Table of Contents
Fetching ...

TOI-2322: two transiting rocky planets close to the stellar rotation period and its first harmonic

M. J. Hobson, A. Suárez Mascareño, C. Lovis, F. Bouchy, B. Lavie, M. Cretignier, A. M. Silva, S. G. Sousa, H. M. Tabernero, V. Adibekyan, C. Allende Prieto, Y. Alibert, S. C. C. Barros, A. Castro-González, K. A. Collins, S. Cristiani, V. D'Odorico, M. Damasso, D. Dragomir, X. Dumusque, D. Ehrenreich, P. Figueira, R. Génova Santos, B. Goeke, J. I. González Hernández, K. Hesse, J. Lillo-Box, G. Lo Curto, C. J. A. P. Martins, A. Mehner, G. Micela, P. Molaro, N. J. Nunes, E. Palle, V. M. Passegger, F. Pepe, R. Rebolo, J. Rodrigues, N. Santos, A. Sozzetti, B. M. Tofflemire, S. Udry, C. Watkins, M. -R. Zapatero Osorio, C. Ziegler

TL;DR

Stellar activity can mimic or obscure small exoplanets in RV data, especially when planetary periods lie near the stellar rotation period. The authors combine ESPRESSO and HARPS RVs with TESS transits and ground-based photometry, modeling the data with both a one-dimensional GP and a multivariate GP that links RVs to activity indicators (FWHM and BIS) to separate planetary signals from stellar activity. They confirm two transiting planets around TOI-2322: TOI-2322 b with P ≈ 11.31 d and TOI-2322 c with P ≈ 20.23 d; b remains mass-constrained only as an upper limit ($M_b \le 2.03\,M_\oplus$) while c is detected with $M_c ≈ 18.1\,M_\oplus$ and $R_c ≈ 1.874\,R_\oplus$, implying Earth-like interior structure for the latter. Transit data are pivotal to constrain periods and epochs and enable robust planetary retrievals in activity-dominated RVs; TOI-2322 provides a valuable benchmark for activity-correction methods and motivates future near-infrared follow-up to probe wavelength-dependent activity signals and planetary atmospheres.

Abstract

Context. Active regions on the stellar surface can induce quasi-periodic radial velocity (RV) variations that can mimic planets and mask true planetary signals. These spurious signals can be problematic for RV surveys such as those carried out by the ESPRESSO consortium. Aims. Using ESPRESSO and HARPS RVs and activity indicators, we aim to confirm and characterize two candidate transiting planets from TESS orbiting a K4 star with strong activity signals. Methods. From the ESPRESSO FWHM, TESS photometry, and ASAS-SN photometry, we measure a stellar rotation period of 21.28 $\pm$ 0.08 d. We jointly model the TESS photometry, ESPRESSO and HARPS RVs, and activity indicators, applying a multivariate Gaussian Process (GP) framework to the spectroscopic data. Results. We are able to disentangle the planetary and activity components, finding that TOI-2322 b has a $11.307170^{+0.000085}_{-0.000079}$ d period, close to the first harmonic of the rotation period, a $\leq 2.03 M_\oplus$ mass upper limit and a $0.994^{+0.057}_{-0.059}$ $\mathrm{R_\oplus}$ radius. TOI-2322 c orbits close to the stellar rotation period, with a $20.225528^{+0.000039}_{-0.000044}$ d period; it has a $18.10^{+4.34}_{-5.36}$ $\mathrm{M_\oplus}$ mass and a $1.874^{+0.066}_{-0.057}$ $\mathrm{R_\oplus}$ radius. Conclusions. The multivariate GP framework is crucial to separating the stellar and planetary signals, significantly outperforming a one-dimensional GP. Likewise, the transit data is fundamental to constraining the periods and epochs, enabling the retrieval of the planetary signals in the RVs. The internal structure of TOI-2322 c is very similar to that of Earth, making it one of the most massive planets with an Earth-like composition known.

TOI-2322: two transiting rocky planets close to the stellar rotation period and its first harmonic

TL;DR

Stellar activity can mimic or obscure small exoplanets in RV data, especially when planetary periods lie near the stellar rotation period. The authors combine ESPRESSO and HARPS RVs with TESS transits and ground-based photometry, modeling the data with both a one-dimensional GP and a multivariate GP that links RVs to activity indicators (FWHM and BIS) to separate planetary signals from stellar activity. They confirm two transiting planets around TOI-2322: TOI-2322 b with P ≈ 11.31 d and TOI-2322 c with P ≈ 20.23 d; b remains mass-constrained only as an upper limit () while c is detected with and , implying Earth-like interior structure for the latter. Transit data are pivotal to constrain periods and epochs and enable robust planetary retrievals in activity-dominated RVs; TOI-2322 provides a valuable benchmark for activity-correction methods and motivates future near-infrared follow-up to probe wavelength-dependent activity signals and planetary atmospheres.

Abstract

Context. Active regions on the stellar surface can induce quasi-periodic radial velocity (RV) variations that can mimic planets and mask true planetary signals. These spurious signals can be problematic for RV surveys such as those carried out by the ESPRESSO consortium. Aims. Using ESPRESSO and HARPS RVs and activity indicators, we aim to confirm and characterize two candidate transiting planets from TESS orbiting a K4 star with strong activity signals. Methods. From the ESPRESSO FWHM, TESS photometry, and ASAS-SN photometry, we measure a stellar rotation period of 21.28 0.08 d. We jointly model the TESS photometry, ESPRESSO and HARPS RVs, and activity indicators, applying a multivariate Gaussian Process (GP) framework to the spectroscopic data. Results. We are able to disentangle the planetary and activity components, finding that TOI-2322 b has a d period, close to the first harmonic of the rotation period, a mass upper limit and a radius. TOI-2322 c orbits close to the stellar rotation period, with a d period; it has a mass and a radius. Conclusions. The multivariate GP framework is crucial to separating the stellar and planetary signals, significantly outperforming a one-dimensional GP. Likewise, the transit data is fundamental to constraining the periods and epochs, enabling the retrieval of the planetary signals in the RVs. The internal structure of TOI-2322 c is very similar to that of Earth, making it one of the most massive planets with an Earth-like composition known.

Paper Structure

This paper contains 26 sections, 2 equations, 20 figures, 9 tables.

Figures (20)

  • Figure 1: TESS target pixel file for TOI-2322 for sector 27. The target star is labelled as 1 and marked by a white cross. All sources from the Gaia DR3 catalogue down to a magnitude contrast of 9 are shown as red circles, with the size proportional to the contrast. The SPOC pipeline aperture is overplotted in shaded red squares.
  • Figure 2: GLS periodograms of the combined ESPRESSO and HARPS RVs (top) and activity indicators (second to bottom: CCF FWHM, CCF contrast, CCF bisector, $\mathrm{H_\alpha}$, $\mathrm{\log R'_{HK}}$, and Na I). The vertical green and purple lines indicate the periods of TOI-2322.02 and TOI-2322.01 respectively. The dotted, dashed, and solid horizontal grey lines indicate the 10%, 1%, and 0.1% FAP levels respectively.
  • Figure 3: Contrast curve and speckle auto-correlation function from the HRCam at SOAR for TOI-2322. The cyan points and solid line indicate the $5\sigma$ contrast curve; the inset shows the speckle auto-correlation function. No nearby sources are detected.
  • Figure 4: GLS periodogram of the TESS PDCSAP light curves (top) and SAP light curves (bottom). The vertical green and purple lines indicate the periods of TOI-2322.02 and TOI-2322.01 respectively, while the vertical red line indicates the period of the highest power seen in the FWHM periodogram. The dotted horizontal grey line indicates the 0.1% FAP level.
  • Figure 5: Light curve (top, separated by camera in colours, and 1-day binned data in black) and GLS periodogram (bottom) of the ASAS-SN g-band photometry. The vertical green and purple lines indicate the periods of TOI-2322.02 and TOI-2322.01 respectively, while the vertical red line indicates the period of highest power seen in the FWHM periodogram. The dotted, dashed, and solid horizontal grey lines indicate the 10%, 1%, and 0.1% FAP levels respectively.
  • ...and 15 more figures