Supernovae Ia have long been used to test cosmological models (and to win Nobel prizes!). Assuming that the supernovae are standard candles, one can compare their apparent magnitude, or flux, with the values predicted by theories. The scatter of the observed fluxes around the mean is considered just noise, arising from a combination of various effects, from peculiar velocities to calibration. This scatter is called intrinsic since it depends on the source. Since supernovae Ia’s lightcurves are supposed to behave in the same way regardless of environment or history, the intrinsic scatter is very likely independent of redshift.
In a recent paper, we show that there are interesting signals in this noise. In fact, the supernova light along its ride to our telescopes is subject to an extra scatter due to the random gravitational fluctuations it crosses along the way. The stronger the fluctuations, the larger the scatter. Since this extra scatter does depend on redshift, while the intrinsic one doesn’t, it is possible to extract it and compare it to what is predicted by models of cosmological perturbations. We find that although current data are not yet good enough to provide interesting results, future datasets with hundred of thousands of supernovae will be able to reveal the lensing scatter and to measure the growth of perturbations.