SPIRou Spectrograph: Exoplanet & Magnetism

• SPIRou spectrograph is a high-resolution, fiber-fed near-infrared instrument that combines spectropolarimetry and precision velocimetry to detect exoplanets and study stellar magnetism.
• Its optimized design features a cryogenic spectrograph, high-throughput fluoride fibers, and advanced pupil slicing, achieving R >70,000 and sub-meter-per-second radial velocity stability.
• The integrated APERO pipeline and robust calibration unit facilitate detailed analysis of stellar activity, accretion phenomena, and atmospheric characterization of exoplanets.

The SPIRou spectrograph is a high-resolution, fiber-fed, near-infrared (nIR) spectropolarimeter and precision velocimeter mounted at the 3.6-m Canada-France-Hawaii Telescope (CFHT), engineered to address forefront research on exoplanet detection around cool stars and the role of magnetism in star and planet formation. Covering the YJHK bands (0.98–2.44 μm) at a resolving power R ≃ 70,000–75,000, SPIRou uniquely integrates broad simultaneous spectral coverage, ultra-stable instrumental design, and full (circular and linear) spectropolarimetric capability. Its science drivers include the detection and characterization of Earth-like planets orbiting low-mass stars (particularly M dwarfs), the analysis of magnetic field geometries in young stars, and the investigation of accretion, disk, and ejection phenomena in star-forming environments. SPIRou serves as a cornerstone for the SPIRou Legacy Survey (SLS), the largest CFHT program to date, and provides vital synergy with major ground- and space-based facilities in planetary and stellar astrophysics.

1. Instrumental Architecture and Performance

SPIRou’s optical path and mechanical architecture are optimized to deliver both high-throughput and sub-meter-per-second radial velocity (RV) precision, alongside robust polarimetric sensitivity. The instrument’s components are as follows:

• Cassegrain Unit: This front-end hosts an atmospheric dispersion corrector (ADC) ensuring negligible chromatic refraction (≤0.03″ up to airmass 2.5), a tip-tilt module stabilizing stellar images (<0.05″ rms), and an achromatic polarimeter comprising dual Fresnel rhombs and a Wollaston prism to separate orthogonal polarization states. The focal reducer adapts the f/8 telescope beam to f/4 for optimal fiber coupling.
• Fiber Link and Pupil Slicer: Light is delivered via high-transmission fluoride fibers (ZrF₄, ≥90% over 0.98–2.4 μm) to a cryogenic bench, then split into pupil slices that preserve the effective spectral resolution, with slicing throughput >90%. Octagonal fibers deliver a scrambling gain ≳1000 to minimize modal noise and illumination instabilities.
• Cryogenic Spectrograph: Housed at 80 K with 2 mK thermal stability, the optical train uses a 150 mm R2 echelle grating (23.2 gr/mm), a three-prism cross-disperser (ZnSe, Infrasil, SiO₂), and a 5-lens dioptric camera to project the full YJHK spectrum onto a Hawaii 4RG-15 HgCdTe 4096² 15 μm detector.
• Spectral Coverage and Resolution: Simultaneous coverage across 0.98–2.44 μm in one shot, with resolving power R > 70,000, velocity scale ~2.28 km/s per pixel, and effective PSF FWHM ~4 km/s.
• RV Precision and Throughput: Baseline internal RV precision is ≤1 m/s, supported by a 15% avg. end-to-end throughput. Thermal and mechanical control yields instrumental drifts <0.7 m/s nightly, reduced to <0.25 m/s with calibration.
• Calibration Unit: Multiple sources (white lamp, UNe hollow-cathode, and a Fabry–Pérot etalon) provide wavelength calibration and simultaneous drift monitoring. The FP module, with measured finesse ~12.8 and a vacuum-sealed enclosure held at a few mK, delivers a dense, stable comb of lines over 980–2350 nm, enabling drift corrections at the <0.3 m/s level (Cersullo et al., 2017, Boisse et al., 2016).

Key formulas characterizing performance: $$R = \frac{\lambda}{\Delta\lambda}, \qquad \Delta v = c \frac{\Delta \lambda}{\lambda} = \frac{c}{R}$$

2. Data Processing and Calibration

The entire data flow is automated through the APERO pipeline (Cook et al., 2022), architected for robust, reproducible nIR reduction and polarimetric extraction:

• Preprocessing: Removal of detector artifacts from raw H4RG ramp images, registration, gain correction, and bad-pixel mapping.
• Calibration: A “reference night” database is built for darks, order mapping, leak maps (calibrating cross-talk), and affine registration of order geometry via hundreds of thousands of FP peaks.
• Optimal Extraction: Spectra are rectified, order-wise extracted (with cosmic-ray rejection), and flat/blaze corrected; background subtraction, and cross-talk corrections are routinely applied.
• Wavelength Solution: A dual-calibrator approach (hollow-cathode for absolute, FP for sampling density) achieves ~0.15 m/s internal errors and ensures cross-order and inter-night stability (Hobson et al., 2021). Quality checks include Littrow and polynomial fits between pixels and wavelengths.
• Telluric Correction: A hybrid model/empirical approach uses TAPAS models, stellar templates, and a library of hot-star spectra to iteratively remove H₂O and “dry” absorption; cross-correlation and PCA reconstruct residuals with high fidelity.
• Polarimetry: Four-exposure sequences with quarter-wave rhomb positions enable full Stokes vector recovery (Q, U, V). LSD (least-squares deconvolution) coadds individual line profiles into high-S/N mean profiles for magnetic diagnostics.
• RV Extraction: Classical CCFs, a state-of-the-art line-by-line (LBL) algorithm (projecting spectrum-to-template residuals onto the template’s derivative per Bouchy et al. 2001), and principal component (wPCA) corrections (“wapiti”) are available. The LBL approach achieves precision at or below 2 m/s over multi-year baselines, especially after removing systematics with wPCA (Moutou et al., 2023, Ould-Elhkim et al., 10 Feb 2025).

3. Science Drivers and Survey Strategy

SPIRou’s design enables several major research programs:

• Exoplanet Detection via RV:
  • Focuses on habitable, terrestrial planets around M dwarfs, exploiting higher RV signal amplitude ($K$) induced by small, close-in orbits:

  $$K = \left(\frac{2\pi G}{P}\right)^{1/3} \frac{M_p \sin i}{M_*^{2/3}}$$ - The RV amplitude is thus maximized for low stellar mass $M_*$, short period $P$ (habitable zones of M dwarfs), and the deep nIR sensitivity enhances detection in optically faint regimes [(Santerne et al., 2013); (Donati et al., 2018)]. - The SPIRou Legacy Survey (SLS) (Donati et al., 2020) is allocated ≥300 nights for systematic blind searches, RV monitoring, and transit follow-up of ~70 nearby M dwarfs, with additional programs on brown dwarfs, massive stars, and planetary atmospheres.

• Magnetic Fields, Star and Planet Formation:

  • Spectropolarimetric sensitivity (Δλ ∝ Bλ²) enables detailed mapping of large-scale fields in young stellar objects (e.g., classical T Tauri stars, cTTS) and protostars, with Zeeman-Doppler Imaging (ZDI) reconstructing field topology and differential rotation (Cortes-Zuleta et al., 2023, Donati et al., 4 Mar 2024).
  • Simultaneous detection of forbidden atomic lines, molecular ro-vibrational transitions, and magnetically split profiles allows concurrent studies of accretion, wind, and jet processes across hot/cool disk regions (Carmona et al., 2013).
• Planetary Atmospheres and Formation:
  • High-dispersion, broad-band nIR spectroscopy permits the detection of weak molecular species (CO, H₂O, CO₂, OH, etc.) via cross-correlation against model templates, disentangling overlapping spectral features and enabling robust retrievals of volume mixing ratios and C/O, C/H, O/H ratios (Pelletier et al., 2021, Boucher et al., 2023).
• Stellar Fundamental Parameters:
  • SPIRou’s high-resolution nIR spectra, modeled with synthetic grids (PHOENIX, MARCS), yield $T_{\mathrm{eff}}$, $\log g$, [M/H] for M dwarfs with typical internal uncertainties of 30 K, 0.05 dex, and 0.1 dex (though systematic differences up to 0.4 dex in $\log g$ and [M/H], and 30–50 K in $T_{\mathrm{eff}}$, are observed based on model choice) (Cristofari et al., 2021).

4. Methodologies and Analytical Frameworks

SPIRou catalyzes the development of advanced analysis frameworks:

• Radial Velocity Measurement:
  • In the LBL method, a high-S/N template is constructed from median-combined data; per-line RVs are computed using first derivatives, then aggregated. Weighted principal component analysis (wapiti) removes residual systematics from RVs, with Bayesian model selection to evaluate signal significance (Ould-Elhkim et al., 10 Feb 2025).
  • RV precision is shown to reach a noise floor of ~0.7 m/s on bright targets after binning; photon/mode noise dominates at this limit (Moutou et al., 2023).
• Gaussian Process Regression:
  • Activity-induced RV jitter is modeled via quasi-periodic GP kernels,

  $$k(\tau) = A\,\exp\left[ - \frac{\tau^2}{2\lambda^2} - \Gamma \sin^2\left(\frac{\pi \tau}{P_\mathrm{rot}}\right) \right]$$ - Ancillary indicators (e.g., Hα, NaD1, dLW, dET) are co-modeled to constrain hyperparameters and isolate planetary signals (Cortes-Zuleta et al., 2023, Ould-Elhkim et al., 10 Feb 2025).

• Spectral and Polarimetric Line Diagnostics:
  • LSD and PCA methods are applied to disentangle Zeeman and telluric features.
  • The longitudinal magnetic field is calculated as:

  $$B_\ell = -2.14 \times 10^{11} \frac{ \int v V(v) dv } { \lambda_0 g_{eff} c \int [I_c - I(v)] dv }$$ - ZDI reconstructs surface magnetic maps with spherical harmonics.

• Atmospheric Retrievals for Exoplanets:
  • High-resolution transmission and emission spectra are analyzed via weighted cross-correlation:

  $$\mathrm{CCF}(v) = \sum_{i=1}^N \frac{f_i\,m_i(v)}{\sigma_i^2}$$ - Likelihoods are calculated using log-χ²-based relations, and joint inference across SPIRou, HST, and Spitzer data is performed to constrain C/H, C/O, and O/H ratios, distinguishing CO from CO₂ absorption at the 2.3 μm band (Boucher et al., 2023, Pelletier et al., 2021).

5. Scientific Results and Examples

SPIRou’s early science and SLS have delivered key results:

• Exoplanet Discovery: Secure detection of an 8.8 ± 0.7 M⊕ planet in a 9.5537-day orbit around Gl 480, with tentative secondary signals (6.4 d) sensitive to activity modeling (Ould-Elhkim et al., 10 Feb 2025). No planets detected in Gl 382 (RV dominated by activity).
• Multiplanet Systems: SPIRou’s RVs and ZDI maps refined orbital models of GJ 876 (Laplace and chaotic resonance) and GJ 1148, and mapped a ~30 G dipole in GJ 876 (Moutou et al., 2023).
• Atmospheric Characterization: In τ Boo b, SPIRou measured super-solar C/H, O/H, C/O ratios, indicating formation beyond the water snowline in a CO-rich disk (Pelletier et al., 2021). In WASP-127 b, robust water detection and a sub-solar C/O ratio were found, discriminating CO₂ from CO atmospheric dominance and favoring oxygen-rich formation (Boucher et al., 2023).
• Stellar Activity and Magnetic Topology: SPIRou nIR spectropolarimetry enabled robust rotation period and differential rotation measurements, ZDI mapping of poloidal/toroidal field evolution, and demonstrated that nIR data can smooth out some spot contrast while offering enhanced Zeeman sensitivity (Cortes-Zuleta et al., 2023, Donati et al., 4 Mar 2024).
• Disk/Accretion Physics: Monitoring veiling in young stars (NIR continuum emission) revealed its correlation with accretion rate and disk inclination, and its variability over nightly and monthly timescales (Sousa et al., 2023).
• Star-Disc Interaction and Planet Formation: In CI Tau, SPIRou showed a stable, predominantly dipolar kG-level poloidal field channelling polar accretion, and established that a 23.86-day RV modulation in CO lines arises from a disk structure rather than a massive planet (Donati et al., 4 Mar 2024).

6. Synergies, Impact, and Future Directions

SPIRou is strategically positioned for maximum scientific leverage:

• Synergy with Space Missions: The nIR RV follow-up of transit candidates from TESS, CHEOPS, and PLATO, together with atmospheric measurements for targets expected to be observed with JWST and future ELTs [(Santerne et al., 2013); (Delfosse et al., 2013); (Donati et al., 2018)].
• Legacy Data: The SLS generates a comprehensive, homogeneous library of high-resolution nIR spectra and magnetic field maps, enabling statistical studies of exoplanet demographics and stellar magnetism (Donati et al., 2020).
• Methodological Innovations: Cross-dataset (nIR-optical) RV analyses, principal component and GP-based activity corrections, and multiwavelength spectropolarimetry improve sensitivity to Earth-mass planets and enable disentangling planetary and stellar activity signals (Ould-Elhkim et al., 10 Feb 2025).
• Integration and Operations: The modularity of APERO and systematic reference calibration ensure long-term stability and facilitate regular reprocessing and integration with future upgrades or instrumentation.
• Broader Astrophysics: SPIRou’s broad coverage and polarimetric sensitivity support studies ranging from stellar atmospheric composition, brown dwarf weather patterns, to magnetospheric accretion, making it a multipurpose facility for stellar, planetary, and disk science.

In summary, SPIRou exemplifies a new class of infrared spectropolarimeters that, through precise thermal/mechanical control, optimized nIR fiber and slicing technology, advanced polarimetry, and robust calibration and analysis pipelines, have enabled a generational advance in both exoplanet and stellar astrophysics. Its technical attributes and methodology, as realized in early SLS results and coordinated survey strategies, establish it as a benchmark instrument in the challenging nIR regime for the foreseeable future.