We present 197 planet candidates discovered using data from the first year of the NASA K2 mission (Campaigns 0-4), along with the results of an intensive program of photometric analyses, stellar spectroscopy, high-resolution imaging, and statistical validation. We distill these candidates into sets of 104 validated planets (57 in multi-planet systems), 30 false positives, and 63 remaining candidates. Our validated systems span a range of properties, with median values of R P = 2.3 R ⊕ , P = 8.6 d, T eff = 5300 K, and Kp = 12.7 mag. Stellar spectroscopy provides precise stellar and planetary parameters for most of these systems. We show that K2 has increased by 30% the number of small planets known to orbit moderately bright stars (1-4 R ⊕ , Kp = 9-13 mag). Of particular interest are 37 planets smaller than 2 R ⊕ , 15 orbiting stars brighter than Kp = 11.5 mag, five receiving Earth-like irradiation levels, and several multi-planet systems -including four planets orbiting the M dwarf K2-72 near mean-motion resonances. By quantifying the likelihood that each candidate is a planet we demonstrate that our candidate sample has an overall false positive rate of 15 − 30%, with rates substantially lower for small candidates (< 2R ⊕ ) and larger for candidates with radii > 8R ⊕ and/or with P < 3 d. Extrapolation of the current planetary yield suggests that K2 will discover between 500 − 1000 planets in its planned four-year mission -assuming sufficient follow-up resources are available. Efficient observing and analysis, together with an organized and coherent follow-up strategy, is essential to maximize the efficacy of planet-validation efforts for K2 , TESS , and future large-scale surveys. 1 We distinguish "confirmed" systems (with measured masses) from "validated" systems (whose planetary nature has been statistically demonstrated, e.g. with false positive probability < 1% ).

Context. Dust is expected to be ubiquitous in extrasolar planetary systems owing to the dynamical activity of minor bodies. Inner dust populations are, however, still poorly known because of the high contrast and small angular separation with respect to their host star, and yet, a proper characterisation of exozodiacal dust is mandatory for the design of future Earth-like planet imaging missions. Aims. We aim to determine the level of near-infrared exozodiacal dust emission around a sample of 42 nearby main sequence stars with spectral types ranging from A to K and to investigate its correlation with various stellar parameters and with the presence of cold dust belts. Methods. We use high-precision K-band visibilities obtained with the FLUOR interferometer on the shortest baseline of the CHARA array. The calibrated visibilities are compared with the expected visibility of the stellar photosphere to assess whether there is an additional, fully resolved circumstellar emission source. Results. Near-infrared circumstellar emission amounting to about 1% of the stellar flux is detected around 13 of our 42 target stars. Follow-up observations showed that one of them (eps Cep) is associated with a stellar companion, while another one was detected around what turned out to be a giant star (kap CrB). The remaining 11 excesses found around single main sequence stars are most probably associated with hot circumstellar dust, yielding an overall occurrence rate of 28 +8 −6 % for our (biased) sample. We show that the occurrence rate of bright exozodiacal discs correlates with spectral type, K-band excesses being more frequent around A-type stars. It also correlates with the presence of detectable far-infrared excess emission in the case of solar-type stars. Conclusions. This study provides new insight into the phenomenon of bright exozodiacal discs, showing that hot dust populations are probably linked to outer dust reservoirs in the case of solar-type stars. For A-type stars, no clear conclusion can be made regarding the origin of the detected near-infrared excesses.