NICMOS Instrument Handbook for Cycle 11

Introduction

---

Multiple exposures with small offsets in the pointing of the telescope are usually necessary or recommended for NICMOS observations. We distinguish three particular circumstances which may require small offsets:

• Dithering to permit the removal of dead or non-calibrated (i.e., non-correctable) pixels on the detectors, and detector's non-uniformities (i.e. sensitivity variations),
• Dithering or chopping to measure the background associated with an astronomical source,
• Mosaicing to map a source larger than a single detector field of view.

The techniques described in this chapter may be used to accomplish any one or any combination of these goals.

Experience with NICMOS has shown that the background is considerably fainter than was expected prior to deployment (see Chapter 4). The background appears spatially uniform (variations no larger than a few percent across the NIC3 field of view) and does not vary much with time (variations of less than 5% on orbit timescales). The description of the thermal background in Chapter 4, Chapter 9 and the Exposure Time Calculator provide a basis for estimating the relative contributions of source and background. It is strongly advised that provision for direct measurement of the background be included in proposals whenever the background is significant relative to the source(s) of interest. The frequency of such measurements should be about once per orbit, and more frequent measurements should be planned when the background must be measured to high accuracy.

All observations at wavelengths longward of 1.7 µm should consider the need for background measurements.

 

Background images are obtained by offsetting the telescope from the target to point to an "empty" region of the sky. The ability to routinely offset the telescope pointing is a fundamental operational requirement for NICMOS. Starting in Cycle 9, HST programs will use a standard pattern syntax, which will replace the old pattern optional parameters, and the even older scan parameters form. The new syntax allows multiple observations (including different filter observations) to be made at each point in the pattern, if desired. Observers should check the "Phase II Proposal Instructions" for the exact syntax of the special requirement that invokes the pattern, and the "pattern parameter form" that describes the motion. For simplicity, a set of pre-defined observing patterns has been built; the exposures taken under them are combined into one or more associations. A pattern, then, is a set of images of the same astronomical target obtained at pointings offset from each other, e.g. for the purpose of removing bad or grot affected pixels from the combined image, mapping an extended target, or for creating background images. The associations of exposures are created for the purpose of simultaneously processing all the images (through a given filter) from a single pattern. Dithered images can thus be easily reassembled into a single image with the effects of minimizing bad pixels, or images taken in the long wavelength regime can be corrected for the thermal contribution, or observations of extended targets combined into a single large map.

Two distinct types of telescope motion are defined:

• Dither: Individual motions are limited to no more than 40 arcsec. These are intended to be used to perform small dithers, to measure backgrounds for compact sources, and to accomplish sequences of overlapping exposures for the construction of mosaics. Such sequences will be assembled into a single final image by the calibration pipeline.
• Chop: Motions up to 1440 arcsec are permitted. These are intended for the measurement of the background at one or more locations significantly removed from the target pointing. Each non-contiguous background pointing will be assembled into its own final image in addition to the target pointing by the calibration pipeline.

Telescope motions involve overheads for physically moving the telescope and, if necessary, for re-acquiring the guide stars. Therefore, significant time overheads may be incurred by observations which need background subtraction or propose to map extended regions of the sky. A careful estimate of the overheads associated with a specific observation or set of observations is necessary to evaluate the number of orbits required (see Chapter 10).