≡ Menu

Directly Imaging a Young ‘Jupiter’

Centauri Dreams continues to follow the fortunes of the Gemini Planet Imager with great interest, and I thank Horatio Trobinson for a recent note reminding me of the latest news from researchers at the Gemini South installation in Chile. The project organized as the Gemini Planet Imager Exoplanet Survey is a three-year effort designed to do not radial velocity or transit studies but actual imaging of young Jupiters and debris disks around nearby stars. Operating at near-infrared wavelengths, the GPI itself uses adaptive optics, a coronagraph, a calibration interferometer and an integral field spectrograph in its high-contrast imaging work.

Launched in late 2014, the GPIES survey has studied 160 targets out of a projected 600 in a series of observing runs, all the while battling unexpectedly bad weather in Chile. Despite all this, project leader Bruce Macintosh (Stanford University), the man behind the construction of GPI, has been able to announce the discovery of the young ‘Jupiter’ 51 Eridani b, working with researchers from almost forty institutions in North and South America. The discovery was confirmed by follow-up work with the W.M. Keck Observatory on Mauna Kea (Hawaii).


Image: Discovery image of 51 Eri b with the Gemini Planet Imager taken in the near-infrared light on December 18, 2014. The bright central star has been mostly removed by a hardware and software mask to enable the detection of the exoplanet one million times fainter. Credits: J. Rameau (UdeM) and C. Marois (NRC Herzberg).

This is a world with about twice the mass of Jupiter, and this news release from the Gemini Observatory is characterizing it as “the most Solar System-like planet ever directly imaged around another star.” The reasons are obvious: 51 Eridani b orbits at about 13 AU, putting it a bit past Saturn in our own Solar System. And although 51 Eridani b is some 100 light years away, Macintosh and colleagues have found a strong spectroscopic signature of methane.

“Many of the exoplanets astronomers have imaged before have atmospheres that look like very cool stars” says Macintosh. “This one looks like a planet.”

Indeed, and we have further evidence that this is a planet rather than a brown dwarf in chance alignment with the star in the form of a recent paper that analyzes the motion of 51 Eridani b and finds it consistent with a forty-year orbit. Moreover, we’re going to be learning a great deal more about this interesting object in years to come, as the paper explains:

Continued astrometric monitoring of 51 Eri b over the next few years should be sufficient to detect curvature in the orbit, further constraining the semimajor axis and inclination of the orbit, and placing the first constraints on the eccentricity. Absolute astrometric measurements of 51 Eri with GAIA (e.g., Perryman et al. 2014), in conjunction with monitoring of the relative astrometry of 51 Eri b, will enable a direct measurement of the mass of the planet. Combined with the well-constrained age of 51 Eri b, such a determination would provide insight into the evolutionary history of low-mass directly imaged extrasolar planets, and help distinguish between a hot-start or core accretion formation process for this planet.


Image: The Gemini Planet Imager utilizes an integral field spectrograph, an instrument capable of taking images at multiple wavelengths – or colors – of infrared light simultaneously, in order to search for young self-luminous planets around nearby stars. The left side of the animation shows the GPI images of the nearby star 51 Eridani in order of increasing wavelength from 1.5 to 1.8 microns. The images have been processed to suppress the light from 51 Eridani, revealing the exoplanet 51 Eridani b (indicated) which is approximately a million times fainter than the parent star. The bright regions to the left and right of the masked star are artifacts from the image processing algorithm, and can be distinguished from real astrophysical signals based on their brightness and position as a function of wavelength. The spectrum of 51 Eridani b, on the right side of the animation, shows how the brightness of the planet varies as a function of wavelength. If the atmosphere was entirely transmissive, the brightness would be approximately constant as a function of wavelength. This is not the case for 51 Eridani b, the atmosphere of which contains both water (H2O) and methane (CH4). Over the spectral range of this GPI dataset, water absorbs photons between 1.5 and 1.6 microns, and methane absorbs between 1.6 and 1.8 microns. This leads to a strong peak in the brightness of the exoplanet at 1.6 microns, the wavelength at which absorption by both water and methane is weakest. Credit: Robert De Rosa (UC Berkeley), Christian Marois (NRC Herzberg, University of Victoria).

Christian Marois (National Research Council of Canada) discusses the nature of the find:

“GPI is capable of dissecting the light of exoplanets in unprecedented detail so we can now characterize other worlds like never before. The planet is so faint and located so close to its star, that it is also the first directly imaged exoplanet to be fully consistent with Solar System-like planet formation models.”

As you would expect, 51 Eridani b is a young planet, young enough that the heat of its formation gives us a solid infrared signature, allowing its direct detection. In addition to being in an orbit that reminds us of the Solar System, the young world is probably the lowest-mass planet yet imaged, just as its atmospheric methane signature is the strongest yet detected. Given that the Gemini Planet Imager Exoplanet Survey is only a fraction of the way through its observing list, we can expect to find more planets in the target area within 300 light years of the Solar System.

The paper is Macintosh et al., “Discovery and spectroscopy of the young jovian planet 51 Eri b with the Gemini Planet Imager,” Science Vol. 350, No. 6256 (2 October 2015), pp. 64-67 (abstract). The follow-up paper is DeRosa et al., “Astrometric Confirmation and Preliminary Orbital Parameters of the Young Exoplanet 51 Eridani b with the Gemini Planet Imager,” accepted at The Astrophysical Journal Letters (preprint).


Comments on this entry are closed.

  • Alex Tolley November 19, 2015, 13:30

    This seems to me to be a good example of why you don’t want to have a view comparable to “looking out the window”. I would want my starship view screen to enhance the images of planets compared to their sun, with either false color or AR overlays to add useful information about conditions.

    I’m not seeing the image processing artifacts issue in the referenced paper. Is there some more explanation of the bright areas due to the image processing?

  • Michael November 19, 2015, 13:39

    I wonder if it is worth looking for the same methane signature in the imagery to spot other gas giants within the system, they should have similar signatures that could enable their detection above the general noise.

  • Horatio Trobinson November 19, 2015, 18:56


    Thank you, Sir, for your hard work and indefatigable dedication over the years.


  • Paul Gilster November 19, 2015, 22:29

    Very kind of you, sir! Much appreciated.

  • David Hall November 20, 2015, 1:27

    The people using SPHERE at the VLT have been doing some great work too: https://plus.google.com/113507009175485747967/posts/5D8WcdMJK5d . It’s certainly a fantastic time to be interested in space!

  • Mark Zambelli November 21, 2015, 6:39

    Thanks David… it’s good to see results from SPHERE as well as GPI.

  • Ashley Baldwin November 22, 2015, 9:08

    SPHERE is a great instrument and just the beginning too. A combination of coronagraphs, adaptive optics and deformable mirrors and polarimeters . Ultimately SPHERE is a prototype of the EPICS imager that will hopefully be a second generation instrument on the E-ELT. SPHERE operates on one of the 8.2m Unit telescopes of the VLT. In the visible spectrum and should be able to image Earth like terrestrial planets in the habitable zone of nearer stars and especially smaller M dwarfs . The polarimeters ( Zimpol on SPHERE and Empol on EPICS) just need photons , the bigger the telescope the better , in order to calculate the orbit of an exoplanet , it’s phase, and even presence of oceans and continents. Some polarisation can be caused by reflection within the telescope itself and is a source of error unless all the reflective surfaces of mirrors within the yelescope optical pathway are perfectly aligned with the axis of the telescope which if done cancels out the effect.

    The Imager technology combines extreme adaptive optics , coronagraphs, polarimetry and the big photon collecting capacity of a 40m mirror. There are however several major technological developments needed though first which is why we won’t see EPICS till 2030 at the earliest. Coronagraphy is the obvious one , both for space and ground based scopes and especially with segmented mirrors. Polarimeters are already nicely developed ( able to separate out an exoplanets light given the fact that starlight becomes polarised once reflected from the planet ) but adaptive optics for an ELT still have a way to go. The big key is the deformable mirrors both large and small that when linked with a wave front sensor are able to make corrective alterations to the incoming wavefront . This is done through a serious of actuators that sit behind the mirror and are attached to it. Small deformable mirrors are effectively MEMS, microelectromechical mirrors , effectively nanotechnology and one of the big breakthroughs of the last twenty years that with massive increase in processing power and CCD sensor arrays have as much as anything contributed to the massive advances in astronomy that have occurred over that time. For a 40m primary mirror a whole sequence of these mirrors are required in the telescopes optical train before the focal plane. “Multi conjugate adaptive optics “, MCAO, with different mirrors making corrections for atmospheric turbulence layers at different heights in the atmosphere between the telescope and the target . Extreme adaptive optic systems require their deformable mirrors to have an order of ten more actuators on them to make the small but critical alterations to the incoming wavefront. The time in which these corrections are made needs to be much quicker too especially for the first mirror in the sequence the so called “tip/tilt mirror that only moves in two planes yet makes the biggest “stroke” or difference of the entire corrective process of the Adaptive Optics system. So called ” Low order wavefront corrections ” which are discrepancies occurring outside of the telescope as opposed to “high order errors ” such as “speckles ” that occurred within the telescope due to imperfections in mirror surfaces and stray light from the incoming wavefront. This tip/tilt mirror is often called the “fast ” mirror as a consequence of its operation and for an ELT this speed is such that the processing power of the wavefront correcting computer needs to be significantly increased with much more sophisticated calculation algorithms. As with the Squrae Kilometer array , the processing power for this does not yet exist , but as Moore’s Law shows no sign of abating just yet and with 15 years between now and operation should see computer for EPICS hit their target . Ironically , 60% of all atmospheric turbulence occurs within 2kms of the ground and 40% within just 200m, which is why ground level adaptive optics look to address this with multiple deformable mirrors and low level laser “dispersion” lasers as part of Ground Layer Adaptive Optics , GLAO. The advantage with this system is that the multiple “guide stars” created by these lasers ( imitation stars close to the target object to act as a bright aiming point for the telescope home in on) allow the telescope to image not only to a good level ( the “Strehl ratio ” which is expressed as a percentage of the best possible viewing . A good AO system would be expected to give at least 60% up to 90% at best ) but also over a wide field of vision. Traditional one laser systems were limited to very small fields of vision. It was a combination of these factors in particular both difficulty and cost that ultimately led to hopes of 100m telescopes , OWLS ( Overwhelmingly Large Telescopes ) , proving a step too far and being downsized to more modest 25-50m ELTs..