TESS phase curve of ultra-hot Jupiter WASP-189 b

TL;DR

WASP-189 b, an ultra-hot Jupiter around a hot A-type star, is studied through TESS phase curves and CHEOPS transits to constrain atmospheric structure and orbital geometry. The authors perform gravity-darkened transit modelling and phase-curve analyses, obtaining a near-polar spin–orbit alignment with $\Psi \approx 89.46^{+1.08}_{-1.08}$ deg and no significant orbital precession over ~2 years. The phase curve yields an occultation depth of $203.4^{+16.2}_{-16.3}$ ppm and a nightside flux consistent with zero, implying dayside emission dominated by thermal radiation with a low Bond albedo and a heat-redistribution efficiency in the range $\varepsilon \approx 0.09$–$0.41$. Atmospheric forward modelling and phase-curve inversion indicate a temperature inversion likely caused by short-wave absorbers, and bandpass differences between CHEOPS and TESS suggest modest altitude-dependent thermal structure, with reflectivity playing a minor role.

Abstract

The thermal structures of highly irradiated ultra-hot Jupiters can deviate substantially from those of cooler hot Jupiters. For planets orbiting rapidly rotating, and consequently oblate, host stars, photometric light curves provide a unique opportunity to measure the spin-orbit angle. Moreover, in systems with significant spin-orbit misalignment, the stellar oblateness can induce observable orbital precession. We wish to study the atmosphere and orbital architecture of an ultra-hot Jupiter WASP-189 b, orbiting around a hot A-type star. We use the photometric phase curves and gravity-darkened transits of WASP-189 b observed with the Transiting Exoplanet Survey Satellite (TESS), complemented with the archival observations from CHaracterising ExOPlanet Satellite (CHEOPS). We detected a phase curve signal with significant occultation depth of 203.4 (+16.2) (-16.3) ppm, while the nightside flux, -71.8 (+36.4) (-36.0) ppm, is consistent with zero at 2-sigma. We invert the phase curve signal to construct the temperature map of the planet. The map was subsequently used to estimate the Bond albedo and heat redistribution efficiency, the expected median ranges of which are found to be 0.19-0.35 and 0.09-0.41, respectively. Finally, we analysed gravity-darkened transits to find that the planet is in polar orbit with the spin-orbit angle of 89.46 (+1.08) (-1.08) deg. We found no hint of orbital precession while comparing our results with those from the literature. Our observations, together with atmospheric modelling, suggest that the dayside emission of WASP-189 b in TESS and CHEOPS bandpasses is dominated by thermal emission from an atmosphere with extremely inefficient heat transport and negligible contribution from reflected light.

Figures (8)

• Figure 1: Detrended and phase-folded TESS data along with the best-fitted model. The light and dark blue points show the unbinned and binned data points. The dark blue and orange lines are the median model and models computed from randomly selected posteriors, respectively. A full phase curve with a transit and occultation is shown in the top panel, while the middle panel shows a zoom-in on the transit (middle left) and occultation/phase curve (middle right). The dashed line in the middle right panel represents the level of stellar flux (i.e., flux level during the occultation). The gravity darkened asymmetric transit and a phase variation, along with an occultation, are clearly visible in the middle panel. The bottom panel shows the residuals after subtracting the median model from the raw data.
• Figure 2: Theoretical occultation spectra of WASP-189 b for two different values of the heat-redistribution factor $\epsilon$. The squares represent the bandpass-integrated occultation depths in the CHEOPS and TESS bandpasses based on the theoretical model calculations. The observational data is shown in black.
• Figure 3: Brightness and temperature maps of the planet. (Top) Equatorial brightness in the units of stellar brightness as a function of longitude. (Middle) 1D temperature distribution on planetary equator as a function of planetary longitude. (Bottom) The median 2D latitude-longitude temperature map of the planet. The dark blue and orange lines in the top and middle plots show the median and randomly selected models from the posterior distribution.
• Figure 4: CHEOPS phase curve and inferred temperature map from it. (Top) The detrended and phase-folded CHEOPS data along with the median (dark blue line, top panel) and models computed from randomly selected posteriors (orange line, top panel). The bottom panel shows the residuals after subtracting the median model. The light and dark blue points show the unbinned and binned data, respectively. (Bottom) The derived median temperature map as a function of latitude and longitude.
• Figure 5: Equatorial temperatures as a function of longitude as observed by TESS (in orange) and CHEOPS (in blue) bandpasses. The solid lines give median values of temperatures, while the shaded regions show the bands of 1$\sigma$ uncertainty in temperatures. We do not plot some nightside longitudes since our modelling cannot properly constrain temperatures on very high longitudes.