Studying the peculiar velocity bulk flow in a sparse survey of Type Ia SNe

Studies of the peculiar velocity bulk flow based on different tools and data sets have been consistent so far in their estimation of the direction of the flow, which also happens to lie in close proximity to several features identified in the cosmic microwave background, providing motivation to use new compilations of Type Ia supernovae measurements to pinpoint it with better accuracy and up to higher redshift. Unfortunately, the peculiar velocity field estimated from the most recent Union2.1 compilation suffers from large individual errors, poor sky coverage and low redshift-volume-density. We show that as a result, any naive attempt to calculate the best-fitting bulk flow and its significance will be severely biased. Instead, we introduce an iterative method which calculates the amplitude and the scatter of the direction of the best-fitting bulk flow as deviants are successively removed and take into account the sparsity of the data when estimating the significance of the result. Using 200 supernovae up to a redshift of z = 0.2, we find that while the amplitude of the bulk flow is marginally consistent with the value expected in a Lambda cold dark matter universe given the large bias, the scatter of the direction is significantly low (at greater than or similar to 99.5 per cent C. L.) when compared to random simulations, supporting the quest for a cosmological origin.