Sigma QUick Infrared Camera User Manual download pdf (Page 17)

WHIRC Data Reduction Manual Version 2.0, 2014 June 26 17

4. Set pixels below a certain value (say 0.05) to 1.0 to avoid excessively large

numbers in the flattened image using imreplace (Fig. 2.5). Alternatively, ensure

that a bad pixel map includes low-valued pixels.

Copy flats to a “fixed” extension: imcopy flat.140420.*.fits

flat.140420.*.%fits%fx.fits% and run fixpix on these images (Fig. 2.9).

Generate a rough image of the pupil ghost by dividing the processed Ks flat by

the J flat: imar flat.140420.fx.k / flat.140420.fx.j flatrat.kj

Knock down some of the bright spots in the ratio by running imreplace with a

lower limit set just above the highest value in the pupil ghost: imcopy flatrat.kj

flatrat.kj.edit; imreplace flatrat.kj.edit value=1 lower=1.2 upper=INDEF

Run fmedian (Fig. 2.10) to produce the smoothed image flatrat.kj.med

Subtract the median/mode from flatrat.kj.med to make the background ~ 0: imstat

flatrat.kj.med ; imar flatrat.kj.med - <stat> flatrat.kj.med

10.

Generate a null image: mkimage whircpupil.fits make 0 2 “2048 2048”

11.

Copy the central 800 × 800 pixels of the pupil image into the null image: imcopy

flatrat.kj.med[624:1424,624:1424] whircpupil[624:1424,624:1424]. This should

give you an image like Fig. 2.11b.

12.

Run rmpupil (Fig. 2.13) on the processed flat (e.g., flat.140420.fx.k) using the

pupil image generated by fmedian as the pupil input. The result should look like

Fig. 2.14b.

2.4.4 Fringes

For at least three of the WHIRC filters (Paβ, Paβ45, and H), Newton’s Rings fringes have

been seen in on-sky images. These almost certainly arise from OH emission lines in the

atmosphere and are thus not seen in dome flats (although they may appear in H band sky

flats). The optical configuration within WHIRC which produces these fringes is not

known, but they are textbook Newton’s Rings centered at the same location on the

detector in all filters where they are seen. The Paβ and Paβ45 fringes appear

monochromatic, whereas the H band fringes show amplitude modulation suggesting

several wavelengths are contributing (most likely the strong OH Q-branch lines in the H

band).

Unfortunately, unlike the pupil ghost, the fringe patterns differ for each filter in both

spacing (wavelength-dependent) and phase. In addition, fringe patterns seen in a given

filter appear to vary slightly in phase from one night to another. Therefore, one cannot

use a generic template to attempt removal of the fringe pattern, but must use the data

from the same night.

1 2 ... 12 13 14 15 16 17 18 19 20 21 22 ... 39 40

No comments

Sigma QUick Infrared Camera User Manual Page 17