Blobs, spiral arms, and a possible planet around HD 169142

Context. Young planets are expected to cause cavities, spirals, and kinematic perturbations in protostellar disks that may be used to infer their presence. However, a clear detection of still-forming planets embedded within gas-rich disks is still rare. Aims. HD 169142 is a very young Herbig Ae-Be star surrounded by a pre-transitional disk, composed of at least three rings. While claims of sub-stellar objects around this star have been made previously, follow-up studies remain inconclusive. The complex structure of this disk is not yet well understood. Methods. We used the high contrast imager SPHERE at ESO Very large Telescope to obtain a sequence of high-resolution, high-contrast images of the immediate surroundings of this star over about three years in the wavelength range 0.95-2.25 μm. This enables a photometric and astrometric analysis of the structures in the disk. Results. While we were unable to definitively confirm the previous claims of a massive sub-stellar object at 0.1-0.15 arcsec from the star, we found both spirals and blobs within the disk. The spiral pattern may be explained as due to the presence of a primary, a secondary, and a tertiary arm excited by a planet of a few Jupiter masses lying along the primary arm, likely in the cavities between the rings. The blobs orbit the star consistently with Keplerian motion, allowing a dynamical determination of the mass of the star. While most of these blobs are located within the rings, we found that one of them lies in the cavity between the rings, along the primary arm of the spiral design. Conclusions. This blob might be due to a planet that might also be responsible for the spiral pattern observed within the rings and for the cavity between the two rings. The planet itself is not detected at short wavelengths, where we only see a dust cloud illuminated by stellar light, but the planetary photosphere might be responsible for the emission observed in the K1 and K2 bands. The mass ofthis putative planet may be constrained using photometric and dynamical arguments. While uncertainties are large, the mass should be between 1 and 4 Jupiter masses. The brightest blobs are found at the 1:2 resonance with this putative planet.