One year of aerosol refractive index measurement from a coastal Antarctic site

Abstract

Though the environmental conditions of the Weddell Sea region and Dronning Maud Land are still relatively stable compared to the fast-changing Antarctic Peninsula, we may suspect pronounced effects of global climate change for the near future (Thompson et al., 2011). Reducing the uncertainties in climate change modeling requires a better understanding of the aerosol optical properties, and for this we need accurate data on the aerosol refractive index (RI). Due to the remoteness of Antarctica only very few RI data are available from this region (Hogan et al., 1979; Virkkula et al., 2006; Shepherd et al., 2018). We calculate the real refractive index of natural atmospheric aerosols from number size distribution measurements at the German coastal Antarctic station Neumayer III. Given the high average scattering albedo of 0.992 (Weller et al., 2013), we assumed that the imaginary part of the RI is zero. Our method uses the overlapping size range (particle diameter D between 120 and 340 nm) of a scanning mobility particle sizer (SMPS), which sizes the particles by their electrical mobility, and a laser aerosol spectrometer (LAS), which sizes the particles by their optical scattering signal at the 633 nm wavelength. Based on almost a complete year of measurement, the average effective refractive index (RIeff, as we call our retrieved RI because of the used assumptions) for the dry aerosol particles turned out to be 1.44 with a standard deviation of 0.08, in a good agreement with the RI value of 1.47, which we derived from the chemical composition of bulk aerosol sampling measurements. At Neumayer the aerosol shows a pronounced seasonal pattern in both number concentration and chemical composition. Despite this, the variability of the monthly averaged RIeff values remained between 1.40 and 1.50. Compared to the annual mean, two austral winter months (July and September) showed slightly but significantly increased values (1.50 and 1.47, respectively). The size dependency of the RIeff could be determined from time-averaged LAS and SMPS number size distributions measured between December 2017 and January 2018. Here we calculated RIeff for four different particle size ranges and observed a slight decrease from 1.47 (D range 116–168 nm) to 1.37 (D range 346–478 nm). We find no significant dependence of the derived RIeff values on the wind direction. Thus we conclude that RIeff is largely independent of the general weather situation, roughly classified as (i) advection of marine boundary layer air masses during easterly winds caused by passing cyclones in contrast to (ii) air mass transport from continental Antarctica under southern katabatic winds. Neumayer, the only relevant contamination source, is located 1.5 km north of the air chemistry observatory, where the measurements were performed. Given that northerly winds are almost absent, the potential impact of local contamination is minimized in general. Indeed our data show no impact of local contamination on RIeff. Just in one case a temporary high-contamination episode with diesel engines operating right next to the measurement site resulted in an unusual high RIeff of 1.59, probably caused by the high black carbon content of the exhaust fumes. To conclude, our study revealed largely constant RIeff values throughout the year without any sign of seasonality. Therefore, it seems reasonable to use a single, constant RIeff value of 1.44 for modeling optical properties of natural, coastal Antarctic sub-micrometer aerosol.