Calibration and performance of the NIKA2 camera at the IRAM 30-m Telescope

Context. NIKA2 is a dual-band millimetre continuum camera of 2 900 kinetic inductance detectors, operating at 150 and 260 GHz, installed at the IRAM 30-m telescope in Spain. Open to the scientific community since October 2017, NIKA2 will provide key observations for the next decade to address a wide range of open questions in astrophysics and cosmology. Aims. Our aim is to present the calibration method and the performance assessment of NIKA2 after one year of observation. Methods. We used a large data set acquired between January 2017 and February 2018 including observations of primary and secondary calibrators and faint sources that span the whole range of observing elevations and atmospheric conditions encountered by the IRAM 30-m telescope. This allowed us to test the stability of the performance parameters against time evolution and observing conditions. We describe a standard calibration method, referred to as the “Baseline” method, to translate raw data into flux density measurements. This includes the determination of the detector positions in the sky, the selection of the detectors, the measurement of the beam pattern, the estimation of the atmospheric opacity, the calibration of absolute flux density scale, the flat fielding, and the photometry. We assessed the robustness of the performance results using the Baseline method against systematic effects by comparing results using alternative methods. Results. We report an instantaneous field of view of 6.5′ in diameter, filled with an average fraction of 84%, and 90% of valid detectors at 150 and 260 GHz, respectively. The beam pattern is characterised by a FWHM of 17.6″ ± 0.1″ and 11.1″ ± 0.2″, and a main-beam efficiency of 47%±3%, and 64%±3% at 150 and 260 GHz, respectively. The point-source rms calibration uncertainties are about 3% at 150 GHz and 6% at 260 GHz. This demonstrates the accuracy of the methods that we deployed to correct for atmospheric attenuation. The absolute calibration uncertainties are of 5%, and the systematic calibration uncertainties evaluated at the IRAM 30-m reference Winter observing conditions are below 1% in both channels. The noise equivalent flux density at 150 and 260 GHz are of 9 ± 1 mJy s1/2 and 30 ± 3 mJy s1/2. This state-of-the-art performance confers NIKA2 with mapping speeds of 1388 ± 174 and 111 ± 11 arcmin2 mJy−2 h−1 at 150 and 260 GHz. Conclusions. With these unique capabilities of fast dual-band mapping at high (better that 18″) angular resolution, NIKA2 is providing an unprecedented view of the millimetre Universe.