Presentation #303.01 in the session Strong and Weak Gravitational Lensing & More — iPoster Session.
Measurements of weak gravitational lensing are limited by the variance in the unlensed shapes of source galaxies, labeled ‘shape noise’. Previous studies by Morales (2006) and Burgh-Day et al. (2015) proposed and expanded the idea of using asymmetries in a lensed galaxy’s velocity field to infer the component of shear 45° offset from the major axis, or γx. To constrain the remaining component along the major axis, γ+, Huff et al. (2013) proposed comparing the expected rotation speed for a galaxy via the Tully-Fisher relation to the line-of-sight rotation speed to infer the inclination and thus the unlensed axis ratio, which is affected by γ+. Wittman & Self (2021) used a Fisher information matrix analysis to forecast the precision of simultaneously inferring shear, inclination, and position angle from velocity and intensity fields given an idealized model of a galaxy as an infinitesimally thin disk. We apply this model to simulated galaxies in the Illustris TNG100 cosmological simulation to measure the injection of scatter by more realistic sources, which contain substructure such as arms, bars, and warps. These galaxies are unlensed, thus the inference of shear indicates the noise and bias emergent from these features. We find that γ+ is highly biased and prone to noise due to disks having a range of intrinsic axis ratios spanning 0.8-1. Inferred γx shows little bias and an rms of 0.03. This is the first realistic estimate of the noise floor for the velocity field lensing technique. We conclude that this technique will achieve significant gains in precision for measurements of γx, yet little gain for γ+.