Mass calibration of the CODEX cluster sample using SPIDERS spectroscopy - I. The richness-mass relation

We use galaxy dynamical information to calibrate the richness-mass scaling relation of a sample of 428 galaxy clusters that are members of the CODEX sample with redshifts up to z similar to 0.7. These clusters were X-ray selected using the ROSAT All-Sky Survey (RASS) and then cross-matched to associated systems in the redMaPPer (the red sequence Matched-filter Probabilistic Percolation) catalogue from the Sloan Digital Sky Survey. The spectroscopic sample we analyse was obtained in the SPIDERS program and contains similar to 7800 red member galaxies. Adopting NFW mass and galaxy density profiles and a broad range of orbital anisotropy profiles, we use the Jeans equation to calculate halo masses. Modelling the scaling relation as lambda proportional to A(lambda) M-200c(B lambda) (1 + z)()lambda), we find the parameter constraints A(lambda) = 38.6(-4.1)(+3.1) +/- 3.9, B-lambda = 0.99(-0.07)(+0.06) +/- 0.04, and gamma(lambda) = -1.13(-0.34)(+0.32) +/- 0.49, where we present systematic uncertainties as a second component. We find good agreement with previously published mass trends with the exception of those from stacked weak lensing analyses. We note that although the lensing analyses failed to account for the Eddington bias, this is not enough to explain the differences. We suggest that differences in the levels of contamination between pure redMaPPer and RASS + redMaPPer samples could well contribute to these differences. The redshift trend we measure is more negative than but statistically consistent with previous results. We suggest that our measured redshift trend reflects a change in the cluster galaxy red sequence (RS) fraction with redshift, noting that the trend we measure is consistent with but somewhat stronger than an independently measured redshift trend in the RS fraction. We also examine the impact of a plausible model of correlated scatter in X-ray luminosity and optical richness, showing it has negligible impact on our results.