NIRCam and MIRI Coronagraphy of HR8799 b
This example science program presents an application of the HCI Roadmap, showing how to create a cross-instrument observing program, using NIRCam and MIRI coronagraphy to observe the outermost exoplanet previously imaged around the young star HR 8799.
This example science program provides a walk-through of a JWST observing program using NIRCam and MIRI coronagraphic imaging, focusing on the overarching science goals from the GTO Program #1194, Characterization of the HR 8799 planetary system and planet search (PI: Charles Beichman) for context—albeit, focusing solely on planet b, the simplest one for JWST. This article discusses how to assess the feasibility and ultimately design a high-contrast imaging (HCI) observing program with JWST, and also links to further articles discussing how to use the JWST Exposure Time Calculator (ETC) to determine exposure times required to meet the science goals and how to set up the program in the JWST Astronomers Proposal Tool (APT) GUI.
The 4 directly-imaged planets around the young A-type star HR 8799 (HR 8799 bcde; Marois et al. 2008a, 2010a) provide a crucial reference point for understanding the physical properties and formation of young gas giants several times the mass of Jupiter. This example science program presents the case of HR 8799 b, the outermost planet of the HR 8799 planetary system, to demonstrate what JWST coronagraphy can bring to the knowledge of planetary atmospheres. The program will use NIRCam and MIRI multi-filter photometry in order to characterize the atmosphere of this benchmark exoplanet (particularly with filters/ bands which are not available from the ground). From this information it will be possible to infer such basic properties as total luminosity, effective temperature, and thus effective radius.
Walkthrough of the JWST HCI roadmap
Main article: HCI Roadmap
The HCI Roadmap guides readers through the process of designing a high-contrast imaging (HCI) observing program with JWST. Here we demonstrate this process, walking the user through the decisions made at each stage for this example science program.
Stage 1: Become familiar with the HCI capabilities and instrument-specific modes of JWST
The first stage in the process of planning any HCI program with JWST is to to familiarize oneself with the JWST HCI capabilities and terminology. Users should familiarize themselves with the various high-contrast imaging articles, in particular the observing modes and optics that enable HCI with JWST, as well as the primary performance metrics, operations, and recommended strategies that are particular to HCI. Users should also consult the Getting Started Guide, which provides more general instructions for planning JWST observations.
Stage 2: Evaluate the required observations with respect to the performance limits and capabilities of the HCI observing modes
(1) What is the wavelength range of interest and how does this influence (or limit) the choice of instrument(s), mask(s) and filter(s)?
NIRCam provides coronagraphic imaging from 1 to 5 µm, a key spectral region for exoplanet study. This program uses 6 of NIRCam's medium bandpass filters, F250M, F300M, F335M, F410M, F430M and F460M, to characterize the atmosphere of HR 8799b, measuring both the continuum and searching for signatures of CH4 (absorption gradient between ~3.5 and ~4.5 µm), CO (absorption features between ~4.5 and ~5.5 µm) and CO2 (strong absorption feature centered at ~3.7µm and gradient between ~4.5 and 5 µm).
MIRI will overcome the limited sensitivity of the largest ground-based observatories, extending planetary characterization to the mid-infrared thermal regime, where objects are too faint to be detected from the ground. For MIRI, the coronagraphic imaging filters are associated directly with each coronagraph and are not interchangeable—selecting the filter selects the coronagraph.This program performs MIRI coronagraphic observations in the F1065C, F1140C and F1550C 4QPM coronagraphic filters to enable: (1) the detection of the NH3 line at 10.65μm; (2) an off-line continuum measurement at 11.4 μm to calibrate the depth of the NH3 line; and (3) a more distant continuum point at 15.5 μm to determine the mid-IR continuum temperature, which traces the atmospheric temperature and cloud structure.
(2) Can choice of instrument(s) and mask-filter combination(s) achieve the required working angles?
Main article: HCI Inner Working Angle
HR8799 b is at a separation of approximately 1.724" from HR 8799. For MIRI, the 4QPMs are able to reach an inner working angle (IWA) of ∼ λ/D—corresponding to 0.34", 0.46" and 0.49" for the 10.65, 11.4 and 15.5 µm coronagraphs, respectively. For NIRCam, the 6 medium band filters are used to characterize HR 8799 b in conjunction with the long wavelength bar occulter (MASKLWB). The panchromatic characterization of planet b could be carried out with a round mask but for training purposes, this article focuses on using the bar. MASKLWB allows to reach an IWA in the range of 0.3"-0.89" which is suitable for planet b.
(3) Are the observations feasible given the contrast limits of the instrument(s)?
The anticipated a planet-to-star flux ratios are ~10-4 and ~10-5 at MIRI and NIRCam wavelengths, respectively. In order to achieve the best possible contrast, this program uses the small grid dither technique to build an optimal PSF reference for subtraction. This program will follow the recommended coronagraphic practice of observing the science target in 2 rolls and observing a nearby PSF reference star in an un-interruptible sequence.
(4) How important is the azimuthal coverage around the science target?
Careful attention is paid to the placement of the planet on each coronagraphic mask (particularly in relation to any instrumental obscurations, such as the boundaries in the MIRI 4QPM). To make these assessments the Coronagraph Visibility Tool (CVT) is used.
(5) Is it possible that the scientific goals can be achieved with non-coronagraphic PSF subtraction?
Main article: JWST Imaging
The angular sensitivity and contrast ratios required by this program can not be achieved through regular imaging; the source is observed with the coronagraphs for the combination of wavelength coverage and sensitivity.
Stage 3: Select a PSF calibration strategy
(1) What are the degrading factors that may limit the PSF calibration?
- wavefront drifts of the observatory
- PSF star color differences
- self-subtraction biases
- imperfect target acquisitions
- line-of-sight jitter and dynamic wavefront error.
All the above are valid points which cannot currently be modeled by the ETC. For the inner-most planets of this HR 8799 system (c, d, e) it would be necessary to address these points, possibly running most sophisticated simulation. For planet b at 1.7", none of those factors are prohibitive and a high SNR can be reached rather easily (within minutes) with each of the coronagraphic filters.
(2) Which observing technique(s) will be included in the PSF subtraction strategy?
The Referenced Differential Imaging (RDI) technique? — Required
The Angular Differential Imaging (ADI) technique? — Recommended
The Small Grid Dithering (SGD) technique? — Optional
In order to draw out the best contrast and achieve the smallest inner working angles of each coronagraph, a nearby and color- and flux-matched PSF reference source is observed using contemporaneous and identically executed observation sequences. This program employs the standard coronagraphic sequence for every instrument, mask and filter combination: an initial observation orientated at a desired nominal aperture position angle (i.e., one that maximizes the spatial coverage of the disk); followed by a second observation with an aperture position angle ~10° relative to the first observation; followed by an observation of the PSF reference star, to assist with suppression of the residuals in the coronagraphic image. All of the observations are linked in a non-interruptible sequence to ensure the PSF calibrator is observed close in time to the science target. Furthermore, to achieve the best possible contrast, the small grid dither technique is employed to build an optimal PSF reference for subtraction from each integration in the observations.
Stage 4: Assess target visibilities and allowed position angles
Main article: JWST Position Angles, Ranges, and Offsets
See also: JWST Observatory Coordinate System and Field of Regard, JWST Field of View, JWST Instrument Ideal Coordinate Systems, JWST Target Visibility Tools, JWST Coronagraphic Visibility Tool Help
(1) Familiarity with JWST position angles, coordinate systems, and pointing constraints.
(2) What viewing constraints are placed on the target?
Main article: JWST Target Viewing Constraints
Coronagraphic observations have additional constraints that go beyond target visibility, such as the placement and orientation of known sources on the coronagraphic masks, or offsets between multi-roll observations. Indeed, some preplanning may save significant time and possibly wasted effort downstream, in the event that certain desired angles or offsets are not available due to observatory level constraints.
Given the use of non-centro-symmetric occulters (i.e., the NIRCam bar occulter and MIRI 4QPMs), it is important to identify dates and aperture position angle (PA) constraints that maximize the distance of planet b with respect to the mask.
(3) What are the target visibilities and allowed position angles versus time?
Main article: JWST Coronagraphic Visibility Tool Help
To determine the target visibilities and available position angles versus time, the JWST Coronagraphic Visibility Tool (CVT)—a GUI-based tool developed specifically for pre-planning and strategizing coronagraphic observations with NIRCam and MIRI—is used. In addition to overall target visibility information, the CVT provides information on location of assumed companions relative to instrumental structures, such as occulting bars in NIRCam or boundaries in the MIRI 4QPM coronagraphs, as a function of time and shows how the instantaneous roll flexibility changes (from approximately ±3.5° to ±7° from nominal) over the visibility period.
After installing and opening the CVT, the CVT is used to determine the observability of HR 8799. In the control panel, the SIMBAD Target Resolver generates HR 8799's RA, Dec and Ecliptic coordinates, and then the Update Plot button calculates and displays the target’s visibility. The plot generated on the left shows the target's visibility windows, where the red highlights on the solar elongation line indicate the valid target's visibility windows and the blue tracks show the allowed position angles for the selected instrument and mask over those windows (see Figure 2 and 3).
(4) Do the observations require any restrictions on the orientation of the instrument field of view (FOV)/ detector being referenced?
In order to determine the ideal placement and orientation of planet b, the CVT is used to judge the planet's positions on the coronagraphic fields of view as a function of time. The CVT currently allows the placement of up to 3 companions relative to the primary target, so we define companion b with PA of 65° and Sep of 1.7" as well as companion c with PA of 330° and Sep of 1" because, at 1" separation, it is likely to be imaged "for free" with our strategy. For each of our planned observations, we select the corresponding instrument and mask (see Figure 2).
In order to maximize the distance of planet b from the coronagraphic masks, we will determine an orientation that places the companion as close as possible to perpendicular to the MASKLWB occulting bar. It turns out that b and c are difficult to image simultaneously and therefore our constraints focus on b solely for NIRCam taking both equally suitable APA range: 59º–75º and 211º–250º.
For MIRI, we will orient the disk at an angle of ~45° from the 4QPM axes, which corresponds to an APA of ~89º to ~104º. This placement guarantees that b (in quadrant 1) is at least 30º away from the 4QPM "vertical" axis (which has a 5º offset). By chance, this is a favorable position for planet c as well.
(5) How does the roll flexibility changes over the visibility period?
The maximum available roll stroke (or ApertureE PA Range: how much the observatory is allowed to rotate and vary the aperture PA at a given time as JWST must remain oriented with respect to the Sun) is always between ~7º and ~14º. For this simplified program, focusing only on planet b, we do not want to have too strict scheduling constraints and therefore we allow 7º as the minimum roll angle. At 1.7" it is enough to avoid self-subtraction of the planet as it represents a linear shift of 0.21" which is 3.4 pixels for NIRCam and ~2 FWHM at 3.35µm. For MIRI, it corresponds to 2 pixels.
(6) Do the goals call for a larger roll offset on the science target than can be obtained instantaneously in a single visibility period?
Main article: HCI Coronagraphic Sequences: Larger roll offset case
We do not find we require the use of a larger roll offset.
Stage 5: Use the Exposure Time Calculator (ETC) to determine observing parameters
Once target visibility is confirmed and a PSF calibration strategy adopted, we use the JWST Exposure Time Calculator (ETC) to determine the optimal exposure specifications for our program. Step-by-step ETC calculation instructions can be found in the Step-by-Step ETC Guide for NIRCam and MIRI Coronagraphy of HR8799 b article.
Stage 6: Select a suitable PSF calibrator
(1) What is the choice of PSF calibrator?
The PSF calibration strategy of this program requires appropriate PSF reference stars.
(a) For NIRCam, HD218261 is identified as a potential PSF reference star, as it is only 1.24º away from HR 8799 and has roughly the same spectral type (F star) and K-band magnitude.
(b) For MIRI, ups Peg (HD 220657) is identified, as it has been successfully used before for HR 8799. It is 4.7º away and brighter.
In GTO Prog. #1194, these 2 reference stars for NIRCam and MIRI are swapped with respect our program, which was elaborated for training purposes.
To identify the PSF reference targets, the following criteria is considered:
(i) Well-known: Is the target a known good PSF reference star?
Selecting a reference PSF source that has been previously observed coronagraphically and found to be single, is recommended. "Good references" are usually stars that are not astrophysically contaminated (i.e., without additional astrophysical signal from a debris disk or companion).
(ii) Schedulability: Do the visibility windows of the science target and PSF calibrator overlap at the time of the desired observation?
In order to observe the science target and PSF reference star in a contiguous, non-interruptible sequence, both objects need to be observable at the same time. With the restrictions on the observability HR 8799 known (from our previous work in the CVT), we now use the CVT to verify that both HD 218261 and HD 220657 are visible at the same time HR 8799, finding that there are 2 periods over which the targets' visibility windows overlap. Since they are less than 5º apart, it is the case.
(iii) Proximity: Is the PSF calibrator in relatively close proximity to the science target?
The JWST PSF is expected to be time variable, which has important consequences on the choice of PSF reference targets. In order to minimize the opportunity for changes in the JWST wavefront between science and PSF reference star observations, a reference star that is in close temporal and physical proximity to the science target should be chosen. Choosing a nearby reference star also minimizes the telescope overheads (by reducing slew time). We find that HD 218261 is an ideal match at a distance of only 1.24°, enabling a slew time of approximately ~600 s between the 2 sources (with HR 8799). With HD 220657 at 4.7º, the slew time is of the order of ~850 s, still very reasonable.
(iv) Avoidance of Binary: Is the PSF calibrator a single and unresolved source?
In order to ensure effective PSF subtraction, it is important to chose a star that is single and not astrophysically contaminated (i.e., without additional astrophysical signal from a debris disk or companion). HD 218261 and HD 220657 are known non-binary sources.
(v) Spectral Type: Does the PSF calibrator share the same spectral properties as the science target?
Spectral mismatch between a science target and its corresponding PSF reference star may lead to the generation of extra noise during the process of photometrically rescaling the reference, allowing for possible under- and over-subtraction of the PSF. The spectral type of HD 218261 is close to HR 8799 with a reasonable delta H-K and delta W1-W2, and therefore the sensitivity loss due to the spectral mismatch should be negligible. For MIRI, this criterion is less important but nevertheless we chose HD 220657 with a close spectral type.
(vi) Brightness: Is the PSF calibrator similar in magnitude to the science target?
It is recommended to use a reference PSF that is at least as bright than the science target, so that the noise isn't scaled during the flux rescaling process. For NIRCam, HD 218261 is 0.1 mag brighter than HR 8799. For MIRI it is even brighter with K = 3.04 but for this program we use the same readout parameters for all observations. Having a brighter reference star will ensure a good PSF subtraction with the same background level (important for MIRI) as long as saturation is not reached, which is true in our case.
(2) Returning to the previous ETC workbook, what are the final exposure parameters for the PSF reference star observations?
Stage 7: Finalize the observing strategy
Imaging HR 8799 b is quite an easy case and the recommended baseline strategy of 2 rolls and one reference star may appear to overshoot. However, this strategy minimizes the risks in case the reference star is not a good one (e.g, a close binary), the roll subtraction will allow to detect the planet. On the other hand, taking the reference star could enable the imaging of more planets than b, closer in as it would mitigate the degree of self-subtraction at small angular distances where the diversity provided by the roll strategy is limited.
Now that a series of technical decisions for the program have been made (such as our PSF calibration strategy, exposure specifications, etc.), an observing strategy that incorporates each of these components whilst also minimizing observing overheads and performance degradation is identified.
(1) What is the total number of observations required for the observing program?
- For our set of NIRCam MASKLWB observations, we will require a total of 3 observations:
(1) an observation of HR 8799 in the NIRCam LW filters at one spacecraft orientation;
(2) following a telescope roll, an observation of HR 8799 in the NIRCam LW filters at a second spacecraft orientation;
(3) an observation of the reference star HD 217261 in the NIRCam LW filters;
- Our MIRI 4QPM observations will require a set of 9 observations:(4, 5, 6) HR 8799 with F1065C, F1140C and F1550C at one spacecraft orientation; (7, 8, 9) HR 8799 with F1065C, F1140C and F1550C at a different spacecraft orientation; (10, 11, 12) The Reference Star HD 218261 with F1065C, F1140C and F1550C, respectively.
(2) How are the observations organized?
Main article: HCI Coronagraphic Sequences
Observing HR 8799 with the NIRCam LW Bar Coronagraph
Observing HR 8799 with 3 MIRI 4QPM Coronagraphs:
Slew to science target
Roll Observatory ~7 to 14°
Slew to PSF reference star
Slew to science target
Roll observatory ~7-14°
Slew to PSF reference star
This order was chosen in order to somewhat minimize overheads while also minimizing time between reference PSF observations and science target observations for each MASK/FILTER setup.
Stage 8: Prepare proposal in the Astronomers' Proposal Tool
Main article: JWST Astronomers Proposal Tool Overview
Currie, T., Burrows, A., Girard, J. H., et al. 2014 ApJ 795 2
Deep thermal infrared imaging of HR 8799 bcde: new atmospheric constraints and limits on a fifth planet
Marois, C., Macintosh, B., Barman, T., et al. 2008 Science 322 1348
Direct imaging of multiple planets orbiting the star HR 8799
Petit dit de la Roche, D. J. M., van den Ancker, M. E., Kissler-Patig, M., et al. 2020 MNRAS 491 1795
New constraints on the HR 8799 planetary system from mid-infrared direct imaging
Rajan, A., Barman, T., Soummer, R., et al. 2015 ApJL 809 2
Characterizing the atmospheres of the HR8799 planets with HST/WFC3
Wang, J., Graham J. R., Dawson R., et al. 2018 AJ 156 192
Dynamical constraints on the HR 8799 planets with GPI