Prediction of galaxy ellipticities and reduction of shape noise in cosmic shear measurements

Abstract The intrinsic scatter in the ellipticities of galaxies about the mean shape, known as ‘shape noise’, is the most important source of noise in weak lensing shear measurements. Several approaches for reducing shape noise have recently been put forward, using information beyond photometry, such as radio polarization and optical spectroscopy. Here we investigate how well the intrinsic ellipticities of galaxies can be predicted using other exclusively photometric parameters. These (such as galaxy colours) are already available in the data and do not necessitate additional, often expensive observations. We apply a regression technique, generalized additive models to the publicly released galaxy property data from CFHTLenS. We find that the individual galaxy ellipticities can be predicted from other photometric parameters to better precision than the scatter about the mean ellipticity. The ellipticity contribution to the shear can apparently therefore be measured to higher precision, comparable to using a larger sample of galaxies. Using only parameters unaffected by lensing (e.g. surface brightness, colour), our best-fitting model leads to a gain (for the ellipticity contribution only) equivalent to having 12 per cent more galaxies in the sample. Allowing parameters correlated with lensing increases the apparent gains (we find 52.5 per cent), but these would likely be negated by correlations between the predictor and measured shear. We caution that the ultimate usefulness of this method will depend on careful treatment of the effect of the point spread function and input parameter measurement. This is to avoid information on the ellipticity contaminating the estimated ellipticity through observational effects rather than true physical correlations.