Automatic search for transiting planets in TESS-SPOC FFIs with RAVEN: over 100 newly validated planets and over 2000 vetted candidates

Abstract

Space-based missions such as TESS are identifying a wealth of short-period ($\lesssim30$ d) transiting planets. Despite the growing number of confirmed and candidate planets, the sample is still incomplete and highly biased, challenging demographic studies. Moreover, there are still a large number of unconfirmed candidates that can end up being false positives. We use the new pipeline RAVEN to perform a uniform search and validation of transiting planet candidates in TESS data. We focus on a magnitude-limited sample of over 2.2 million main sequence stars well characterised by Gaia and observed by TESS in its Full Frame Images during its first 4 years of operations (sectors 1 to 55). We aim to detect candidates with periods within $0.5-16$ days. RAVEN detects candidates with a box least squares algorithm, classifies them into transiting planets and false positives using machine learning models trained with realistic simulations, and performs statistical validation. We present several samples of candidates with different levels of vetting and validation. We newly validate 118 planets, including 31 newly detected here. We also present a sample of over 2000 candidates not validated but with high probability of being planets, including $\sim1000$ new candidates, a small sample of newly identified mono- and duo-transiting candidates, and a sample of large radii ($>8~\mathrm{R_{\oplus}}$) candidates with high planet probability suited for further follow-up. Our samples of vetted and validated transiting planet candidates represent a major effort towards improving the candidate sample from TESS.

Figures (13)

• Figure 1: Gaia properties of our stellar samples: apparent Gaia magnitude $G$ (from DR3, top left panel), distance (computed by inverting the DR3 parallax, top right, note that the parallax values of the stars in our sample have uncertainties better than 20%), stellar radius (from DR2, bottom left), and effective temperature (from DR2, bottom right). Different colours/line styles represent different samples used in this work: black dotted lines correspond to all stars (after SDE and MES cuts, see Sect. \ref{['sec:bls']}), blue dashed dotted lines show the subsample of stars with candidates with NSFP mean classifier probability $\geq0.9$ (see Sect. \ref{['sec:nsfp_vetting']}), pink dashed lines show the subsample stars with vetted candidates (see Sect. \ref{['sec:res_cand_vet']}), and yellow solid lines show the subsample stars with validated candidates (see Sect. \ref{['sec:res_cand_val']}). The numbers in parenthesis in the legend show the number of stars in each sample. The inset panels show a zoom in of the stars in the vetted and validated samples. Note that the y-axes of the main panels are in logarithmic scale while, in the insets, they are in linear scale.
• Figure 2: Classification results of all candidates for the Planet-NSFP classifiers. Each histogram shows the posterior probability of a different classifier: pink dash-dotted line for the GBDT, blue dashed line for the GP, and solid black line and grey-filled for the mean of both GBDT and GP. Candidates with probability close to 1 are classified as Planet, and candidates with probability close to 0 are classified as NSFP. The inset panel shows a zoom in on the candidates with probabilities above 0.8.
• Figure 3: Distribution of the main BLS properties, from top to bottom: period $P$, transit duration $t_\mathrm{dur}$, depth $\delta$, signal detection efficiency SDE, multiple event statistics MES, and candidate peak number, colour-coded with the candidate peak number. The different columns show different samples of candidates, from left to right: all candidates (after SDE and MES cuts), subsample of candidates with non-simulated false positive (NSFP) mean classifier probability $\geq0.9$ (see Sect. \ref{['sec:nsfp_vetting']}), vetted candidates (see Sect. \ref{['sec:res_cand_vet']}), and validated candidates (see Sect. \ref{['sec:res_cand_val']}). In the first row of panels, the top right numbers show the total number of candidates in each sample. In the last row of panels, the numbers on each bin show the number of candidates per BLS peak number.
• Figure 4: Vetted sample of 2170 candidates in period-radius space. The values shown are the results of our juliet fits (see Sect. \ref{['sec:juliet']}). Black circles correspond to new candidates orbiting stars not known to host any TOI/CTOI, blue triangles show candidates matching known TOIs (i.e. recovered), pink squares show candidates on stars known to have a TOI that do not match the known TOI candidates (i.e. non-recovered, some of which could be new candidates), yellow down triangles show candidates matching known CTOIs (i.e. recovered), and green diamonds show candidates on stars known to have a CTOI that do not match the known CTOI candidates (i.e. non-recovered, some of which could be new candidates). Solid grey lines and grey-shaded area show the Neptunian desert limits according to mazeh2016desert, and dashed grey lines show the recently derived limits between the Neptunian desert, ridge, and savannah from castrogonzalez2024neptune. Dotted grey vertical and horizontal lines show our pipeline's detection/validation limits: periods from 0.5 to 16 d and radii below $8~\mathrm{R_{\oplus}}\xspace$.
• Figure 5: Validated sample of 143 candidates in period-radius space. All symbols show the same as Fig. \ref{['fig:cand_per_rp_vet']} but for the validated sample rather than the vetted one. Note that the validated sample is included in the vetted one. Note that the CTOI TIC 270471727 BLS peak 1 (green diamond) is actually a CTOI that we recover (TIC 270471727.01), despite the $t_0$ not matching (see Appendix \ref{['sec:cand_vet_toi_other_recovery']}).