Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

At the stellar noise frontier: a transit survey of 121 TESS M3--M6 dwarfs

Yohann Tschudi

Abstract

M-dwarf stars are the most favorable hosts for detecting small transiting planets, yet mid-to-late M-dwarfs that acquired sufficient TESS multi-sector coverage only through recent Cycle 6+ observations represent a newly accessible discovery space. This paper presents a systematic transit survey of 121 M3-M6 dwarfs (Teff = 2700-3400 K) selected as "newly enabled" targets -- stars with <=2 archival TESS sectors that only recently crossed the multi-sector detection threshold, covering P = 0.5-100 d. The sample was selected from 498,312 TIC M-dwarfs via a 9-step funnel. The pipeline combines TLS with a signal validation cascade, TRICERATOPS vetting, Gaia DR3 verification, and three empirical signal reliability tests. Pipeline validation achieved 100% recovery (16/16 planets) on 10 known systems with zero false positives. The survey identifies 20 transit-like signals across 16 systems, none with prior TOI designations. The reliability framework classifies 2 as Tier 1 (High Robustness), 7 as Tier 2 (Moderate), and 10 as Tier 3 (Noise-Susceptible); one monotransit is excluded. No candidate SDE significantly exceeds its host star's noise floor. The global false alarm rate is 17.4% (21/121; Wilson 95% CI: [11.6%, 25.1%]).The 2 Tier 1 candidates are priorities for RV confirmation. The 10 Tier 3 candidates require additional TESS sectors to establish signal persistence; 9 systems need high-resolution imaging.

At the stellar noise frontier: a transit survey of 121 TESS M3--M6 dwarfs

Abstract

M-dwarf stars are the most favorable hosts for detecting small transiting planets, yet mid-to-late M-dwarfs that acquired sufficient TESS multi-sector coverage only through recent Cycle 6+ observations represent a newly accessible discovery space. This paper presents a systematic transit survey of 121 M3-M6 dwarfs (Teff = 2700-3400 K) selected as "newly enabled" targets -- stars with <=2 archival TESS sectors that only recently crossed the multi-sector detection threshold, covering P = 0.5-100 d. The sample was selected from 498,312 TIC M-dwarfs via a 9-step funnel. The pipeline combines TLS with a signal validation cascade, TRICERATOPS vetting, Gaia DR3 verification, and three empirical signal reliability tests. Pipeline validation achieved 100% recovery (16/16 planets) on 10 known systems with zero false positives. The survey identifies 20 transit-like signals across 16 systems, none with prior TOI designations. The reliability framework classifies 2 as Tier 1 (High Robustness), 7 as Tier 2 (Moderate), and 10 as Tier 3 (Noise-Susceptible); one monotransit is excluded. No candidate SDE significantly exceeds its host star's noise floor. The global false alarm rate is 17.4% (21/121; Wilson 95% CI: [11.6%, 25.1%]).The 2 Tier 1 candidates are priorities for RV confirmation. The 10 Tier 3 candidates require additional TESS sectors to establish signal persistence; 9 systems need high-resolution imaging.
Paper Structure (43 sections, 1 equation, 5 figures, 24 tables)

This paper contains 43 sections, 1 equation, 5 figures, 24 tables.

Table of Contents

  1. Introduction
  2. Target Selection Strategy
  3. The "Newly Enabled M3--M6" Sample: 9-Step Selection Funnel
  4. Sample Composition by Spectral Type
  5. Selection Criteria
  6. Methodology
  7. Data Acquisition and Preprocessing
  8. Transit Detection
  9. Harmonic Correction
  10. Signal Validation Cascade
  11. Quality flag system.
  12. Harmonic nest cleanup.
  13. Candidate Validation
  14. Statistical Vetting and Parameter Fitting
  15. Physical Verification via Gaia DR3
  16. ...and 28 more sections

Figures (5)

  • Figure 1: Gaia DR3 physical verification decision tree. Candidates passing statistical vetting are classified based on aperture contamination ($G_{\rm max}$ from Eq. \ref{['eq:gmax']}) and astrometric quality (RUWE).
  • Figure 2: Validation lightcurves (1/2). Top row: L 98-59 c, TOI-700 d (HZ). Bottom row: Gliese 12 b (HZ), TOI-406 b.
  • Figure 3: Validation lightcurves (2/2). Top row: TOI-782 b, GJ 357 b. Bottom row: GJ 3473 b, TOI-6086 b.
  • Figure 4: Phase-folded lightcurves for the two Tier 1 candidates (TIC 211401412), two Tier 2 candidates downgraded by Fourier FAP (TIC 117955697, TIC 77634567), and the monotransit candidate TIC 72750190. Top row: TIC 117955697 P1 ($P = 1.83$ d, Tier 2) and P2 ($P = 3.79$ d, Tier 2). Middle row: TIC 211401412 P1 ($P = 0.42$ d, Tier 1) and P2 ($P = 0.63$ d, Tier 1). Bottom row: TIC 77634567 ($P = 4.12$ d, Tier 2) and TIC 72750190 ($P = 50.3$ d, monotransit, tentative HZ). Grey points: individual TLS phase-folded measurements. Blue circles with error bars: mean-binned data (60 bins, SEM). Orange curve: batman transit model Kreidberg2015 with quadratic limb darkening (TESS bandpass; Claret2017); shape parameters from MCMC posterior medians, depth scaled to match the TLS-observed transit depth. TIC 72750190 uses a TLS box-model estimate (MCMC did not converge; single transit).
  • Figure 5: Phase-folded lightcurves for the 14 candidates not shown in Fig. \ref{['fig:phasefold_tier1']} (PLANET_CANDIDATE and NEEDS_HR_IMAGING), ordered by orbital period. Format as in Fig. \ref{['fig:phasefold_tier1']}. Grey: individual measurements; blue: mean-binned data; orange: batman model with MCMC shape parameters. The three HZ candidates (TIC 186664624 P2, TIC 139471702, TIC-80315364; all Tier 3) are included here. TIC 72750190 ($P = 50.3$ d, tentative HZ, monotransit) is excluded: MCMC did not converge due to a single partial transit.