NIRISS Target Acquisitions (TAs) position the source with sub-pixel accuracy on a specific part of the detector. A TA is required if a subarray is used and is strongly recommended when using full-frame readout with the SOSS and AMI observing modes.
On-board target acquisitions (TAs) position a source with sub-pixel accuracy on a specific location on the detector. TAs are supported for two of the observing modes for NIRISS: Single Object Slitless Spectroscopy (SOSS) and Aperture Masking Interferometry (AMI). While a TA is only required for those observing modes when a subarray is used, it is strongly recommended to perform a TA in full-frame readout mode as well, so as to ensure that the target is always placed on the same detector pixel. The steps in the NIRISS TA procedure are discussed in some detail below.
- The target is located. Three 64 × 64 pixel subarray exposures of the sky are taken. The individual images are offset from one another using small integer pixel offsets in both X and Y to allow for removal of bad pixels via dithering. The on-board TA software processes the images as needed (to realign the images, flag bad pixels, remove cosmic rays, and subtract the background level), and applies a centroiding algorithm to determine the target coordinates (see "Target Location Algorithm" below).
- The spacecraft is then moved to place the target at the nominal center of the TA subarray being used, which is listed in on-board tables.
- The target is re-imaged with a single (undithered) exposure. This is a so-called "TA verification" exposure.
- If the observing mode is AMI, the spacecraft is moved to the science aperture which is located on a separate (but nearby) subarray. This is done to avoid any persistence from the TA images. (This step is not performed for the SOSS observing mode, for which the spectra are projected onto a different location on the detector, so that the telescope does not have to be moved.)
A TA exposure sequence produces scientific data, which include the images of the target produced in steps 1 and 3. These data will be returned to the observer along with the science data.
Target location algorithm
To locate the target, the on-board software first passes a square checkbox of 3 × 3 pixels (0.196 × 0.196 arcsec2) over the TA subarray image and determines the intensity contained within the checkbox at each pixel in the subarray. The software then selects the checkbox with the maximum intensity and determines the flux-weighted centroid of the target using a 5 × 5 pixel window centered on that checkbox (see Figure 1 below).
Generally, the TA will be performed on the science target of the observation, though it is allowed to perform the TA on an object within 60 arcsec of the science target.
Target acquisition modes
TAs are performed with either the CLEARP or non-redundant mask (NRM) element in the pupil wheel and the F480M filter in the filter wheel. In general, the CLEARP element is used for faint TA targets while the NRM element, which limits throughput to 15% of the CLEARP element, is used for bright targets. There are additional considerations at play in selecting the best TA mode for a given target brightness for science observations in a given observing mode (i.e., SOSS vs. AMI), namely the sensitivity of the TA procedure to the presence of saturated pixels in a given TA mode, and differences between the CLEARP and NRM elements in terms of sampling of the telescope pupil. Table 1 lists the target magnitude ranges recommended for the various TA modes.
Table 1. Recommended magnitude ranges for NIRISS TA Modes
|TA Mode||PW Element|
|SOSSBRIGHT||NRM||3 ≤ M ≤ 6.1|
|SOSSFAINT||CLEARP||6.1 ≤ M ≤ 14.5|
|AMIBRIGHT||NRM||3 ≤ M ≤ 9.3|
|AMIFAINT||CLEARP||9.3 ≤ M ≤ 14.5|
It is recommended to choose a TA exposure time via the Exposure Time Calculator (ETC) that achieves a signal-to-noise ratio (SNR) of at least 30. The latter enables a centroid accuracy of ≤ 0.15 pixel. However, increasing the TA exposure time to get a higher SNR should be considered while planning observations for which accurate centroiding is deemed crucial. For example, the centroiding accuracy improves to about ≤ 0.10 pixel at SNR = 50 and to about ≤ 0.05 pixel at SNR = 100. More information on the use of the NIRISS TA ETC is given in the JWST ETC NIRISS Target Acquisition article.
Goudfrooij, P., 2016, Technical Report JWST-STScI-004784: "NIRISS Operations Concepts: Target Acquisitions".
Goudfrooij, P., 2017, Technical Report JWST-STScI-005934: "NIRISS Target Acquisition: the sensitivity of centroid accuracy to the presence of saturated pixels".