real clusters Abell 370 and Abell 1689 using the latest available strong lensing data from the Hubble Space Telescope. We compared our results with existing reconstructions obtained by different parametric, free-form or hybrid methods. In our mass models for both the clusters, we found signatures of deviations from mass-follows-light scenarios which are also recovered by several other reconstruction methods. We conclude that on spatial scales of about ~100 kpc and above, most, and probably all mass reconstructions agree. The discrepancies between reconstructions start on scales below ~100 kpc.enClusters of GalaxiesCosmologyLens ReconstructionStrong Gravitational LensingThesis or Dissertation