HCI Small Grid Dithers

Small grid dithering (SGD), in JWST coronagraphic imaging, uses the fine steering mirror (FSM) to make very small offsets between exposures. The multiple datasets produced by this optional technique, at the cost of additional observing time, may be used in post-processing to produce higher fidelity reference PSF subtractions.

On this page

See also: NIRCam Small Grid Dithers, MIRI Coronagraphic Imaging Dithering, MIRI Coronagraphic Imaging Target AcquisitionNIRCam Coronagraphic Target Acquisition 

Target acquisition (TA) is an important factor that contributes to contrast performance in high-contrast imaging applications and typically depends on the specific instrument. For JWST, coronagraphic TAs rely on measuring the target's image centroid at a position away from the focal plane mask, then performing a small angle maneuver (SAM) to place the target behind a selected coronagraphic mask. Therefore, the accuracy of the TA is directly limited by the SAM accuracy, which is expected to be ~6–8 mas per axis (1-sigma radial) (see JWST Pointing Performance table for details).

For JWST high-contrast imaging, it is standard procedure to subtract a scaled image of an occulted point spread function (PSF) reference star to remove the speckles that remain in an occulted science target observation. However, the accuracy with which a science target and a subsequent PSF reference target can be placed behind an occulter is limited by the accuracy of the target acquisition procedure. Since the placement of a science target behind a given occulter may be slightly different from the placement of the PSF reference target, speckles in these observations may be slightly different, thus compromising the quality of the PSF reference subtraction from the science target observation.

In cases requiring the highest quality PSF matching, a technique called small grid dithering (SGD) can be invoked. The technique uses the fine steering mirror (FSM) to make a number of very small (10 mas for MIRI; 10–20 mas or NIRCam) offsets of the target in a grid pattern around the nominal TA position, with an observation executed at each step, thus creating a mini library of PSFs obtained at the same epoch as the science observation. Post-processing of the ensemble of observations can be used to model a more precise speckle pattern to use for the subtraction, at the expense of the additional observational overheads. An excellent article by LaJoie et al. (2016) is available for those who need to know the details of the technique, where it can be most effectively used, and the tradeoffs involved. A technical report by Soummer et al. (2014) also provides more information.

Since the purpose of using SGD is to obtain an improved reference PSF to subtract from the science observation, it is anticipated that most users will only apply SGD to the PSF reference target, and even then only in cases where the highest quality PSF subtractions are required for the science. Using the SGD technique on a science target is not disallowed, but you will need to explain in your proposal why you think it is beneficial for your use case. 

Specifying SGDs in APT 

See also: MIRI Coronagraphic Imaging APT TemplateNIRCam Coronagraphic Imaging APT Template

The expected use case for SGDs will be to (1) take the science observation(s), then (2) use an SGD dither pattern to obtain a set of slightly offset exposures on the PSF reference star. The ensemble of PSF star SGD exposures covers the expected region of possible misalignments from the science observation TA.

The SGD technique is expected to have the most benefit for MIRI 4QPM observations, owing to the requirement of very precise and accurate target placement relative to the apex of the mask. However, simulations have shown significant benefits for all MIRI and NIRCam coronagraphic modes (LaJoie et al. 2016). 

Words in bold are GUI menus/
panels or data software packages; 
bold italics are buttons in GUI
tools or package parameters.

MIRI and NIRCam instrument teams have created a set of pre-defined SGD options that can be selected in the appropriate APT observation template, using the pull-down menu for specifying dithers. Table 1 shows the available options for each mode, including the dither name used in APT.

Table 1. SGD dither option names in APT

Coronagraph/filterAPT dither type SGD options

Coronagraphic maskAPT dither pattern SGD options

From this table, it should be clear that the use of SGD comes with a price: depending on the grid chosen, the number of grid points (and hence observations) of the PSF reference star can be as high as 9 instead of one. The good news is that the FSM offsets are tiny compared with an FGS guider pixel, so no reacquisition of the guide star is needed. The FSM motions themselves take relatively little time, so it is mainly the additional observation time that is required. Because of this efficiency hit, you should only select the SGD technique in cases where the highest suppression of target star light is needed, but in those cases, significant improvements can be garnered (LaJoie et al. 2016).


Use of SGD data in the data processing

For observational sequences that include SGD data on the reference star, there are 2 possible processing algorithms to derive an improved PSF reference model for subtracting the residuals in the science image. They are KLIP (Karhunen-Lo`eve image projection) and LOCI (locally optimized combination of images). Initially, the pipeline will use the KLIP algorithm as part of standard processing. These algorithms use the small variations in the PSF speckle pattern from each SGD step to produce a model PSF that best matches the speckle pattern in the science target observation.

For details, refer to the publications by Lafreniere et al. (2007) and Soummer, Pueyo & Larkin (2012).


Lafrenière, D., et al., 
2007, ApJ, 660, 770
A New Algorithm for Point-spread Function Subtraction in High-Contrast Imaging: A demonstration with Angular Differential Imaging

LaJoie, C-P, et al. 2016, SPIE 9904 Space Telescopes and Instrumentation: Optical, Infrared, and Millimeter Wave
Small-grid dithers for the JWST coronagraphs

Soummer, R., Pueyo, L. Larkin, J.,  2012,  ApJL, 755, L28
Detection and Characterization of 
Exoplanets and Disks Using Projections on Karhunen-Loeve Eigenimages

Soummer, R., et al., 2014, SPIE 9143 JWST-STScI-004142
Small-Grid Dithering Strategy for Improved Coronagraphic Performance with JWST

Notable updates

  • Added offset sizes (in mas) for MIRI and NIRCam 

  • Minor corrections

  • Added references and updated text
Originally published