Validation and Application of the Bayesian Fourier Domain (BFD) Shear Measurement Method

Shear measurement for weak lensing studies relies on the ability to precisely and accurately measure galaxy shapes. Both moment- and model-based methods are sensitive to many forms of bias including model bias, noise bias, and selection bias and require calibration using simulations or "meta-calibration" from the data itself. The Bayesian Fourier Domain (BFD) shear measurement algorithm is rigorously derived to be unbiased for background-limited observations of low S/N galaxies, typical of the galaxies used for weak lensing analyses. Using validation simulations of a simplified galaxy population, we show that the BFD shear measurement algorithm is unbiased to within |m| < 1e-3, the target of next generation weak lensing surveys such as Euclid, LSST, and WFIRST. We also describe extension of the method to perform a joint measurement of shear and magnification, demonstrating the power of BFD to measure both weak lensing effects. Finally, we report on efforts to develop BFD for the Euclid and Dark Energy Surveys, including additional testing and application to real data.