Julien speaking to a crowd looking away from camera downward.

Julien de Wit

Julien’s primary interest and expertise lie in the field of data science where Math and Science are brought together to make sense of newly accessible pieces of Reality! Over the past five years, he has developed and applied new analysis techniques to map exoplanet atmospheres, study the radiative and tidal planet-star interactions in eccentric planetary systems, and constrain the atmospheric properties and mass of exoplanets solely from transmission spectroscopy.

Assistant Professor - MIT

Contact Information:

Email: jdewit@mit.edu

Office: 77 Massachusetts Ave, 54-1724, Cambridge, MA 02139

Screen Shot 2020-03-23 at 1.07.49 PM.png

Featured Research Figures


This figure is the first figure of the first paper I ever published. Many thanks to Professor Michael Gillon, and to Professor Sara Seager for the introduction to both the field and Prof. Gillon. This paper introduces a framework to derive maps of extra-solar planets as robustly as possible, while accounting for degeneracies between the map parameters and, notably orbital parameters (see next Figure). The figure above specifically introduces the various scanning processes that an eclipsed planet is going through over the course of its orbit. The green dotted lines indicate the scanning processes during the exoplanet occultation ingress/egress. The red dashed line indicates the scanning process resulting from its rotation. The component labeled “combined” shows the specific grid generated by these three scanning processes. For more info, click here.

Fig 2, 1.png

This figure is from the same paper as introduced above and highlight the aforementioned degeneracy between the map parameters and the system parameters. Left: the 2D confidence interval of the stellar density vs the product of the orbital eccentricity sqrt and the sinus of the periastron argument is shown to substantially change depending on the assumption made regarding the brightness distribution of the planet seen observed in eclipse. Right: Deviations in the deviations in occultation ingress/egress from a median fit are highlighted through perturbation of four system parameters in order to contextualize the aforementioned degeneracy. The system parameters can be used to compensate for an anomalous occultation that emerges from, e.g., a non-uniformly-bright exoplanet leading to biased estimates of the system parameters. For more info, click here.

Fig 2, 2.png
Fig 3, 1.png

First map of an exoplanet made in the visible. Many thanks to Professor Brice-Olivier Demory for the opportunity. Left: Longitudinal brightness map of the hot-Jupiter Kepler 7b in the visible derived from its Kepler light-curve phase folded over 4 years. It reveals a very bright half dayside on the western side of the substellar point. Right: Artistic representation of the planet compared to Jupiter (credit: NASA). Our findings lead to insights in the circulation regime and cloud particles of the planet: warm air is brought from the substellar point to the night side by the eastward equatorial winds, once on the night side the temperature drops enough for condensate with high reflectivity to form and be brought back to the dayside where they reflect the stellar light until reaching the substellar point and be sublimated, and the cycle continues. For more info, click here.

Fig 3, 2.jpg

First map of a terrestrial exoplanet. Many thanks to Professor Brice-Olivier Demory for the opportunity. Left: Longitudinal brightness map of the super-Earth 55 Cnc e in the infrared derived from its Spitzer light-curve. It reveals a very bright half dayside on the eastern side of the substellar point. Right: Artistic representation of the planet showing the presence of a larger pool of magma ocean (bright in the infrared) on the east of the substellar point (credit: NASA). For more info, click here.

Fig 4, 1.png
Fig 4, 2.jpg
fig 5, 1.png
Fig 5, 2.png

First transient map of an exoplanet. Many thanks to Professor Nikole Lewis for the opportunity. Left: Snapshots of the visible hemisphere of the eccentric Jupiter HD 80606 b during its periastron passage. The brightness temperature scale runs from 500 K (black) to 1500 K (white). Right: HD 80606 b’s transient heating observed in Spitzer/IRAC 4.5 and 8.0 μm bands. (A) HD 80606 b’s relative thermal fluxes in Spitzer/IRAC 4.5 and 8.0 μm bands as a function of the time from periastron passage. Data as dots, best-fit models as lines. Conversion of the relative planetary fluxes to hemisphere-averaged brightness temperature are

shown on the right for each channel. (B) HD 80606 b’s hemisphere-averaged brightness temperature as a function of the time from periastron passage derived from the planetary fluxes introduced in (A). The dashed lines represent the equilibrium temperature of a planet assuming full and instantaneous energy redistribution for an absorptivity of 0.8 and 0.2. For more info, click here.

Fig 6, 1.png
Fig 6, 2.png

First detection of plant-induced stellar pulsations. Many thanks to Professor Nikole Lewis for the opportunity. Left: ~40 p.p.m. pulsations found in 18 phase-folded occultations of the eccentric hot-Jupiter HAT-P-2 b. Surviving the phase-folding over a period of a couple of years and being see in occultation, these pulsations must be harmonics of the planet’s orbital frequency and yet of stellar origin. Right: Periodogram of the Spitzer 4.5 micron photometry after subtraction of the best fit models highlighting the presence of signals corresponding to HAT-P-2 b’s 79th and 91st orbital harmonics. Such pulsations are not expected and suggest that further investigations of the system and its planet-star interactions. For more info, click here.


Turning now to transmission spectroscopy, my very first gateway to exoplanetary sciences (back in 2010!) thanks to Professor Sara Seager. Many many thanks to her. I am fond of this figure as it simply introduced the basics of transmission spectroscopy while presenting for the first time the central role that the Euler-Mascheroni constant (\gamma_{EM}) plays in the underlying mathematics. For more info, click here.

Figure 8.png

Figure born of the hope to get access to JWST observations of temperate terrestrial planets before the end of my PhD (back in 2012..). It shows the capability to retrieve with good precision the atmospheric properties (temperature, pressure, composition) of an Earth-sized planet around a M7V star from ~200-hrs of in transit observation—transmission spectrum in panel A.  For more info, click here.

Fig 9.png

Central piece to the insights in transmission spectroscopy introduced in de Wit & Seager 2013. This multi-panel figure shows the dependence on pressure and temperature of an absorption line profile at a given frequency. If curious to no more check Figure S.2 here.

Fig 10.png

Finally, the first temperate terrestrial exoplanets that are amenable for atmospheric studies are detected! Many thanks to Professor Michael Gillon and the whole TRAPPIST team. The figure here contextualizes the remarkable TRAPPIST-1 system. It shows the expected signal in transmission for the first three planets found and their expected SNR for JWST observations relative to GJ 1214 b. For more info, click here and here

Fig 11, 1.jpg
Fig 11, 2.png

On May the 4th 2016, the Hubble Space Telescope is pointed for the first time ever to temperate terrestrial worlds beyond the edge of the Solar System to initiate the characterization of their atmospheres. Left: Artistic representation of the simultaneous transit of TRAPPIST-1 b and c. Right: Joint transmission spectra of the planets allowing us to reject the presence of hydrogen-rich atmospheres, which would have rendered such planets inhabitable due to hydrogen’s strong greenhouse gases effects. Figure produced by Professor Amaury Triaud. For more info, click here.

Fig 12.png

The exploration of TRAPPIST-1’s planetary system continues with the Hubble Space Telescope. It is an ongoing adventure, which is soon to be followed by one with JWST’s. Stay tuned! Figure produced by Professor Hannah R. Wakeford. For more info, click here and here