Effects of asphericity and substructure on the determination of cluster mass with weak gravitational lensing

Weak gravitational lensing can be used to directly measure the mass along a line-of-sight without any dependence on the dynamical state of the mass, and thus can be used to measure the masses of clusters even if they are not relaxed. One common technique used to measure cluster masses is fitting azimuthally-averaged gravitational shear profiles with a spherical mass model. In this paper we quantify how asphericity and projected substructure in clusters can affect the virial mass and concentration measured with this technique by simulating weak lensing observations on 30 independent lines-of-sights through each of four high-resolution N-body cluster simulations. We find that the variations in the measured virial mass and concentration are of a size similar to the error expected in ideal weak lensing observations and are correlated, but that the virial mass and concentration of the mean shear profile agree well with that measured in three dimensional models of the clusters. The dominant effect causing the variations is the proximity of the line-of-sight to the major axis of the 3-D cluster mass distribution, with projected substructure only causing minor perturbations in the measured concentration. Finally we find that the best-fit "universal" CDM models used to fit the shear profiles over-predict the surface density of the clusters due to the cluster mass density falling off faster than the r^{-3} model assumption.