Hot Jupiters are giant Jupiter-like exoplanets that orbit their host stars 100 times more closely than Jupiter orbits the Sun. These planets presumably form in the outer part of the primordial disk from which both the central star and surrounding planets are born, then migrate inwards and yet avoid falling into their host star. It is, however, unclear whether this occurs early in the lives of hot Jupiters, when they are still embedded within protoplanetary disks, or later, once multiple planets are formed and interact. Although numerous hot Jupiters have been detected around mature Sun-like stars, their existence has not yet been firmly demonstrated for young stars, whose magnetic activity is so intense that it overshadows the radial velocity signal that close-in giant planets can induce. Here we report that the radial velocities of the young star V830 Tau exhibit a sine wave of period 4.93 days and semi-amplitude 75 metres per second, detected with a false-alarm probability of less than 0.03 per cent, after filtering out the magnetic activity plaguing the spectra. We find that this signal is unrelated to the 2.741-day rotation period of V830 Tau and we attribute it to the presence of a planet of mass 0.77 times that of Jupiter, orbiting at a distance of 0.057 astronomical units from the host star. Our result demonstrates that hot Jupiters can migrate inwards in less than two million years, probably as a result of planet–disk interactions.

We report results of an extended spectropolarimetric and photometric monitoring of the weakline T Tauri star V830 Tau and its recently-detected newborn close-in giant planet. Our observations, carried out within the MaTYSSE programme, were spread over 91 d, and involved the ESPaDOnS and Narval spectropolarimeters linked to the 3.6-m Canada-France-Hawaii, the 2m Bernard Lyot and the 8-m Gemini-North Telescopes. Using Zeeman-Doppler Imaging, we characterize the surface brightness distributions, magnetic topologies and surface differential rotation of V830 Tau at the time of our observations, and demonstrate that both distributions evolve with time beyond what is expected from differential rotation. We also report that near the end of our observations, V830 Tau triggered one major flare and two weaker precursors, showing up as enhanced red-shifted emission in multiple spectral activity proxies.With 3 different filtering techniques, we model the radial velocity (RV) activity jitter (of semi-amplitude 1.2 km s −1 ) that V830 Tau generates, successfully retrieve the 68 ± 11 m s −1 RV planet signal hiding behind the jitter, further confirm the existence of V830 Tau b and better characterize its orbital parameters. We find that the method based on Gaussian-process regression performs best thanks to its higher ability at modelling not only the activity jitter, but also its temporal evolution over the course of our observations, and succeeds at reproducing our RV data down to a rms precision of 35 m s −1 . Our result provides new observational constraints on scenarios of star / planet formation and demonstrates the scientific potential of large-scale searches for close-in giant planets around T Tauri stars.

We report the analysis, conducted as part of the MaTYSSE programme, of a spectropolarimetric monitoring of the ∼0.8 Myr, ∼1.4 M disc-less weak-line T Tauri star V410 Tau with the ESPaDOnS instrument at the Canada-France-Hawaii Telescope and NARVAL at the Télescope Bernard Lyot, between 2008 and 2016. With Zeeman-Doppler Imaging, we reconstruct the surface brightness and magnetic field of V410 Tau, and show that the star is heavily spotted and possesses a ∼550 G relatively toroidal magnetic field.We find that V410 Tau features a weak level of surface differential rotation between the equator and pole ∼5 times weaker than the solar differential rotation. The spectropolarimetric data exhibit intrinsic variability, beyond differential rotation, which points towards a dynamo-generated field rather than a fossil field. Long-term variations in the photometric data suggest that spots appear at increasing latitudes over the span of our dataset, implying that, if V410 Tau has a magnetic cycle, it would have a period of more than 8 years.Having derived raw radial velocities (RVs) from our spectra, we filter out the stellar activity jitter, modeled either from our Doppler maps or using Gaussian Process Regression. Thus filtered, our RVs exclude the presence of a hot Jupiter-mass companion below ∼0.1 au, which is suggestive that hot Jupiter formation may be inhibited by the early depletion of the circumstellar disc, which for V410 Tau may have been caused by the close (few tens of au) M dwarf stellar companion.