Lead: Madhura Killedar

Participants: Hareth Mahdi, Pascal Elahi, Jesus Vega

Aim: explore cluster weak lensing mass bias and uncertainty (B&U), dependence on several cluster lens properties, and the dependence of (some of) these results on the numerical code used. How does B&U depend on choice of overdensity and corresponding structure in the cluster (use one cluster to explore this). How does B&U vary with various cluster centres? Consider the benefit of including information from strong lensing and cluster member galaxies (talk to Matt Owers).

Input post-processing data:

• lensing maps: surface density (convergence), shear, reduced shear, deflection angles
• cluster properties: X-ray flux, dynamical state, most-massive substructure, substructure fraction

Comparisons

Alternative NFW fits to CL00019: Jesus Vega's results from previous study (see strong lensing)

Complementary mass estimates: Richness, Dynamical (Lyndsay Old), Caustics (Matt Owers), X-ray (Edwin Lau), SZ (Federico Sembolini)

Current state of progress:

• lensing maps made (by Hareth) for ~20 clusters, 5 Mpc across at 0.4", and few lines of sight (at z=0, shifted to higher-z)
• Bayesian mass measurement software (by Madhura)

Notes:

Earlier tests on mock NFW lenses showed a very strong degeneracy with c & M.  Degeneracy is/was due to:

• intrinsic shape noise
• a log-uniform mass prior
• structure in the cluster, so that it is not truly NFW
• too narrow field of view, thus lacking external shear (radius 1Mpc)
• bad resolution (~50 kpc)

I've been comparing to the incorrect overdensity (due to redshift shifting) so I'm not sure yet that I'm recovering the correct mass. I'll now make the overdensity choice flexible so that this can be accounted for.

PLOT BELOW: Uniform prior on mass (not log mass). Fitting routine assumes a higher level of noise sigma=0.15

Sample shear map of one of the nIFTy clusters 00228 (at z=0 but shifted to z=0.5):