A multi-site comparison of aerosol properties, including ice-nucleating particles, measured during autumn field campaigns with online and offline techniques

Although the coverage of ice-nucleating particle (INP) measurements grows more comprehensively distributed by the year, studies measuring comprehensive aerosol properties often focus extensively on one or two sites at most, and methodology between studies even within the same lab is rarely the same as methods are refined and study aims differ. This inconsistency in part contributes to a sizeable uncertainty in effective radiative forcing estimation in regard to aerosol cloud interactions. To complement our insufficient knowledge of aerosol properties, we compiled and assessed a multitude of aerosol measurements from four sites: Graciosa Island to represent a predominantly marine site (data from autumn 2020), central Oklahoma to represent a mid-latitude terrestrial site (data from autumn 2019), and Utqiagvik, Alaska (data from autumn 2021) and Ny-Ålesund, Svalbard (data from autumn 2019 and 2020) to represent Arctic sites in different longitudes.

Here we report both total aerosol concentration as well as information related to two distinct aerosol concentrations: INPs and cloud concentration nuclei (CCN). INPs were measured using a single online and two offline methods, while CCN and total aerosol concentrations were measured with the same methods and similar instrumentation. This dataset can allow a broad-level comparison of several contrasting sites which would be expected to have vastly different INP and CCN activated fractions and total concentrations due to the variety in aerosol sources around the globe. Our spatial variation analysis indicates the aerosol concentrations vary by up to two orders of magnitude between sites (approximately 10¹ to 10³ particles per cubic centimeter), while INP and CCN concentrations measured by two comparable methods vary by an order of magnitude (approximately 10¹ to 10² INPs per liter, and approximately 10¹ to 10² CCN per cubic centimeter) and INPs show much less variation than the ranges reported by previous studies. This small range may be due to similarities in INP composition, even as total aerosol population composition varies. For instance, INPs at all sites include a population of dust aerosols (either locally sourced or long-range transported). On the other hand, the total aerosol sources are more divergent between sites, especially between continental and marine-dominant sites where the greatest differentiation is seen. As well, previous studies have focused on a yearly average rather than a single season, which may also lead to greater variation. There is also variation in the heat-sensitivity of the INP samples, with continental samples from Graciosa Island showing the least heat sensitivity. Carefully comparing a large dataset containing a variety of aerosol property information including INP and CCN concentrations will allow us to determine patterns in the global distribution of aerosols that will be important as future climate models are developed.