Source Clustering in Cosmic Shear: Estimating the Correction from Foreground-Associated Sources

Source galaxies are not distributed randomly on the sky; they cluster, and they cluster with the foreground structure that lenses them, which perturbs a shear measurement at a level that precision analyses must estimate. Using an overlap model in which the correlated fraction of the source density declines with multipole, this paper estimates the fractional addition to the convergence power as 0.092 at multipole 100, 0.076 at multipole 300 and 0.042 at multipole 3000, corresponding at multipole 300 to a bias of 3.8 percent in the amplitude if the effect were ignored. The size is below the leading systematics but above the floor that forthcoming surveys can tolerate unmodelled, and its scale dependence, falling toward small scales, distinguishes it from a calibration error. All quantities are computed from stated assumptions, not measured.