Abstract
Weak gravitational lensing provides a way of estimating the mass and mass distribution of galaxy clusters at large radii. The clusters act as lenses, so that light emitted by background sources, is bent and distorted by the gravitational potential of the cluster. At large radii, the induced distortions in the image of the background galaxies, can be observed as a subtle systematic shift in their ellipticity. This is parametrized in the reduced gravitational shear, $g$. In this thesis, we perform a weak lensing analysis on three selected clusters, PSZ1 G311.65--18.48 (z=0.443), SDSS J1226+2152 (z=0.435), and SDSS J1723+3411 (z=0.443), all imaged by the Hubble Space Telescope. First, objects and their magnitudes are detected. Artifacts are filtered out by only allowing objects detected in multiple filters. By putting an upper an lower limit on their signal-to-noise ratio, and the sizes of the objects, stars and some cluster galaxies, are also rejected. Next, we remove the red sequence, formed by cluster galaxies in a color-magnitude diagram. Stars and are only affected by the point-spread-function (PSF) of the telescope optics, and most form a column in a magnitude versus radius diagram. They are used to determine the shape of the PSF. The final catalog, contains only background galaxies, and their PSF corrected reduced shear, is measured through a modified KSB method \citep{LuppinoKaiser1997}. The azimuthally averaged reduced shear, is found at different radii, and converted into a 2D surface mass distribution. We find that a NFW profile best fit our data, and obtain a mass estimate, $M_{200}$, through a best-fit analysis. The NFW profile depends on the parameters $r_{200}$ and $c_{200}$, the latter of which is very difficult to determine. While keeping $c_{200}$ fixed, and estimated from the relation found by \citet{Duffy2008}, we report the following results: For PSZ1 G311.65--18.48, we find $\sigma_v\approx 1066.4\,\si{km.s^{-1}}$, and $M_{200}\approx 1.1\times 10^{15}\, M_\odot h^{-1}$. This mass is about $1/2$ of that extrapolated from \citet{Dahle2016}, but well within our combined margins of error. For the SDSS clusters, we were only able to obtain an upper and lower limit of the mass. SDSS J1226+2152 is a part of a larger multi-cluster structure. This most likely biases our shear measurements to smaller values, and results in an unexpectedly small mass, $M_{200}\approx2.1\times\,10^{13}M_\odot h^{-1}$. We suspect this is the same reason why our measured cluster galaxy velocity dispersion, $\sigma_v\approx 323.0\,\si{km. s^{-1}}$, is so much smaller than that found by \citet{Bayliss2011}. For SDSS J1723+3411, we calculate a mass of $M_{200}\approx4.5\times10^{14}\,M_\odot h^{-1}$, and a velocity dispersion of $\sigma_v = 518.9\si{km.s^{-1}}$. Compared to the strong lensing analysis done by \citet{Kubo2010} on the same cluster, our $\sigma_v$ values are very similar. If we extrapolate their mass estimate with an SIS profile, our mass is $\approx0.6$ that of theirs.