Seasonal trends of Ice Nucleating Particles at Ny-Ålesund: a study of condensation-freezing by the Dynamic Filter Processing Chamber

Abstract. This study presents atmospheric ice nucleating particle (INP) data from the Gruvebadet (GVB) observatory in Ny-Ålesund (Svalbard). Aerosol particle sampling activities were conducted over three years (2018–2020), for a total of 6 intensive campaigns, covering three seasons (spring, summer and autumn). Ambient INP concentrations (nINP) were measured offline on the collected filters, in condensation freezing mode (water saturation ratio of 1.02), by means of the Dynamic Filter Processing Chamber (DFPC). Three activation temperatures (Ts) were considered: -15, -18 and -22 °C.

Overall, in the PM₁₀ size range, DFPC-measured nINP ranged from 0.3 to 315 m^-3 in the considered T range, in agreement with previous observations in the Arctic environment. Regarding the ice-nucleation efficiency of the investigated aerosol particles (referring to the range between 0.1 and 10 µm), the estimated activated fraction (AF) resulted between 10^-8 and 10^-5, obviously increasing as the T decreases.

The seasonality of the ice nucleating properties of Arctic aerosol particles was investigated by merging the results of the 6 campaigns. Our data show a moderate summertime increase of nINP at T = -15 °C. No such summertime increase was observed at T = -18 and -22 °C. On the other hand, the AF of atmospheric aerosol particles presents a clearer seasonal evolution, with maxima observed in late summer and early autumn. Finally, we report a marked seasonal evolution in the contribution of super-micrometer INPs. Coarse INPs increase significantly their contribution from spring (15–20 %) to summer (~60 %), while lower levels typically characterize the autumn season (20–50 %). Our calculations also show that coarse particles have at least two orders of magnitude higher AF compared to sub-micrometre ones.

The correlation with anthropogenic long range transport tracer black carbon, the contribution of ground types inferred from satellite data, the low-traveling back trajectory analysis and the aforementioned considerations regarding the varying seasonal contributions of sub- and super-micrometre INPs all indicate that the primary sources of springtime INPs at GVB are mostly located outside the Arctic. In contrast, local INP sources dominate during summer and early autumn. When land and sea are mostly free from snow and ice, both marine and terrestrial sources result important INP contributors at GVB. Regarding marine sources in particular, our analysis identifies potential marine INP sources located in the seawaters surrounding and immediately to the South of the Svalbard archipelago down to the waters around Iceland. Such sources apparently dominate nINP in summer and early autumn outside the major terrestrial INP bursts.