JWST’s First Directly Imaged Exoplanet

Title: JWST Early Release Science: High Contrast Imaging of the Exoplanet HIP 65426 b from 2−16 µm

Authors: A. L. Carter, S. Hinkley, J. Kammerer, et al.

First author’s institution: University of California, Santa Cruz

Status: Open access

This has been the summer of JWST. NASA’s newest major space telescope — a huge step up in size and capabilities from past observatories — finally began its science observations in June after a lengthy commissioning process. So far, we’ve seen stunning vistas of star forming nebulae and an incredible deep field showing lensed galaxies in JWST’s first images. Now, we’re starting to get exciting results from the Early Release Science (ERS) teams, who proposed creative ways to use the telescope in its first few months of observations, such as insight into some of the oldest galaxies we’ve seen.

Although JWST is somewhat known for its galaxy-hunting capabilities since mid-infrared observations allow us to see the distant redshifted parts of the Universe, it’s also going to be incredible for exoplanet science. One of the first “images” released was a transit spectrum of WASP-96b’s steam-filled atmosphere, and a recent ERS result made the first unambiguous detection of CO₂ in its atmosphere, too. 

JWST is also going to do wonders for direct imaging — the notoriously tricky way of detecting exoplanets, where we actually resolve light from the exoplanet itself instead of observing the host star. The NIRCam and MIRI instruments have coronagraphs, small optics that block the light from a bright host star so you can see the comparatively faint planets orbiting them. NIRISS also has the ability to do a cutting-edge technique called non-redundant aperture masking and NIRSpec and MIRI have a type of instrument called an Integral Field Unit (IFU), all of which can help directly detect planets. Lucky for us (or more accurately, thanks to the folks who planned this mission), it’s also easier to directly detect exoplanets in the infrared, making JWST extremely well-suited to this task.

The JWST Direct Imaging ERS team (ERS-1386) is now testing out these direct imaging capabilities, with the goal of determining how well these modes of observing are performing and coming up with advice for observers who want to use these modes in the future. The expectation is that JWST will be able to image planets smaller than Jupiter — which is a big deal! So far, we haven’t been able to spot a planet smaller than ~2 Jupiter masses from the ground.

In today’s bite, we share a hot-off-the-presses result from the JWST Direct Imaging ERS Team: the first directly imaged exoplanet observed with JWST, HIP 65426 b. This is also the first direct detection (ever!) of an exoplanet at wavelengths longer than 5 microns.

HIP 65426 b is a Super-Jupiter sized planet that was already known to us, originally discovered with ground-based observations around 2017. It’s part of the Lower Centaurus-Crux association, a grouping of stars that were all born near each other, including the relatively famous PDS 70 b. Moving group associations like this make it possible for us to estimate stellar ages — for example, HIP 65426 is around 14 million years old. This star and its planetary companion were chosen as a relatively easy target to test out JWST’s capabilities, and see what more we could learn about this planet! The team observed it with NIRCam, which covers 2 to 5 microns across five different filters, and MIRI, covering 11-16 microns over two filters — and it was detected in all seven filters, as seen below.

The “hamburger-like” shape in some of the wavelengths is an expected effect, just an artifact from the Lyot stop (part of the coronagraph). There aren’t actually multiple real sources there, unfortunately! In addition to the images, the team presented astrometry, photometry, model fits, and more for this observation.

Since JWST has such a large span of wavelength coverage in the infrared, they were able to constrain this exoplanet’s bolometric luminosity (its energy output across all wavelengths) like never before. No matter what model atmosphere they used, their result was the same thanks to the incredible data they had on hand! Their photometry of the planet was also incredibly precise — 7% precision for NIRCam and 16% for MIRI, compared to the ground-based observations of this star at 13-32% precision. The team also investigated the planet’s spectral energy distribution (SED) using the new photometry, alongside old measurements, shown below.

Using models of the planet’s thermal evolution and atmospheric models, they derived the mass (7.1 Jupiter masses), temperature (1282 K), and radius (1.45 Jupiter radii) of HIP 65426 b. The constraint on radius is ~3x more precise than before the JWST data! They also fit for the planet’s orbit using the package orbitize! and found a semi-major axis of 87 AU and inclination of 100 degrees, which agrees with but doesn’t significantly improve on past measurements.

To further quantify just how well JWST is doing things, they computed contrast curves for these observations, shown below. Contrast curves show the contrast (how faint of a planet you can detect, compared to its star, at 5 sigma confidence) versus the separation (how far the planet is from the star). So far, JWST appears to be outperforming expectations by about a factor of 10!

The team also estimated what kinds of planets JWST would be able to detect around this star. From their calculations, NIRCam can easily find a sub-Jupiter mass planet between 150-2000 AU from its star, and might be able to find something as small as 0.4 Jupiter masses. That’s still about 120x bigger than an Earth-like planet, but it’s significantly smaller than what we’ve been able to find with direct imaging before! MIRI is a little less sensitive, able to detect 1-2 Jupiter mass planets from 150-2000 AU. 

Since one of the goals of the ERS program is to make recommendations for best use of the instruments going forward, we’re going to describe a few of these technical pieces of advice here, too. If technical details of coronagraphs are not your jam, feel free to skip this next paragraph! 

One of the tricky parts of direct imaging is always aligning the coronagraph to perfectly cover the star, and they report that coronagraph alignment is still being perfected for NIRCam, whereas the MIRI procedure is further refined, accurate down to 0.1 pixels. The other notorious step of direct imaging is data processing, and this team tried both reference star differential imaging (RDI) and angular differential imaging (ADI) from different “rolls” of the spacecraft. They ended up using a combo of ADI and RDI for this paper, but suggest that future observations can just pick one or the other depending on their needs. RDI worked better in this case, but ADI may be better for wider separations. To do the data processing, they used spaceKLIP, an adaptation of pyKLIP, an image processing algorithm widely used for direct imaging work.

There is so much more to come with JWST observations, especially in the realm of direct imaging. For example, the authors suggest that for fainter M-type stars, JWST may be able to image even smaller planets than estimated in this investigation. They say in the paper, “It will be possible  to detect Uranus and Neptune mass objects beyond 100−200 AU, and Saturn mass objects beyond ∼10 AU [around M dwarfs].” They also mention that JWST’s incredibly precise infrared data might allow measurements of CH₄ and CO, providing a fascinating window into the complex chemistry of giant planet atmospheres. 

This result is clearly a harbinger (and a thrilling one at that!) of much more to come — the team even mentions other ERS results coming soon, including a different look into HIP 65426 b, observations of a circumstellar disk, and spectroscopy of another substellar object. Direct imaging, welcome to the era of JWST!

Astrobite edited by: Yoni Brande

Featured image credit: NASA/ESA/CSA, A Carter (UCSC), the JWST ERS 1386 team, and A. Pagan (STScI)