Appendix A
PCA on SCUBA-2 data

This document has outlined the process of reducing POL-2 data and its reliance on PCA during the makemap process. There are two main reasons why running makemap with PCA is not the default method for all data observed using SCUBA-2:

it is very time consuming to run because of the faster scanning speed and consequent higher sample rate of non-POL-2 data, and
the optimal number of principal components to remove has not yet been determined.

All the relevant tools are still provided however, and so interested users may wish to try using this method and to compare the results.