Searching for close-in planets around TWA 7 with SPIRou

Abstract: We outline in this paper observations of the young pre-main-sequence low-mass star TWA 7, hosting a debris disk and a distant planet. Using data collected with the near-infrared SPIRou spectropolarimeter / precision velocimeter at the Canada-France-Hawaii Telescope from early 2019 to mid 2021, we detected the magnetic field of TWA 7 from the circularly polarized Zeeman signatures and Zeeman broadening of atomic spectral lines, and the rotational modulation of its longitudinal component at the known stellar rotation period (of 5.012+-0.007 d). We then modeled the large-scale and small-scale magnetic properties of TWA 7 using Zeeman-Doppler imaging. We found that TWA 7 hosted a mainly poloidal field that significantly evolved from 2019 to 2021, the dipole component getting stronger (increasing from 0.5 kG in 2019 to 0.7 kG in 2021) and less inclined to the stellar rotation axis (from 22° in 2019 to 15° in 2021). We also analyzed the radial velocities of TWA 7 derived from the SPIRou data, and found a tentative planet signature at a period of 15.2 d (with aliases at 20.8 and 30.4 d), very close to the detection limit of our data and that would correspond to a 0.17 Mjup planet at a distance of 0.09 au if confirmed. We finally report modulation of the 1083 nm He I and 1282 nm Pa-beta lines of TWA 7 with a period of 6.6 d, different from the rotation period and potentially hinting at the presence of a close-in planet triggering star-planet interactions.

• The paper presents a detailed SPIRou survey that uncovers strong, evolving magnetic fields in TWA 7, exhibited by clear Zeeman signatures and periodic modulation.
• It employs high-cadence near-infrared spectropolarimetry and radial velocity analysis to refine stellar parameters and tentatively detect a close-in mini-Saturn.
• The study highlights challenges in decoupling planet-induced signals from stellar activity and underscores the need for multi-epoch, multi-method observations.

Comprehensive SPIRou Survey of Close-in Planets and Magnetism in TWA 7

Introduction and Scientific Context

The paper presents a detailed spectropolarimetric and radial velocity (RV) survey of TWA 7, a 10 Myr, pre-main sequence M2 star in the TW Hydrae association with a resolved debris disk and a recently imaged distant planet at $\sim$52 au. The objectives are (i) to characterize the star’s magnetic field at both large and small scales, (ii) to refine stellar and disk parameters, and (iii) to search for evidence of close-in planets using near-infrared (NIR) precision velocimetry with SPIRou at CFHT. The system serves as an archetype for studying planet formation and disk-planet interactions in young, low-mass stellar systems that exhibit prominent magnetic activity and dynamic evolutionary processes.

Observational Strategy and Stellar Parameter Determination

A high-cadence series of 51 NIR spectropolarimetric sequences were acquired from early 2019 to mid 2021, sampling three distinct observing seasons. The longitudinal field was robustly detected at all epochs from circular polarization and Zeeman broadening. Dedicated analysis using the ZeeTurbo code provided key physical parameters: $T_{\rm eff}=3435\pm60$ K, $\log g=4.28\pm0.10$, [Fe/H]$=0.21\pm0.10$, $R_\star=0.92\pm0.12\,R_\odot$, and $M_\star\sim0.46\,M_\odot$, confirming TWA 7’s location on the PMS, fully convective branch. The projected stellar spin axis aligns with a near face-on ($i\approx13^\circ$) debris disk, indicated by the resolved imaging. The rotation period, refined from both magnetic and temperature proxies, is $5.012\pm0.007$ d—fully consistent with independent photometric estimates.

Magnetic Field Characterization

Zeeman Diagnostics and Variability

Both Stokes I (broadening) and Stokes V (Zeeman signatures in polarization) exhibit strong, temporally variable signatures. The mean small-scale field, $\langle B \rangle=3.2\pm0.3$ kG (ZeeTurbo), exceeds previously published values for TWA 7 and is indicative of substantial local photospheric magnetism, typical among very active, fully convective PMS M dwarfs [Kochukhov 2021, (2511.16609)].

The longitudinal, large-scale field ($B_\ell$) showed unambiguous seasonal and rotational modulation with semi-amplitudes $\sim50$ G and mean values rising from $-34$ G (2019) to $-205$ G (2021). This is notable given the star’s low $v \sin i$ and nearly pole-on geometry, which generally suppresses variability in integrated longitudinal measurements.

Zeeman-Doppler Imaging (ZDI)

The ZDI technique reveals an axisymmetric, predominantly poloidal topology dominated by the dipole, which strengthens and becomes more aligned with the rotation axis over the observed seasons ($B_{\rm dipole}$ increases from 0.48 kG to 0.69 kG, inclination to axis decreases from $22^\circ$ to $15^\circ$). The toroidal component is weak but exhibits a polarity reversal, providing evidence of a potentially cyclic field evolution analogous to solar/stellar dynamo cycles but at much higher field strengths and earlier evolutionary stages [Brun & Browning 2017]. The small-scale field inferred from ZDI ($\sim$2.2 kG) is below the ZeeTurbo estimate, a divergence likely due to ZDI’s inherent sensitivity limitations for strong, small-scale components in slowly rotating, pole-on targets.

Radial Velocity Analysis and Candidate Planet Detection

RV Modeling and Activity Filtering

The collected SPIRou RVs exhibit low-amplitude rotational modulation and prominent activity-driven jitter. Using a quasi-periodic Gaussian process regression framework, the authors isolated residual periodicities post-activity filtering. The most significant peak is observed at $15.21\pm0.02$ d, with additional power at 20.8 and 30.4 d (window function and alias correlations carefully analyzed).

The favored interpretation is the tentative detection of a mini-Saturn ($M_p=0.17_{-0.03}^{+0.04}\,M_{\rm Jup}$) at $0.092\pm0.006$ au, whose RV semi-amplitude is $K=5.4_{-1.0}^{+1.3}$ m/s and whose inferred orbital inclination matches the disk. The Bayes factor for this signal reaches 11.4 and residuals are significantly reduced compared to activity-only models. However, its position near harmonics of both the sampling window and integral multiples of the stellar rotation period undermines robust confirmation. Complementary injection-recovery experiments show the detection is at the survey’s sensitivity threshold. Further physical interpretation is deferred pending additional, denser phase coverage and expanded multi-wavelength datasets.

Chromospheric Activity Proxies and Star-Planet Interactions

Analysis of NIR activity indicators (He I 1083 nm and Pa β 1282 nm) reveals a periodicity at 6.6 d, which is not present in the stellar rotation, RV, or temperature datasets. This is distinct from established accretion- or flare-related variability and is suggestive of periodic star-planet magnetic or magnetospheric interaction effects, potentially related to one of the planet RV aliases or the planet’s own rotation if tidally synchronized. Such a signal’s amplitude and coherence, coupled with its appearance only in specific line proxies, warrant further investigation via simultaneous multi-band monitoring and theoretical modeling of possible star-planet interaction regimes (e.g., sub-Alfvénic wind conditions at the candidate planet’s location [Vidotto et al. 2013]).

Implications for PMS Stellar Magnetism and Planet Detection

These results reinforce several salient conclusions about young, active, low-mass stars:

• Magnetic field evolution: The combination of intense, variable small- and large-scale surface magnetism and evidence for evolving field geometry supports dynamo models that predict cycles and rapid topology changes in fully convective PMS M dwarfs [Yadav et al. 2015; Shulyak et al. 2015].
• Planet formation and migration constraints: The tentative detection of a close-in, low-mass giant in the presence of a distant directly imaged planet and structured debris disk provides new constraints on core accretion and migration models, particularly those involving magnetospheric cavities, disk trapping, and early pile-up of close-in planets [Romanova et al. 2019; Lau et al. 2024].
• Stellar activity and planet detectability: The paper demonstrates the efficacy and limitations of current NIR RV and spectropolarimetric techniques for exoplanet detection in strongly active young stars: while activity filtering and multiple proxies significantly improve sensitivity, RV diagnostics remain fundamentally limited by instrumental systematics, sample cadence, and aliasing.
• Unexplained periodicities and interactions: The discovery of a strong, non-rotational periodic activity signal in He I and Pa β represents a provocative clue about star-planet magnetic/atmospheric coupling or alternative activity cycles in young stars.

Conclusion

This study yields a comprehensive magnetic field and RV census of TWA 7, identifying strong, evolving magnetism comparable to other active, young M dwarfs and presenting a candidate for a close-in planet with parameters near the survey's sensitivity floor. The work underscores the complexity of disentangling planetary signatures from magnetic activity in the PMS regime and highlights the critical need for multi-epoch, multi-modal spectropolarimetric and RV monitoring to (a) constrain the frequency and nature of close-in planets in young systems, (b) elucidate the temporal evolution of dynamo-generated fields, and (c) probe possible star-planet interaction mechanisms. The ongoing characterization of TWA 7 sets a benchmark for comparative planetology and stellar magnetism studies in the early stages of low-mass stellar evolution (2511.16609).