Demographics of Close-In TESS Exoplanets Orbiting FGK Main-sequence Stars

TL;DR

Using four years of TESS-SPOC FFIs cross-matched with Gaia, this study delivers a high-precision census of close-in exoplanets around FGK main-sequence stars. The authors combine a rigorous detection, vetting (RAVEN), and completeness framework with two inference approaches—IDEM with a weighted KDE and a hierarchical Bayesian model—to map occurrence density on a $10\times10$ grid in the period–radius plane, reporting $9.4^{+0.7}_{-0.6}\%$ overall, $0.39^{+0.03}_{-0.02}\%$ hot Jupiters, and $0.08\pm0.01\%$ for the Neptunian desert; Gaia within 100 pc yields $15.4^{+1.6}_{-1.5}\%$ and $0.42^{+0.16}_{-0.12}\%$ for hot Jupiters. The results are broadly consistent with Kepler but achieve markedly tighter uncertainties, illustrating the power of a uniform, bright-star sample for exoplanet demographics. The findings provide a robust empirical baseline to test planet-formation and evolution theories and highlight directions for extending the census with more TESS data and Gaia-derived volume-limited samples.

Abstract

Understanding the demographics of close-in planets is crucial for insights into exoplanet formation and evolution. We present a detailed analysis of occurrence rates for close-in (0.5-16 day) planets with radii between 2 and 20$\,R_{\oplus}$ around FGK main-sequence stars. Our study uses a comprehensive sample from four years of TESS Science Processing Operations Center full-frame image data cross-matched with Gaia, analysed through our rigorous detection, vetting, and validation pipeline. Using high-confidence planet candidates, we apply a hierarchical Bayesian model to determine occurrence rates in the two-dimensional orbital period-radius plane. Our results are presented using 10-by-10 bins across the period-radius parameter space, offering unprecedented resolution and statistical precision. We find an overall occurrence rate of $9.4^{+0.7}_{-0.6}\%$. When using identical binning, our occurrence rate posteriors distributions align with Kepler's but have a magnitude smaller uncertainties on average. For hot Jupiters, we estimate the overall occurrence rate of $0.39^{+0.03}_{-0.02}\%$. This value is consistent with the previous Kepler FGK-type result within $1σ$. We find an overall occurrence rate of Neptunian desert planets of $0.08\pm0.01\%$, to our knowledge the first such determination. Additionally, in a volume-limited Gaia subsample within 100 pc in the same parameter region, we measure an overall planet occurrence rate of $15.4^{+1.6}_{-1.5}\%$ and a hot Jupiter occurrence rate of $0.42^{+0.16}_{-0.12}\%$. Our results establishes an improved foundation for constraining theoretical models of exoplanet populations.

Figures (16)

• Figure 1: Effective temperature distributions of our TESS-SPOC FGK main sequence stars. Shaded regions represent the temperature boundaries for different FGK spectral types.
• Figure 2: Observational bias, detection efficiency, NSFP completeness, signal-to-noise ratio (SNR) completeness, FP completeness, and combined efficiency of transiting planets in our stellar sample as functions of orbital period and planet radius. Each subplot title describes its process; higher values indicate more planets retained.
• Figure 3: NSFP completeness, FPs completeness, and combined efficiency as a function of the orbital period and planet radius. The NSFP and FPs models are trained on our uniformly injected sample.
• Figure 4: wKDE in period and radius space, with the colour scale normalized logarithmically for better visibility and higher dynamic range. Two Neptunian desert boundaries and a potential "ridge" region estimated from Kepler are shown with dashed black lines mazehDearthShortperiodNeptunian2016 and dashed red lines castro-gonzalezMappingExoNeptunianLandscape2024. The bandwidth of KDE is 0.3, which is manually tuned for clearly showing the general features like hot Jupiters and desert boundaries. The range of our colour scale is limited to match that of Fig. \ref{['fig:occurrence-rate']} to have consistent visibility.
• Figure 5: Occurrence rates of planet candidates around TESS-SPOC FGK main-sequence stars. Each cell displays the average occurrence as a percentage (number of planets per 100 stars), along with the corresponding uncertainties (the difference between the 16th and 84th percentiles and the median). For clarity, text in cells with an occurrence rate below 0.01% is displayed in white. The colour of each cell encodes the occurrence rate density, ${\mathrm{d}^2 N_\mathrm{p}}/(\mathrm{d}\ln P\,\mathrm{d}\ln R_\mathrm{p})$, per 100 stars. The logarithmic colour scale is truncated at 0.5 to enhance visibility. We also provide an https://cuikaiming.com/TESS-SPOC-Occurrence-Rate/ that allows users to easily select any region and obtain its total occurrence rate.