SIMBAD references

We investigate in this paper the error on the cluster redshift estimate as a function of: (1) the number of galaxy structures along the line of sight; (2) the morphology of the clusters (regular/substructures); (3) the nature of the observed galaxies (cD/normal galaxies); (4) the number of observed galaxies; and (5) the distance of the clusters. We find that if we use cD galaxies we have errors of less than 2% (at the 1-σ level) for the cluster distances except when the clusters are very complex. In those cases when we use five redshift measurements along the cluster lines of sight to compute the mean redshift, the mean redshift estimate error is less than 5% for the clusters closer than z=0.30. The same error of 5% for clusters in z=[0.30;0.60] requires about 12 redshifts measurements.