Biased dark energy constraints from neglecting reduced shear in weak-lensing surveys

The weak gravitational lensing of distant galaxies by a large-scale structure is expected to become a powerful probe of dark energy. By measuring the ellipticities of large number of background galaxies, the subtle gravitational distortion called "cosmic shear" can be measured and used to constrain dark energy parameters. The observed galaxy ellipticities, however, are induced not by shear but by reduced shear, which also accounts for slight magnifications of the images. This distinction is negligible for present weak-lensing surveys, but it will become more important as we improve our ability to measure and understand small-angle cosmic shear modes. I calculate the discrepancy between shear and reduced shear in the context of power spectra and cross spectra, finding the difference could be as high as 10% on the smallest accessible angular scales. I estimate how this difference will bias dark energy parameters obtained from two weak-lensing methods: weak-lensing tomography and the shear ratio method known as offset-linear scaling. For weak-lensing tomography, ignoring the effects of reduced shear will cause future surveys considered by the Dark Energy Task Force to measure dark energy parameters that are biased by amounts comparable to their error bars. I advocate that reduced shear be properly accounted for in such surveys, and I provide a semi-analytic formula for doing so. Since reduced shear cross spectra do not follow an offset-linear scaling relation, the shear ratio method is similarly biased but with smaller significance.