Tolman Test from z = 0.1 to z = 5.5: Preliminary results challenge the expanding universe model

Abstract

We performed the Tolman surface-brightness test for the expansion of the universe using a large UV dataset of disk galaxies in a wide range of redshifts (from 0.03 to 5.7). We combined data for low-z galaxies from GALEX observations with those for high-z objects from HST UltraDeep Field images. Starting from the data in publicly- available GALEX and UDF catalogs, we created 6 samples of galaxies with observations in a rest-frame band centered at 141 nm and 5 with data from one centered on 225 nm. These bands correspond, respectively, to the FUV and NUV bands of GALEX for objects at z = 0.1. By maintaining the same rest-frame wave-band of all observations we greatly minimized the effects of k-correction and filter transformation. Since SB depends on the absolute magnitude, all galaxy samples were then matched for the absolute magnitude range (-17.7 < M(AB) < -19.0) and for mean absolute magnitude. We performed homogeneous measurements of the magnitude and half-light radius for all the galaxies in the 11 samples, obtaining the median UV surface brightness for each sample. We compared the data with two models: 1) The LCDM expanding universe model with the widely-accepted evolution of galaxy size R prop H(z)-1 and 2) a simple, Euclidean, non-expanding (ENE) model with the distance given by d=cz/H0. We found that the ENE model was a significantly better fit to the data than the LCDM model with galaxy size evolution. While the LCDM model provides a good fit to the HUDF data alone, there is a 1.2 magnitude difference in the SB predicted from the model for the GALEX data and observations, a difference at least 5 times larger than any statistical error. The ENE provides a good fit to all the data except the two points with z>4.