My previous entry described the process of generating simulated SZA data and analyzing it. The plots below give results from 3 different sets of ymaps and catalogs. Note that the whs, sw and bia sets have 15, 90 and 100 deg^2 of coverage respectively.
Improvements to the matching algorithm are very likely necessary. Ideas that come to mind: making cluster fit model elliptical rather than circular, merging catalog entries which are close enough that they are one object from the SZA point of view - but only if they are close in redshift...
Here are the first few match maps for each ymap set