We use statistical inference theory to explore the constraints from future galaxy weak lensing (cosmic shear) surveys combined with the current CMB constraints on cosmological parameters, focusing particularly on the running of the spectral index of the primordial scalar power spectrum, ${\alpha }_s.$ Recent papers have drawn attention to the possibility of measuring ${\alpha }_s$ by combining the CMB with galaxy clustering and/or the Lyman-$\alpha$ forest. Weak lensing combined with the CMB provides an alternative probe of the primordial power spectrum. We run a series of simulations with variable runnings and compare them to semianalytic nonlinear mappings to test their validity for our calculations. We find that a “reference” cosmic shear survey with $f_{\mathrm{sky}}=0.01\mathrm{}$ and $6.6\times {10}^8$ galaxies per steradian can reduce the uncertainty on $n_s$ and ${\alpha }_s$ by roughly a factor of 2 relative to the CMB alone. We investigate the effect of shear calibration biases on lensing by including the calibration factor as a parameter, and show that for our reference survey, the precision of cosmological parameter determination is only slightly degraded even if the amplitude calibration is uncertain by as much as 5%. We conclude that in the near future weak lensing surveys can supplement the CMB observations to constrain the primordial power spectrum.

• Received 25 August 2003

DOI:https://doi.org/10.1103/PhysRevD.69.083514