I’m looking at scaling relations between bulk properties (mass, temperature, luminosity, etc.) for galaxy clusters. It’s really hard to measure the mass of a cluster, so you need some sort of observable mass proxy. You need to really understand the mass-observable relation, and its scatter, to know how a cluster survey is sampling the mass distribution of clusters, and it’s that mass distribution of clusters that tells you about cosmology. I did this for mass and luminosity, and I have the first quantitative measure of the scatter.
I’m also looking at how selection criteria in cluster surveys bias the survey towards certain sub-populations of clusters (merging clusters, for instance).
(Oh dear, reading that I’m not sure it makes any sense.)