A comprehensive characterisation of natural aerosol sources in the high Arctic during the onset of sea ice melt

Abstract

The interactions between aerosols and clouds are still one of the largest sources of uncertainty in quantifying anthropogenic radiative forcing. To reduce this uncertainty, we must first determine the baseline natural aerosol loading for different environments. In the pristine and hardly accessible polar regions, the exact nature of local aerosol sources remains poorly understood. It is unclear how oceans, including sea ice, control the aerosol budget, influence cloud formation, and determine the cloud phase. One critical question relates to the abundance and characteristics of biological aerosol particles that are important for the formation and microphysical properties of Arctic mixed-phase clouds. Within this work, we conducted a comprehensive analysis of various potential local sources of natural aerosols in the high Arctic over the pack ice during the ARTofMELT expedition in May-June 2023. Samples of snow, sea ice, seawater, and the sea surface microlayer (SML) were analysed for their microphysical, chemical, and fluorescent properties immediately after collection. Accompanied analyses of ice nucleating properties and biological cell quantification were performed at a later stage. We found that increased biological activity in seawater and the SML during the late Arctic spring led to higher emissions of fluorescent primary biological aerosol particles (fPBAPs) and other highly fluorescent particles (OHFPs, here organic-coated sea salt particles). Surprisingly, the concentrations of ice nucleating particles (INPs) in the corresponding liquid samples did not follow this trend. Gradients in OHFPs, fPBAPs, and black carbon indicated an anthropogenic pollution signal in surface samples especially in snow but also in the top layer of the sea ice core and SML samples. Salinity did not affect the aerosolisation of fPBAPs or sample ice nucleating activity. Compared to seawater, INP and fPBAP concentrations were enriched in sea ice samples. All samples showed distinct differences in their biological, chemical, and physical properties, which can be used in future work for an improved source apportionment of natural Arctic aerosol to reduce uncertainties associated with their representation in models and impacts on Arctic mixed-phase clouds.

Fig. 1. Sources of natural primary aerosol particles over the high Arctic Ocean during the onset of sea ice melt. Primary particles can be lofted to the atmosphere via wind stress on surfaces (e.g., blowing snow, wave breaking) and bubble bursting (e.g., in leads or melt ponds). Once in the atmosphere, particles can interact with solar radiation and influence cloud properties by acting as cloud condensation nuclei (CCN) or ice nucleating particles (INP).

Fig. 2. Set-up of the experiment. Aerosol particles were atomised using an atomiser, the aerosols were diluted with dry particle-free air and successively analysed using the multiparameter bioaerosol spectrometer (MBS), soot-particle aerosol mass spectrometer (SP-AMS), single-particle soot photometer (SP2) and scanning electrical mobility spectrometer (SEMS), and using filter sampling with transmission electron microscopy (TEM) analyses performed after the expedition. The relative humidity was monitored at the inlet of the SEMS.

Fig. 3. Change in biological activity within the seawater throughout the expedition. Top of (panel A) shows the different sampling periods (in transit and floe ice camps). (A) Continuous normalized chlorophyll-a concentration measured by the ship's clean sea water supply (ship hull) and chlorophyll-a measured from the surface lead water samples at the remote ice stations. (B) 16S rRNA (prokaryote) and 18S rRNA (eukaryote) gene copies per mL sample of sea surface microlayer (SML) and lead water from the remote ice stations. (C) Ice nucleating particle (INP) concentration as measured from surface lead water for distinct freezing temperatures. (D) Contribution (in ‰) of fluorescent primary biological aerosol particles (fPBAPs) to the coarse mode (particle diameter >0.8 μm) aerosol generated using surface lead water, surface lead slush and lead water collected at a depth of 5 m from leads at the remote ice stations.

Fig. 4. Overview of all analysed source samples sorted by source type. (A) Percentage contribution of fluorescent biological aerosol particles (fPBAPs) and other highly fluorescent aerosol particles (OHFPs) to the coarse-mode aerosol (particle diameter >0.8 μm) measured by the MBS. (B) Particle number size distribution measured by the SEMS and MBS. (C) Aerosol chemical mass fraction measured by the SP-AMS. (D) Refractory black carbon (rBC) mass concentration measured by the SP2, divided by the average total number concentration recorded for each sample. Aqua, blue, green, and grey indicate sea ice, lead water, SML, and snow source samples, respectively, in panels A, B and D. The results of the individual samples are shown in the ESI.

Fig. 5. Relationship between bioaerosols, ice nucleating particles and salinity for all source samples and along the ice core thickness. (A and B) Ice nucleating particle (INP) concentration categorised by source type and section for two freezing temperatures, −15 °C and −20 °C, versus the contribution of fluorescent primary biological aerosol particles (fPBAPs) to the coarse-mode aerosol (particle diameter >0.8 μm). (C) fPBAP contribution to the coarse-mode aerosol as a function of sample salinity. (D) 16S rRNA gene copies per mL sample (prokaryote) and INP concentration at different sections of the ice core, and fPBAP contribution generated from these samples. Note that the cell and INP analysis was not done for all the ice core samples that were analysed for the fPBAP contribution.