Detecting Horizon-Scale Gradients in Large-Scale Structure

We explore the possibility that there might be very long-wavelength spatial gradients in the large-scale structure of the Universe. This investigation is motivated by various observations, including a hemispherical power asymmetry in the cosmic microwave background and a potential anisotropy in the distribution of Type 1a supernovae. We use the Continuous-Function Estimator (CFE), a generalized estimator for the two-point correlation function, to simultaneously estimate the correlation function and its dependence on position within a survey volume. We first perform a null test on mock data to determine the expected variance of clustering across the sky. We then generate mock catalogs with injected gradients in galaxy clustering, and perform recovery tests to detect the direction and magnitude of the gradient. We compare our method to a standard two-point function estimation approach, dividing the catalog into patches and performing a least-squares fit to the clustering amplitudes in each patch. We find that we are able to detect more subtle gradients more precisely with the CFE approach than the standard technique. We discuss the implications of this work for constraining or measuring gradients in large-scale structure in current and future surveys.