Confirmation Of Water Emission In The Dayside Spectrum Of The Ultrahot Jupiter WASP-121b


(a) Published eclipse depth measurements for WASP-121b across the red optical and near-infrared wavelength range covered by the TESS and HST WFC3 passbands. Note in particular the improved WFC3 G141 signal-to-noise achieved for the present study with five eclipse observations, compared to the original data presented in Evans et al. (2017) for a single eclipse observation. (b) Corresponding planetary emission extending out to longer wavelengths including the Spitzer IRAC passbands. The errorbars for the IRAC measurements are not much larger than the marker symbols on this vertical scale. In both panels, the dark yellow line shows the spectrum assuming the planet radiates as a blackbody with a best-fit temperature of 2700 K and the pale yellow envelope indicates blackbody spectra for temperatures of 2330 K and 2970 K. The latter encompass a plausible range of emission under limiting assumptions for the albedo and day-night heat recirculation. As labeled in panel (a), other solid lines show best-fit models obtained for the three retrieval analyses described in the main text, which all incorporate the updated WFC3 G141 spectrum, along with that obtained for our previous retrieval analysis published in Mikal-Evans et al. (2019). Spectral emission features due to H− and H2O are also labeled in panel (a).

We present four new secondary eclipse observations for the ultrahot Jupiter WASP-121b acquired using the Hubble Space Telescope Wide Field Camera 3.

The eclipse depth is measured to a median precision of 60ppm across 28 spectroscopic channels spanning the 1.12-1.64 micron wavelength range. This is a considerable improvement to the 90ppm precision we achieved previously for a single eclipse observation using the same observing setup. Combining these data with those reported at other wavelengths, a blackbody spectrum for WASP-121b is ruled out at >6-sigma confidence and we confirm the interpretation of previous retrieval analyses that found the data is best explained by a dayside thermal inversion. The updated spectrum clearly resolves the water emission band at 1.3-1.6 micron, with higher signal-to-noise than before.

It also fails to reproduce a bump in the spectrum at 1.25 micron derived from the first eclipse observation, which had tentatively been attributed to VO emission. We conclude the latter was either a statistical fluctuation or a systematic artefact specific to the first eclipse dataset.

Thomas Mikal-Evans, David K. Sing, Tiffany Kataria, Hannah R. Wakeford, Nathan J. Mayne, Nikole K. Lewis, Joanna K. Barstow, Jessica J. Spake

Comments: Accepted in MNRAS
Subjects: Earth and Planetary Astrophysics (astro-ph.EP)
Cite as: arXiv:2005.09631 [astro-ph.EP] (or arXiv:2005.09631v1 [astro-ph.EP] for this version)
Submission history
From: Thomas Evans
[v1] Tue, 19 May 2020 17:58:14 UTC (274 KB)
https://arxiv.org/abs/2005.09631
Astrobiology, Astrochemistry

Please follow Astrobiology on Twitter.


  • submit to reddit