Dithering is discouraged for the JWST NIRISS aperture masking interferometry (AMI) observing mode. Persistence and other detector effects are likely to contaminate dithered data. Dithering may be requested if these effects do not compromise the science and if specialized data reduction methods are intolerant of missing data from bad pixels.

Main article: NIRISS Aperture Masking Interferometry
See also: NIRISS AMI Template APT Guide

Because of persistence (latent images), dithering is discouraged for the NIRISS aperture masking interferometry mode. Most AMI observations will likely not benefit from dithering. Dithering is not common practice in ground-based aperture masking. The usual two-point "nodding" scheme common to ground-based IR observations are used to subtract sky and thermal background from the data. This is not necessary for AMI data. The angular distance between typical nods is much larger than the PSF size, so image persistence effects in ground-based data are not severe. AMI's SUB80 array is of the order of the size of the AMI PSF wings. Dithering introduces more unknowns in pixel and instrument response. 

Given that AMI's strength is in minimizing effects of instrumental artifacts, observers typically prefer to calibrate out the fewest possible uncertainties (pixel responses, residual flat fielding errors, amplifier drift, and so on).  In addition, most ground-based experience is with pixel sizes much finer than the angular Nyquist sampling of the optical system.  This fine image plane sampling makes for good correction of bad pixels, and intra-pixel capacitance effects are less problematic. AMI data are barely Nyquist-sampled, so bad pixel correction becomes more difficult, and inter-pixel capacitance affects the analysis more.

The AMI template allows the elimination of flat fielding errors by repeated use of the same source position within the same pixel in every observation. Therefore the template always offers the same fixed pixel location(s), which means that fewer individual pixel peculiarities need to be calibrated out of the data. In general, dithering is not recommended.

However, dithering is available to AMI users who feel they need it. Dithering can increase the observable area around a target. Or it may turn out that image calibration quality exceeds expectations, or yet-to-be-tested forward modeling data analysis methods can account for relevant detector effects. If so, dithered AMI data might help to mitigate some of the effects listed below:

  • limited field of view of AMI's subarray readout
  • residual errors in the flat field
  • the effects of hot pixels or bad pixels
  • the effects of cosmic ray hits that are not identified in standard processing
  • inter-pixel capacitance variations

Since these effects occur on different spatial scales on the detector, dither patterns involving spatial offsets of many pixels as well as small fractions of a pixel may be required to treat them. The two approaches—to dither or not to dither—are likely to have advantages and disadvantages, particularly for imaging faint emission around bright objects.  With no dithers, a dominant source will remain in the same position on the detector and persistence will therefore not significantly affect the resulting data (except for the well-known Bohlin effect, a form of flux- and history-dependent non-linearity). However, bad pixels on the detector may suffer from poorer correction when compared to dithered observations. When dithering, though, the persistence from a bright central source may well affect image reconstruction. Early on-sky data should cast more light on whether dithering benefits some types of observations.

The dither patterns for the AMI mode are implemented as "primary" dithers that perform ~30-pixel offsets (with up to four positions within the 80 × 80 pixels science subarray) in conjunction with "secondary" dithers that are sub-pixel offsets (0.20–0.33 pixels) designed to obtain adequate pixel phase coverage. The offsets for the various dither step sizes and patterns are listed in Tables 1–4. Note that the dither offsets on this page are in the native detector coordinate system rather than the “Science” coordinate system. The dithers will be implemented correctly prior to launch.

Primary dithers

Table 1. Offsets for primary AMI dithers

Primary 1 DithersΔX (arcsec)ΔY (arcsec)

Subpixel dithers

Table 2. Offsets when not using subpixel dithers with the NRM

  Subpixel Positions = NONE
Dither Step #ΔX (arcsec)ΔY (arcsec)

Table 3. Offsets for a 5-step subpixel dither pattern with the NRM

  Subpixel Positions = 5
Dither Step #ΔX (arcsec)ΔY (arcsec)

Table 4. Offsets for a 9-step subpixel dither pattern with the NRM

  Subpixel Positions = 9
Dither Step #ΔX (arcsec)ΔY (arcsec)

Dither patterns for direct imaging with AMI mode

In addition to the main AMI dither patterns mentioned above, which are for observations with the non-redundant mask (NRM), the AMI mode also offers three additional dither patterns for optional direct imaging to mitigate effects of bad pixels, intra-pixel sensitivity variations, persistence and flat field errors. The spatial offsets of these dither patterns with respect to the first pointing (dithers=NONE) are tabulated below.

Table 5. Offsets for dithers in AMI direct imaging mode

Pattern NameStep #ΔX (pixels)ΔY (pixels)

ΔX (arcsec)

ΔY (arcsec)
Image Dithers = NONE

Image Dithers = 2


Image Dithers = 3



Image Dithers = 4





Bold italics style indicates words that are also parameters or buttons in software tools like the APT and ETC. Similarly, a bold style represents menu items and panels.



Latest updates
    Table 5 updated to show all dither patterns available for direct imaging with AMI mode.
    Also updated the heading and text above table 5.