[News] Potential Data Problems

[Emmanuel Bertin]

Just to throw in my 2 cents about the fringing matter (from the point of view of
CFH cameras).
I totally agree with Mark. From my limited experience, the fringe pattern is
"almost" perfectly stable from night to night, at least during a whole observing
run (a few weeks). Its intensity however varies wildly and does not scale very
well with the average background level. It is better to adjust the intensity on
each image using a robust procedure (that ignores stars and very large scale
gradients). When you do that (which is not the easiest, especially in crowded
fields) you should end up with a perfectly fringe-free image in 99% of the
cases. In the remaining 1%, well, you might have a secondary residual pattern.
This secondary pattern looks also quite stable in shape. It might be a part of
the O component mentioned by Mark, or some phase quadrature term. The thing is,
if you do a robust PCA analysis of your flat-fielded science images, you will
find these two patterns, and the less significant terms are compatible with
Poisson noise (+ local artifacts). This is what I observed on both WFI and
CFH12k images (not checked on MEGACAM yet).
Cheers,
					Emmanuel.