Evolution of the size and composition of ice nucleating particles within the synoptic context of the Arctic melt onset

The Arctic region is undergoing rapid changes caused by a warming climate and positive radiative feedback loops associated with rapidly-declining sea ice. Clouds play a key role in melt onset timing and annual extent of sea ice loss by modifying the amount of radiation that reaches the surface. The characteristics of Arctic mixed-phase clouds, including lifetime and partitioning of cloud particle phase, are sensitive to ice nucleating particles (INPs), a cloud-active aerosol capable of initiating freezing of cloud droplets at temperatures above homogeneous freezing (-38 °C). INP concentration and freezing temperature (T) are necessary parameters for modeling and validating observations of cloud ice. Observations of INPs are therefore critical for reducing the uncertainties associated with aerosol-cloud interactions for predicting the future Arctic climate. We present an overview of temperature-resolved INP concentrations observed during the ARTofMELT (Atmospheric rivers and the onset of sea ice melt) expedition from May-June 2023. Included in this overview are concentrations of INPs from total aerosol filters (collected continuously on the icebreaker Oden and directly on the sea ice downwind of open leads) and from size-resolved aerosol collected on Oden. Information regarding particle size is pertinent for revealing the aerosol populations acting as INPs and, alongside back-trajectory and meteorological data, their sources. The concentration of INPs from the Oden total aerosol filters reached a maximum of ~1 L^-1 at the coldest detectable temperature (-29 °C) and a minimum of ~0.0001 L^-1 at temperatures near -10 °C. The total aerosol filters deployed on the ice were less effective at detecting the warm-temperature (rarest) INPs due to the shorter sampling periods. However, an INP maximum of ~1000 L^-1 at T = -29 °C was reached on May 11 downwind of a lead, potentially due to wave breaking in strong winds. The total concentrations of INPs from the size-resolved samples were lower than the concentrations observed from both locations with total aerosol filters, likely due to the size cut-off in the size-resolved samples (0.34-12 μm in particle diameter). The variability in INP size distribution showed associations with wind speed and direction. At T = -20 °C, the largest size stage (2.96-12 μm) had the highest fraction of INPs during a period at the beginning of the expedition that encompassed a series of surface cyclones (May 11-18). The INP number concentrations in the smallest smallest size stage (0.15-0.34 μm) eclipsed those on the largest as a larger storm passed the Oden on May 13. Further analysis into INP size and composition, from heat and chemical treatments of samples, will be used to assess their sources.