Estimates of mass absorption cross sections of black carbon for filter-based absorption photometers in the Arctic

Abstract. Long-term measurements of black carbon (BC) are warranted for investigating changes in its emission, transport, and deposition. However, depending on instrumentation, parameters related to BC such as aerosol absorption coefficient (b_abs) have been measured instead. Most ground-based measurements of b_abs in the Arctic have been made by filter-based absorption photometers, including multi-angle absorption photometers (MAAP), particle soot absorption photometers (PSAP), continuous light absorption photometer (CLAP), and Aethalometers. The measured b_abs can be converted to atmospheric mass concentrations of BC (M_BC) by assuming the value of the mass absorption cross section (MAC = b_abs/M_BC). However, the accuracy of conversion of b_abs to M_BC has not been adequately assessed. Here, we introduce a systematic method for deriving MAC values from b_abs measured by these instruments and independently measured M_BC. In this method, M_BC was measured with a filter-based absorption photometer with a heated inlet (COSMOS). COSMOS-derived M_BC (M_BC (COSMOS)) is traceable to a rigorously calibrated single particle soot photometer (SP2) and the absolute accuracy of M_BC (COSMOS) has been demonstrated previously to be about 15 % in Asia and the Arctic. The necessary conditions for application of this method are a high correlation of the measured b_abs with independently measured M_BC, and long-term stability of the correlation slope, which represents the MAC. In general, b_abs – M_BC (COSMOS) correlations were high (r² = 0.84–0.96 for hourly data) at Fukue in Japan, Barrow in Alaska, Ny-Ålesund in Svalbard, Pallastunturi in Finland, and Alert in Canada, and stable up to for 10 years. We successfully estimated MAC values (11.0–15.2 m² g⁻¹ at a wavelength of 550 nm) for these instruments and these MAC values can be used to obtain error-constrained estimates of M_BC from b_abs measured at these sites even in the past, when COSMOS measurements were not made. Because the absolute values of M_BC in these Arctic sites estimated by this method are consistent with each other, they are applicable to study spatial and temporal variation of M_BC and to evaluate performance of numerical model calculations.