Change page style: 

NICI Flats

You are here

NICI Flat Fields require special consideration because of the Focal Plane Masks. Imperfections in the masks (scratches, dust specks), as well as the occulting spots themselves, of course are in the focal plane and therefore cause artifacts in science images. The imperfections are different for each mask, and the visible pattern changes every time the mask is moved. Therefore, the best flat field strategy must be chosen for each observing mode; there is no single strategy which gives good results for every application.

Note especially two important points regarding flat fields:

  • The Clear mask is a clear substrate, which is required to maintain correct focus, with its own imperfection pattern.
  • It is not possible to take a flat without a coronagraph mask, but with the same pattern of imperfections as a certain mask.

The "Flat Fields" group in the NICI OT library contains several types of GCAL flat templates. The first three sets in this library contain examples for Dual, Red, and Blue-channel flats which may be inserted into observing sequences and taken along with the science data with the masks and filters in exactly the same positions. For standard imaging (with no coronagraph mask) this gives the best results at the expense of higher observing overheads.

The library examples show the proper GCAL and exposure time settings for different filters. The exposure times are set so that the shutter-open exposures will have about 3000 ADU or 36000 electrons. Where possible, the coadd is set to10 so the total electrons is about 3.6e5 and the S/N ~ 600. For some filters the exposure times are so long that 10 coadds becomes prohibitive. Adjust Ncoadds for higher or lower S/N as desired.

Another strategy we have adopted for ASDI and ADI coronagraphic imaging is to take a set of flats through the Clear mask while dithering the mask to several positions. The images are then median-filtered to produce a master flat which represents the detector component of the flat field, with the Clear mask imperfections filtered out. Images with any of the occulting masks can then be divided by the master flat without introducing a strong artifact at the occulting spot position, while the field rotation helps remove the remaining mask features. While not perfect, this appears to give acceptable results for detecting faint point sources within the main target halo.

An example Clear Mask Dither Sequence is contained in the last group of the library. With several mask positions and long integration times designed to achieve 0.1% flat accuracy, the sequence requires more than 30 min to execute and therefore is run during the day. Such a sequence should not be entered into an observation, but should be requested in an Observing Note and/or by making a request to your contact scientist. 

Gemini Observatory Participants