HCI Coronagraphic Sequences

JWST coronagraphic observations are normally executed together to minimize point spread function changes in science and PSF reference star observations.

On this page

See also: HCI APT Coronagraphic Sequence Examples, JWST High-Contrast Imaging Proposal Planning, NIRISS AMI Observations of Extrasolar Planets Around a Host StarMIRI and NIRCam Coronagraphy of the Beta Pictoris Debris DiskMIRI and NIRCam Coronagraphy of HR8799 b

Most applications of high-contrast imaging require careful subtraction of diffracted light from the bright point source, the "host" source, in order to see the science targets that are the nearby faint "companion" sources. Recognizing the importance of optical stability, the STScI Coronagraphy Working Group has recommended that coronagraphic observations be grouped into sequences, to ensure they execute together in time. The goal is to minimize changes in the optics that might alter the point spread function (PSF). This goal leads to the concept of coronagraphic sequences. Details of these sequences will depend on the science goals of your program, including whether single or multiple filters/coronagraphs are needed, and whether small or large offsets in position angle are called for. Some details and examples are provided below.

There are aspects and options of coronagraphic observations—such as target acquisitions (TAs), full frame astrometric (FFA), background observations and small grid dithers—that are specified within each observation template, but are not included explicitly in this article, which concentrates on organizing at the observation level.

For an example of an APT program that demonstrates the functionality of coronagraphic sequences, see HCI APT Coronagraphic Sequence Examples, which include instructions on downloading an accompanying demonstration proposal available in APT.

The standard coronagraphic sequence

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

The standard coronagraphic sequence comprises 3 observations for each science target:

  1. Observation of the science target in one spacecraft orientation
  2. Second observation of the science target in a different spacecraft orientation (e.g., 10° roll)
  3. Observation of a PSF reference star (to enable improved PSF subtraction in data processing)

The standard sequence uses a Sequence Observations ... Non-interruptible APT special requirement to ensure the observations execute together and in the order shown, and to minimize possible thermal variations differentially affecting the acquired PSFs. The goal is to obtain the lowest limiting contrast by minimizing the opportunity for changes in the JWST wavefront between the 3 observations. The obvious inference from this is that the science target and PSF reference target must be schedulable in the same visibility windows. Visibility can be verified using one of the Target Visibility tools.

In the standard sequence, the 2 observations of the science target are referred to as a roll dither, which is done to mitigate hot pixels. Removal of hot pixels may be important if you are looking for point-like companions, but less important if you are looking at extended structure. The roll dither strategy is recommended as the default. The 10° value shown above is provisional and assumes that the observation is scheduled at a time when the whole observatory can roll ±5° from its nominal position angle. In practice, the allowed offset from nominal varies from ±3.5° (7° total) to ±7° (14° total), as a function of solar elongation (longitude of the sun) at the time of the observation. Thus, forcing the roll offset toward the upper end of the range becomes very restrictive to schedule because the windows of time where larger roll offsets can be accommodated get very small. (See JWST Position Angles, Ranges, and Offsets for more information.)

The PSF reference star observation is used to calibrate wavefront errors and other uncertainties in the TA process, and to support PSF subtraction in pipeline data processing. According to STScI policy, PSF reference star observations are non-proprietary; any exceptions must be justified in the technical description of your proposal. The goals of this policy are to facilitate the community's deeper understanding of the coronagraphic PSFs and to build a library of PSFs for all to use.

This standard sequence—or a derivative of it, as described below for other cases—is recommended for all coronagraphic programs. If you want to depart from this strategy, you must provide an explanation in the technical justification section of your proposal. 

A single filter-occulter observation sequence

Here is a specific example of the standard 3-observation coronagraphic sequence using MIRI with a 4QPM coronagraphic mask and the F1065 filter:

  1. Science target, MIRI, 4QPM, F1065
  2. Science target, MIRI, 4QPM, F1065, 10° roll
  3. Reference PSF target, MIRI, 4QPM, F1065

The 10° roll between observations 1 and 2 is the "roll dither," calling for 2 observations of the science target with a 10° position angle offset between them. Note that a roll dither is not performed on the reference star because, in most cases, it would not significantly improve results. Nevertheless, the user has the option of adding a roll dither to the reference observation as well, but at a cost in efficiency, because a 10° roll counts as a 10° slew of the observatory, even though the target remains the same.

In APT, the relative roll angle between the 2 science observations is specified using an Aperture PA Offset ... APT special requirement, which specifies the desired angle between the 2 science observations (e.g., +10° or –10°, in our example). The sign specifies the direction of the roll (see JWST Position Angles, Ranges, and Offsets). At a sufficiently large separation from the bright host star, the two roll-dithered science images can be differenced in post-observation processing to directly obtain a PSF subtraction. However, it is assumed that in most cases the goal is to detect structure around the host; therefore, a separate PSF reference observation is needed for PSF subtraction.

The 3 observations in the standard sequence are also linked in APT using the Sequence Observations ... Non-interruptible special requirement to ensure that they are executed in order.

Standard sequence implementation for multiple filters and occulters

The on-orbit stability of JWST is not yet known. As a consequence, the current policy is conservative, sometimes requiring many science and PSF reference star exposures in the same time frame. If implemented strictly, this strategy can result in inefficient scheduling.

For example, consider the use of coronagraphy to characterize the atmosphere of a known exoplanet using multiple filters. If the rules are strictly applied, the result is multiple standard coronagraphic sequences as outlined above, one for each choice of filter and occulter. To identify alternative strategies, a study was conducted by STScI to quantify the overheads associated with 2 possible observing strategies:

  1. An optimal wavefront stability strategy, where standard sequences are consecutive in each filter to minimize the chance of any wavefront changes. This strategy increases the number of slews and rolls for the telescope. 

  2. An optimal efficiency strategy, where observations for a given target are organized—in each filter and occulter—to minimize the number of rolls and slews. This strategy increases the time between observation of a target in a given filter and the corresponding reference PSF star observation in the same filter, but it results in more efficient use of the observatory. 

The detailed study found that the optimal wavefront stability approach can require up to several hours more overhead per science target, depending on the number of filters and occulters being used. Hence, the optimal efficiency strategy is recommended, unless on-orbit operations reveals PSF variability on short times scales.

Building a coronagraphic sequence with multiple filters

Using the optimal efficiency strategy, here is how a user could implement multiple filters-occulters in a standard sequence. The following non-interruptible sequence of 6 observations is an example:

  1. Science target, MIRI, 4QPM, F1065
  2. Science target, MIRI, 4QPM, F1140
  3. Science target, MIRI, 4QPM, F1065, 10° roll
  4. Science target, MIRI, 4QPM, F1140, 10° roll
  5. Reference PSF target, MIRI, 4QPM, F1065
  6. Reference PSF target, MIRI, 4QPM, F1140 

This example involves 2 MIRI filters, but if NIRCam were the instrument, more filters could also be added to such a sequence.

This choice of ordering the observations—changing the filters-occulters before changing the target—minimizes slew overheads.

Using the APT Special Requirement tab, the user should link multi-filter observations with the Sequence Observations ... Non-interruptible special requirement in the same way as for the single filter case.

Linking too many observations together into a sequence can make the total execution time long, to the point that the observations cannot be scheduled. Therefore, you should seek to strike a balance between efficiency and the pragmatic aspects of scheduling observatory activities.

You may find it instructive to inspect the reports that are generated by running Smart Accounting in APT. These files provide a more detailed breakdown of where various overheads are being charged and will help you understand the tradeoffs in efficiency for the different models. See the articles on the APT Visit Planner and APT Smart Accounting for more information.

Alternate sequences for other science cases

Larger roll offset case

Depending on the science case, it may be necessary to obtain a larger roll offset on the science target than can be obtained in a single pair of roll-dithered observations. For example, the occulter in a given coronagraph may block part of the scene that the observer wants to see, and the ~10° roll is not sufficient. In this case, the only option is to break the Sequence Observations ... Non-interruptible requirement and schedule an observation at another time, when the larger position angle change can be accommodated.

In this case, you should structure 2 coronagraphic sequences, which are then linked together with special requirements in APT to accomplish the science. Here is a specific example:

  1. Science target, MIRI, 4QPM, F1065, initial PA
  2. Reference PSF target, MIRI, 4QPM, F1065, initial PA
    (Obs 1 and 2 Sequence Observations ... Non-interruptible
  3. Science target, MIRI, 4QPM, F1065, PA offset by 30°
  4. Reference PSF target, MIRI, 4QPM, F1065, PA offset by 30°
    (Obs 3 and 4 Sequence Observations ... Non-interruptible)

For this scenario, you have the option of whether or not to include the ~10° roll dither in the individual sequences or simply schedule 2 pairs of science and PSF reference star observations, as shown above. The Coronagraphic Visibility Tool can be used to assess the availability of multiple position angles, and estimate what the time separation will be. It is expected that each sequence will contain an observation of a relevant PSF reference star, since the PSF will likely change between the 2 epochs. This information can then be used with the special requirements in APT to request the needed observations. See HCI APT Instructions for details.

Shared reference survey case

If a set of science targets are clustered on the sky in close proximity, it may be possible to economize by observing more than one science target in sequence before observing one or more PSF reference targets. This is called the shared reference case because it breaks the pairing of individual science PSF observations in the normal coronagraphic sequence. An example might be a grouping of stars in a star-forming region:

  1. Science target #1, MIRI, 4QPM, F1065
  2. Science target #2, MIRI, 4QPM, F1065
  3. Science target #3, MIRI, 4QPM, F1065
  4. Science target #4, MIRI, 4QPM, F1065
  5. Science target #5, MIRI, 4QPM, F1065
  6. PSF Reference target, MIRI, 4QPM, F1065

All 6 targets need to be schedulable at the same time, and the Sequence Observations ... Non-interruptible special requirement would be placed on this entire set. The single PSF reference observation would get used for all 5 science targets. Some users may wish to schedule 2 PSF reference targets in order to guarantee at least one good PSF observation if the characteristics are uncertain. Or the observation of the PSF reference star could be scheduled in the middle of the sequence, at the user's discretion. The point is, there is no one-to-one pairing of science and PSF reference observations as is done in the standard sequence.

Self-referenced survey case

The self-referenced survey case is a slight variation of the shared reference case. Here, the user decides to observe a set of targets not knowing which may show surrounding structure and which may not. The assumption is that some targets will be for science, but others—the ones not showing surrounding structure—will be used for PSF reference observations. Unfortunately, the question of which are which cannot be determined until after the observations are obtained. So for instance:

  1. Science target #1, MIRI, 4QPM, F1065
  2. Science target #2, MIRI, 4QPM, F1065
  3. Science target #3, MIRI, 4QPM, F1065
  4. Science target #4, MIRI, 4QPM, F1065
  5. Science target #5, MIRI, 4QPM, F1065
  6. Science target #6, MIRI, 4QPM, F1065

All 6 targets need to be schedulable at the same time, and the Sequence ... Non-interruptible special requirement would be placed on this entire set. Since this is a special case, with no explicit PSF reference observations, 2 things need to happen: you need to carefully explain your assumptions in the technical justification portion of your proposal, and a box to this effect must be checked in APT. You also still need to select one of the targets to use as an initial PSF observation, to be used by the data processing system in its initial processing. See HCI APT Instructions for details.

Use of the small grid dithering technique

See also: HCI Small Grid Dithers, NIRCam Small Grid Dithers

The target acquisition accuracy with JWST cannot guarantee the same precise alignment of science targets and PSF reference targets on the coronagraphic occulters. For cases where the highest accuracy of PSF subtractions is desired, users can choose to apply a strategy of small grid dithers (SGDs).

SGDs can mitigate possible subpixel coronagraph misalignments between science and reference images. This strategy utilizes a defined set of subpixel dithered exposures to optimize coronagraphic PSF subtraction, which occurs in subsequent data processing. SGDs provide increased speckle diversity, which can be used to reconstruct an optimized, synthetic, reference PSF using one of the advanced PSF subtraction algorithms (Lafrenière et al. 2007; Soummer et al. 2012). While SGDs can be selected for observations of either a science target or a PSF reference star, it is envisioned that most users will apply it only to the PSF reference observation.

The SGD's subpixel dithers are executed with the fine steering mirror (FSM) under fine guidance. They are accurate to ~2–3 mas (1-σ/axis). Because they are executed using the FSM, their overhead is small. Nevertheless, the requested observation time increases as the number of dither points in the selected SGD pattern. The allowed (pre-defined) SGDs are readily available for MIRI and NIRCam by selecting the appropriate dither pattern in the relevant coronagraphic imaging template in APT. Simulations indicate that performance gains using the SGD strategy, compared to the standard undithered scenario, range from a factor of 2 for NIRCam to more than a factor of 10 for the MIRI 4QPM coronagraphs

Details can be found in Lajoie et al. 2016.


Lafrenière, D., Marois, C., Doyon, R., Nadeau, D., Artigau, E., 
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
Small-grid dithers for the JWST coronagraphs

Soummer, R., Pueyo, L. Larkin, J., 2012, ApJ, 755, L28

Detection and Characterization of 
Exoplanets and Disks Using Projections on Karhunen-Loeve Eigenimages

Latest updates

Originally published