P12	January 1998

¹ Institut d'Astrophysique et de Géophysique - Université de Liège, Avenue de Cointe 5, 4000 Liège, Belgium

^* Chercheur Qualifié FNRS, Belgium

⁺ Based in part on observations collected at the European Southern Observatory (La Silla, Chile)

Abstract

     On the basis of a new sample of quasar optical polarization measurements, we have found that, in a region of the sky, the quasar polarization vectors are not randomly oriented as naturally expected, but appear concentrated around one preferential direction.

     In order to verify this surprising although preliminary result, we have compiled a large sample of quasar polarization measurements from the literature. With quite severe criteria to eliminate at best the contamination by our Galaxy, a sample of 170 quasars with good quality polarization measurements has been defined. Maps in redshift slices reveal a few regions where the polarization vectors are apparently aligned. To handle the problem more quantitatively, non-parametric 3D statistical tests were designed, as well as a method for visualizing spatially the results. The significance is evaluated through Monte-Carlo simulations.

     Applied to our sample of 170 polarized quasars, two different statistical tests provide evidence, with significance levels of 0.005 and 0.015 respectively, that the optical polarization vectors of quasars are not randomly distributed over the sky but are coherently oriented on very large spatial scales. This orientation effect appears spatially delimited in the 3D Universe, mainly occuring in a few groups of 10-20 objects. The polarization vectors of objects located along the same line of sight but at different redshifts do not appear accordingly aligned. Essentially for this reason, instrumental bias and contamination by interstellar polarization in our Galaxy are unlikely to be responsible for the observed effect.

     The very large scale at which this local orientation effect is observed indicates the presence of correlations in objects or fields on spatial scales ~ 1000 Mpc at redshifts z 1-2, suggesting an effect of cosmological importance. Several possible and testable interpretations are discussed.