Jean-Baptiste Ruffio

Assistant Research Scientist, Astronomy & Astrophysics department, University of California, San Diego 

 Exploring exoplanets with direct imaging and novel high-resolution spectroscopy techniques.

As an astronomer interested in extra-solar planets, I strive to understand their diversity, evolution, and underlying formation mechanisms. My goals include the detection of the first solar system analogs, the study of their atmospheres, and pave the way for the search for biosignatures. I explore these new worlds through the largest direct-imaging surveys and high-resolution spectroscopy. Direct imaging allows us to take images of faint exoplanets that would otherwise be lost in the glare of their bright host star. High-resolution spectroscopy allows us to detect the distinct spectral features of molecules that are present in exoplanet atmospheres (like a molecular barcode). These detailed features are then used to measure their composition and better understand their climate. High-resolution spectroscopy also allows us to measure the wobble of these planets (ie, radial velocity) and therefore look for satellites, or exomoons, orbiting them. I have been pushing the frontiers of direct imaging and high-resolution spectroscopy by developing rigorous data analysis and statistical techniques.

Studying the atmospheres of the coldest exoplanets and brown dwarfs with JWST

One of the next milestones in exoplanet science is the detection of gas giants that are similar to the ones in our own solar system. We are currently able to study the atmospheres of planets that are either 10 times closer to their star (mostly transit spectroscopy), or several times larger and further away (direct imaging). Measuring the composition of the first true solar system analogs would bridge the gap between these two populations.  This is necessary to identify the trends in atmospheric composition that would allow us to better understand how they form. This might just be possible with JWST.

JWST presents a unique opportunity to observe directly imaged exoplanets and brown dwarfs at wavelengths >3 μm and thereby better constrain their composition and atmospheric physics. I have developed tools to enable the first moderate resolution spectra with NIRSpec (R~2,700; 2.9−5.2 μm) targeting substellar companions at high contrast, when the companion is much fainter and close to its star. The figure above features the faint T-dwarf companion HD 19467 B, which has clear spectral features of CO2, CH4, CO and H2O in its spectrum (Figure below). In order to mitigate systematics caused by spatial undersampling, we developed a framework to forward model the companion signal and host starlight directly in detector images. We demonstrated a sensitivity to companions that are 2x10-6 fainter than their stars at 1′′ (figure above). The achieved performance will enable detailed spectroscopy of most known directly imaged exoplanets, and even allow spectroscopy of sub-Jupiters up to ~1 Gyr at 10 pc and 1’’ separation. 

JWST-TST High Contrast: Achieving direct spectroscopy of faint substellar companions next to bright stars with the NIRSpec IFU - NASA/ADS ( 

Searching for exomoons

Io in front of Jupiter taken by the Cassini spacecraft. 

Credit: NASA/JPL/University of Arizona

The development of novel instruments combining the power of high-resolution spectroscopy and high-contrast imaging is enabling the first direct radial velocity measurements of planets. This will allow us to detect wobble of these planets caused by orbiting satellites. 

The figure below shows the future prospects for exomoon detections around the brown dwarf companion HR 7672 B. From ~1.5 nights of observations with the KPIC instrument, we have demonstrated a sensitivity to satellites with a mass ratio of 1-4% at separations similar to the Galilean moons (Dashed blue). The figure also includes simulated sensitivity for future instruments like Keck/KPIC II, Keck/HISPEC, and TMT/MODHIS (colored curves) assuming 6 nights of observations over a 25-day period. 

The mass ratios of the Galilean satellites are shown as black dots for comparison. Their possible scaled-up mass ratios accounting for the larger mass of the brown dwarf compared to Jupiter are shown as grey crosses. The Roche limit is computed for both a rigid and a fluid satellite shown as the inner and outer greyed region respectively.  The black dashed lines represent the astrometric sensitivity of VLTI/Gravity and the vertical gray scale bars represent direct imaging of satellites.

Detecting Exomoons from Radial Velocity Measurements of Self-luminous Planets: Application to Observations of HR 7672 B and Future Prospects - NASA/ADS ( 

Measuring the composition of exoplanets

The formation of directly imaged planets, whether from accretion of planetesimals or from the collapse of the circumstellar disk, remains poorly understood. Different formation models predict different atmospheric compositions, so spectroscopic characterization of exoplanets might be used to inform their formation pathway. 

Steps leading to the detection of water (H2O) and carbon monoxide (CO) in the atmospheres of the HR 8799 planets using the OSIRIS instrument at the Keck observatory and the new data reduction framework that I developed during my PhD. It allowed the study of planets that were so far inaccessible and enabled the measurements of the radial velocities of the planets themselves. I am developing a Python module for high-contrast imaging at high spectral resolution called BREADS (

Radial Velocity Measurements of HR 8799 b and c with Medium Resolution Spectroscopy - NASA/ADS (
Deep Exploration of the Planets HR 8799 b, c, and d with Moderate-resolution Spectroscopy - NASA/ADS ( 

(left) Using over a decade of Keck/OSIRIS observations, we have obtained the best moderate resolution spectra of the HR 8799 planets. (Right) While classical core accretion models predict super-stellar C/O, we showed that the four planets have similar C/O ratio and consistent with stellar.

High-resolution spectroscopy

A particularly exciting prospect for high-contrast imaging has been the recent development of dedicated high-resolution spectroscopic facilities. I have been working on the Keck planet imager and characterizer (KPIC; R=35000), which is the first instrument of its kind to come online. After two years on sky, our team has already opened new frontiers for exoplanet characterization. We have detected 20+ low-mass companions at high spectral resolution and measured the first spins of the HR 8799 planets. Exciting applications of high-resolution spectroscopy include the direct detection of new populations of planets, the characterization of their atmospheres, spin measurements and mapping their surface features with Doppler imaging, and even the search for exomoons from measurement of their radial velocities. It is a very active area of instrument development and innovative data analysis techniques. 

Looking for new planets

During my PhD at Stanford University with Bruce Macintosh, I designed a statistically motivated planet detection algorithm for high-contrast imaging based on forward modeling and matched filtering. The open-source package is publicly available in python ( I analyzed the data for one of the largest and most sensitive searches for young gas giant planets around 524 nearby stars (Gemini Planet Imager Exoplanet Survey) enabling its derivation of planet occurrence rates. We showed that giant planets are more common around higher-mass stars and form a distinct population from brown-dwarf companions suggesting that they form differently (Nielsen et al. 2019).  

Improving and Assessing Planet Sensitivity of the GPI Exoplanet Survey with a Forward Model Matched Filter - NASA/ADS ( 

Get in touch at jruffio at