TESS Planet Occurrence Rates Reveal the Disappearance of the Radius Valley Around Mid-to-Late M Dwarfs

Abstract

We present the deepest systematic search for planets around mid-to-late M dwarfs to date. We have surveyed 8134 mid-to-late M dwarfs observed by TESS with a custom built pipeline and recover 77 vetted transiting planet candidates. We characterize the sensitivity of our survey via injection-recovery and measure the occurrence rate of planets as a function of orbital period, instellation, and planet radius. We measure a cumulative occurrence rate of $1.10\pm0.16$ planets per star with radii $>1\, R_\oplus$ orbiting within 30 days. This value is consistent with the cumulative occurrence rate around early M dwarfs, making M dwarfs collectively the most prolific hosts of small close-in planets. Unlike the bimodal Radius Valley exhibited by close-in planet population around FGK and early M dwarfs, we recover a unimodal planet radius distribution peaking at $1.25\pm0.05 \, R_\oplus$. We additionally find $0.954\pm0.147$ super-Earths and $0.148\pm0.045$ sub-Neptunes per star, with super-Earths outnumbering sub-Neptunes 5.5:1, firmly demonstrating that the Radius Valley disappears around the lowest mass stars. The dearth of sub-Neptunes around mid-to-late M dwarfs is consistent with predictions from water-rich pebble accretion models that predict a fading Radius Valley with decreasing stellar mass. Our results support the emerging idea that the sub-Neptune population around M dwarfs is composed of water-rich worlds. We find no hot Jupiters in our survey and set an upper limit of 0.012 hot Jupiters per mid-to-late M dwarf within 10 days.

Figures (17)

• Figure 1: $M_{K_s}$/$G_{\mathrm{BP}}-G_{\mathrm{RP}}$ color-magnitude diagram of our original target sample, with color cuts plotted with the red dashed lines. Targets hosting TOIs are encircled.
• Figure 2: Distributions of $T_{\mathrm{mag}}$, distance, effective temperatures, radii, masses and of our 8134 targets. The median of each column and row is marked with a dashed red line, which is also reported in the column titles along with each parameters' 16th and 84th percentiles.
• Figure 3: Selected light curves from TICs 98796344 (top) and 328081248 (bottom) respectively after flare removal. The trends recovered by a Gaussian process (top) and median filter (bottom) are overplotted in black, and the transit events in each light curve are highlighted in gold.
• Figure 4: Example in-transit centroid offsets for TOIs 562.01 (left) and 3494.01 (right). The star marker in each panel marks the out-of-transit centroid and the black points mark the relative in-transit centroid positions of each TCE. The contours mark the 1, 2, and $3\sigma$ confidence intervals of the estimated in-transit centroid distribution. TOI-562.01 has a 97% on-target probability and TOI-3494.01 has a 0% on-target probability.
• Figure 5: Injection/recovery maps over each of our six injection recovery target bins that span different TESS magnitudes (columns) and non-rotating (top row) and rotating targets (bottom row) showing our pipeline's sensitivity to transiting planet signals across period--$R_p/R_\star$ space. Each cell in this recursive structure contains at least 50 injected planets with the black grid denoting the boundaries of each cell.