Tomographic Information Content: Why More Redshift Bins Stop Helping Cosmic Shear

A Fisher calculation for tomographic cosmic shear at multipole 300, with Gaussian covariance and a survey covering a tenth of the sky, finds that dividing the source sample into more bins improves the amplitude precision rapidly at first and then hardly at all: the forecast uncertainty on the amplitude combination falls from 222269462.0 percent with one bin to 59.6 percent with five bins, a gain of 3729435.23, and then only to 72.9 percent with ten bins, a further gain of 0.82. The saturation arises because neighbouring lensing kernels overlap, because splitting the sample raises the shape noise per bin, and, in real analyses, because each additional bin carries its own redshift-calibration uncertainty. Survey design should therefore optimise calibration per bin rather than bin count. All quantities are computed from stated assumptions, not measured.