Comparison of Filter-Based and in-Situ Black Carbon Measurements during the GOAmazon 2014/15 Campaign

Black carbon (BC) is an important warming agent in the atmosphere and thus much effort from the scientific community has been expended to accurately measure the contribution of this aerosol type to the overall radiative forcing. Filter-based measurements of BC are used worldwide as these instruments are inexpensive, easy to operate, and robust. However, they are prone to several errors, since the measurement is based on light transmission change through a filter and therefore determines aerosol light absorption indirectly.  For example, particles collected on the filter, and the filter material itself, will scatter light and result in a reduction in light transmission that would, absent proper corrections, be erroneously ascribed to BC absorption. As part of the Department of Energy-sponsored GOAmazon 2014/15 project, an Aethalometer, a Particle Soot Absorption Photometer (PSAP), and a Single Particle Soot Photometer (SP2) were deployed. The Aethalometer and PSAP are both filter-based techniques that use light transmission to infer aerosol absorption, and the SP2 measures properties of refractory BC (rBC) using laser-induced particle incandescence. Since the SP2 is an in-situ technique, it is not subject to the artifacts of filter-based measurements and therefore can be utilized as the reference measurement to better quantify the measurement artifacts associated with filter-based instruments.

A comparison between the absorption coefficient (determined using the Aethalometer and PSAP) and the rBC mass concentration (determined using the SP2) was conducted to further investigate the dependence of the filter-based measurement artifacts on different ambient conditions (biomass burning (BB) and non-BB) and different aerosol types. Currently available correction schemes for the Aethalometer and PSAP were used and analyzed for the whole period to understand which scheme works best under different conditions, and to investigate improvements that could be made. Preliminary results suggest that the Aethalometer overestimates the absorption coefficient, especially under high loading BB conditions, and this pattern follows with the available corrections schemes tested.