This dataset contains corrected particle number concentration data measured during the year long MOSAiC expedition from October 2019 to September 2020. The measurement was performed in the Swiss aerosol container on the D-deck of Research Vessel Polarstern using an inlet with an upper cut-off of 1 µm (called also interstitial inlet). Data were collected by a Condensation Particle Counter (CPC) model 3025 (TSI Inc.) in 10 s resolution with a lower cut-off diameter of 3 nm. More detailed information about its position and the measurement setup can be found in (Beck et al., 2022). The data are cleaned from calibration and zero-check filter periods and the time-stamps of all data points was synchronized to its closest 10 s interval. Some periods of the measurements were corrected for a step in the signal (see file description).Participation of the Swiss Container was co-financed by the Swiss Polar Institute and University of Helsinki.

Some periods of the measurements were affected by repeated step changes in the particle number concentrations. This was caused by a valve, which was positioned in the line behind (downstream) the CPC and switched between a total air inlet and an interstitial inlet. During odd hours (i.e. 1 am, 3 am etc.) the valve was open to the interstitial inlet, during even hours to the total inlet. During the affected periods (less than ten periods with a duration of less than 12 hours, except for a three-day period from Dec. 08, 2019 to Dec. 10, 2019), the measured particle number concentrations behind the interstitial inlet are lower compared to the previous and the following hour behind the total inlet. To correct for these step changes in concentration, we calculated two correction factors, i.e. at the beginning and at the end of the affected period. The two correction factors are derived by dividing the median particle number concentrations of 3 minutes before (after) the start (end) of the affected period by the median particle number concentration of 3 minutes after (before) the start (end) of the affected period. Each data point of the affected period was then corrected by multiplying its particle number concentration by the linearly interpolated correction factor at the corresponding timestamp. The corrected data set may still contain minor artefacts related to the step change correction, i.e. small deviations of the number concentration in the order of < 10 %. The counting uncertainty of a CPC is nominally 10 %. Depending on how the dataset is used this might be relevant or not. The dataset has a time resolution of 10 sec. More detailed information about the measurement setup can be found in (Beck et al., in prep.). The data columns include Time, Latitude, Longitude, Particle number concentration and a correction flag (0=uncorrected, 1 = corrected).We used 10 sec time resolution positional data from the following datasets: Rex, M (2020), Haas, C (2020), Kanzow, T (2020), Rex, M (2021a,b).