Radius valley scaling among low mass stars with TESS

TL;DR

The study investigates whether the radius valley observed around Sun-like stars extends to low-mass M dwarfs and how it scales with host stellar mass. Using a volume-limited sample of TESS TOIs within 120 pc and precise Bioverse stellar parameters, the authors recalibrate planet radii and apply Kernel Density Estimation and Gaussian Mixture Models to map the planet size distribution across GKM hosts. They find a clear M-dwarf radius valley at $1.64 \pm 0.03\,R_{\oplus}$ with roughly $45\%$ depth and derive a scaling relation for GKM stars of $R_p \propto M_*^{0.15 \pm 0.04}$, while M dwarfs show a flatter slope ($\beta = 0.12 \pm 0.06$), suggesting additional formation pathways beyond photoevaporation. The results align with pebble accretion and water-world formation scenarios and demonstrate that large, homogeneous datasets enable robust population-level insights into planet formation around the lowest-mass stars.

Abstract

The Transiting Exoplanet Survey Satellite (TESS) has been highly successful in detecting planets in close orbits around low-mass stars, particularly M dwarfs. This presents a valuable opportunity to conduct detailed population studies to understand how these planets depend on the properties of their host stars. The previously observed radius valley in Sun-like stars has not been unambiguously detected among M dwarfs, and how its properties varies when compared with more massive stars remains uncertain. We use a volume-limited sample of low mass stars with precise photometric stellar parameters from the bioverse catalog of TESS Objects of Interest (TOIs) confirmed planets and candidates within 120 pc. We detect the radius valley around M dwarfs at a location of 1.64 $\pm$ 0.03 $R_{\oplus}$ and with a depth of approximately 45${\%}$. The radius valley among GKM stars scales with stellar mass as $R_p \propto M_*^{0.15\pm 0.04}$. The slope is consistent, within 0.3$σ$, with those around Sun-like stars. For M dwarfs, the discrepancy is 3.6$σ$ with the extrapolated slope from the Kepler FGK sample, marking the point where the deviation from previous results begins. Moreover, we do not see a clear shift in the radius valley between early and mid M dwarfs. The flatter scaling of the radius valley for lower-mass stars suggests that mechanisms other than atmospheric mass loss through photoevaporation may shape the radius distribution of planets around M dwarfs. Comparison of the slope with various planet formation and evolution models matches well with pebble accretion models including waterworlds, indicating a potentially different regime of planet formation that can be probed with exoplanets around the lowest mass stars.

Figures (9)

• Figure 1: Histogram comparing TESS Objects of Interest (TOIs) and Kepler Objects of Interest (KOIs) from the NASA Exoplanet Archive, with star counts indicated in parentheses. The inset plot (top left) highlights the significant increase in low-mass stars identified by TESS, which enhances our ability to study these stars in greater detail.
• Figure 2: Kernel Density Estimates along with the histograms of Planet radii distribution of the entire sample and M,K,G stellar types.
• Figure 3: GMM fits to the unbinned radius data, along with corresponding histograms, are presented for the entire low-mass star sample and for M, K, and G stellar types. The Super-Earth and Mini-Neptune peaks, as well as the radius valley, are indicated in each plot.
• Figure 4: Slope values, $\beta = \left( \frac{\partial \log R_{\text{gap}}}{\partial \log M_*} \right)$ from various studies, grouped by stellar mass ranges (M dwarfs, FGKM dwarfs, FGK dwarfs) in the radius- stellar mass plane. The colored boxes represent the uncertainty ranges, while the horizontal dashes indicate the slope values for each study. A rise in the slope values can be observed as we go from M dwarfs to Sun-like stars.
• Figure 5: GMM fits along with their histograms for early and late M dwarfs.