http://arxiv.org/abs/1712.03444
Previous studies have shown that our velocity in the rest frame of galaxies at high redshift does not converge to that deduced from the CMB temperature-dipole anisotropy. In this work we determine the dipole in the galaxy catalogue derived from the Wide-field Infrared Survey Explorer (WISE) survey. After reducing star contamination to ~0.1% by rejecting sources with high apparent motion as well as those close to the Galactic plane, we eliminate low redshift sources in order to suppress the non-kinematic, clustering dipole. We remove sources near the super-galactic plane, and those which are within 1” of 2Mass Redshift Survey (2MRS) sources at z<0.03. We enforce cuts on the angular extent of the sources to preferentially select distant ones. As we progress along these steps, the dipole converges in direction towards that of the CMB, ending up within 5{\deg} of it. Its magnitude also progressively reduces as nearby structures are removed but stabilises at ~0.012, corresponding to a velocity >1000 km/s, if it is solely of kinematic origin. However, previous studies have shown that only ~70% of the velocity of the Local Group as inferred from the CMB dipole is due to sources at z<0.03. We examine the Dark Sky simulations to quantify the prevalence of such environments and find that <3.1% of Milky Way-like observers in a {\Lambda}CDM universe should observe the bulk flow (>240 km/s extending beyond z=0.03) that we do. We construct mock catalogues from the Dark Sky simulations in the neighbourhood of such peculiar observers in order to mimic our final galaxy selection, and quantify the residual clustering dipole. After subtracting this the remaining dipole is 0.0048+/-0.0022, corresponding to a velocity of 430+/-197 km/s which is consistent with the CMB. However the cause of such a large clustering dipole, the sources of which are at z>0.03, remains to be established.
M. Rameez, R. Mohayaee, S. Sarkar, et. al.
Tue, 12 Dec 17
59/78
Comments: Prepared for submission to MNRAS