Magnetic field measurements in a sample of Class I and flat-spectrum protostars observed with SPIRou

Context. Magnetic fields play a crucial role throughout stellar evolution in regulating angular momentum, channelling accretion, and launching jets and outflows. While the magnetic properties of classical T Tauri stars (CTTSs) have been extensively characterised, those of their progenitors, Class I and flat-spectrum (FS) protostars, remain poorly constrained due to their embedded nature, which provides an observational challenge. Aims. Our aim was to detect and characterise the large-scale magnetic fields in a sample of Class I and FS protostars. These young stars are expected to host strong magnetic fields generated by dynamo processes in their largely convective interiors. Methods. We have used SPIRou, a high-resolution spectropolarimeter working in the near-infrared domain, to analyse the polarised light of Class I and FS protostars. We used the least-squares deconvolution (LSD) technique to perform the magnetic analysis and measure the longitudinal magnetic fields from circularly polarised Stokes V profiles. Results. We report new large-scale magnetic field detections in five FS protostars. Including the previous detection of the large-scale magnetic-field in the V347 Aur FS-protostar, 40% of our final sample of 15 protostars is confirmed to be magnetic. These magnetic stars show clear Zeeman signatures, with longitudinal field strengths ranging from ∼80 to ∼200 G in absolute value. The remaining stars exhibit no detectable Stokes V signature, but the estimated upper limits on a hidden dipolar field range from 500 G to more than 5 kG. For stars in which no magnetic fields are detected, it is still conceivable that a magnetic field exists, but is intrinsically weak, highly complex and dominated by small-scale structures, or cancelled out in integrated spectropolarimetric signals due to opposing polarities. Conclusions. We show that Class I and FS protostars can host large-scale magnetic fields with strengths that may be weaker than in more evolved CTTSs. This supports the idea that magnetic processes are already active during the main accretion phase and may influence star–disk interactions from the earliest stages.