TESS Investigation -- Demographics of Young Exoplanets (TI-DYE) IV: a Jovian radius planet orbiting a 34 Myr Sun-like star in the Vela association

TL;DR

This TI-DYE IV paper validates TOI-6448 b, a Jovian-radius planet with $R_P \approx 0.76\text{--}0.78\,R_J$ on a $P \approx 14.84$ day orbit around a Sun-like star whose age is tightly constrained to $34 \pm 3$ Myr as a member of Vela Population IV. The team combines TESS light curves, ground-based photometry in multiple bands, and high-resolution spectroscopy to robustly validate the planet, quantify false-positive scenarios, and extract precise planetary and stellar parameters. An injection-recovery analysis indicates about 80% completeness for planets in the same period-radius regime, underscoring the sincerity of the detection, while a multi-method age analysis (isochronal, variability-based EVA, and gyrochronology) yields a coherent young-age estimate. The results strengthen the empirical picture that young exoplanets at $<50$ Myr tend to be inflated and occupy a regime distinct from older Kepler planets, with implications for formation scenarios and early radii evolution, and set the stage for discovering dozens more in Vela, Taurus-Auriga, Sco-Cen, and Orion.

Abstract

The discovery of infant (< 50 Myr), close-in (<30-day period) planets is vital in understanding the formation mechanisms that lead to the distribution of mature transiting planets as discovered by Kepler. Despite several discoveries in this age bin, the sample is still too small for a robust statistical comparison to older planets. Here we report the validation of TOI-6448 b, an 8.8 +/- 0.8 Re planet on a 14.8 day orbit. TOI-6448 was previously identified to be a likely member of Vela Population IV. We confirm the star's membership and re-derive the age of the cluster using isochrones, variability, and gyrochronology. We find the star, and thus planet, to be 34 +/- 3 Myr. Like other young planets, TOI-6448 b lands in a region of parameter space with few older planets. While just one data point, this fits with prior findings of an excess of 5-11Re planets around young stars far beyond what can be explained by reduced sensitivity at young ages. Our ongoing search of Vela, Taurus-Auriga, Sco-Cen, and Orion are expected to reveal dozens more < 50 Myr transiting planets.

Figures (12)

• Figure 1: Fit to the SED of TOI-6448, showing the best-fit template (black) filler model (blue), observed photometry (red) and synthetic photometry (green). For the photometry, the vertical error bars are the photometric uncertainties, while the horizontal error bars represent the filter width. The bottom panel shows the residuals in units of standard deviations. The fit is marginalized over a range of templates and extinctions; this one is a G1.5V with low extinction ($A_V=0.05$).
• Figure 2: Left) Phase-folded TESS light curve of TOI-6448 (gray) binned to 10-minute intervals (purple) for clarity. The best-fit transit model is shown as the bright, opaque red line with 25 sample fits pulled from the posterior shown as the dark, translucent red lines. The GP has been removed from the data and model. Right) Representative section of the TESS light curve (gray) binned to 10-minute intervals (purple) for clarity. The red line shows the GP model, and the pink lines show the locations of the transit events.
• Figure 3: Corner plot of the eccentricity ($e$) and argument of periastron ($\omega$) fit from the TESS data. Though the eccentric fit is preferred, the $e$ posterior agrees with a circular orbit.
• Figure 4: Ground-based light curves (blue) binned to 10 minute intervals (green) for clarity. The transit model is shown in purple. In each set, the top plot is the raw data, the middle plot has the stellar variability model removed, and the bottom plot is the residuals. Bottom right: Transit depth, measured by $(R_p/R_*)^2$, posterior from each transit fit.
• Figure 5: Injection-recovery map in period-radius space for TOI-6448. The green star marks TOI-6448 b. Blue points indicate recovered planets and red points are planets which were not recovered. Only 20% of the individual injected signals are shown for clarity. The background is color-coded by overall completeness for a given bin.