Relative errors in derived multi-wavelength intensive aerosol optical properties using cavity attenuated phase shift single-scattering albedo monitors, a nephelometer, and tricolour absorption photometer measurements

Abstract

Aerosol intensive optical properties, including Ångstrom exponents for aerosol light extinction (EAEs), scattering (SAEs) and absorption (AAEs) as well as and the single-scattering albedo (SSA), are indicators for aerosol size, chemical composition, radiative behaviour and particle sources. Derivation of these parameters requires the measurement of aerosol optical properties at multiple wavelengths, which usually involves the use of several instruments. Our study aims to quantify the uncertainties in the determination of these intensive properties using an optical closure approach. In our laboratory closure study, we measured the full set of optical properties for a range of light-absorbing particles with different properties externally mixed with ammonium sulfate to generate aerosols with controlled SSA values. The investigated absorbing particle types were fresh combustion soot emitted by an inverted flame soot generator (SOOT; fractal agglomerates), Aquadag (AQ; compact aggregates), Cabot black (BC; compact agglomerates) and an acrylic paint (magic black, shape unknown). The instruments used in this study were two cavity attenuated phase shift particle monitors for single-scattering albedo (CAPS PMSSA's; hλCombining double low line450, 630gnm) for measuring light-extinction and light-scattering coefficients, one integrating nephelometer (λCombining double low line450, 550, 700gnm) for light-scattering coefficients, and one tricolour absorption photometer (TAP; λCombining double low line467, 528, 652gnm) for filter-based light-absorption coefficients. One key finding is that the coefficients of light absorption, scattering and extinction derived from combing the measurements of two independent instruments agree with measurements from single instruments; the slopes of regression lines are equal within reported uncertainties (i.e. closure is observed). Despite closure for measured absorption coefficients, we caution that the estimated uncertainties for absorption coefficients, propagated for the differential method (DM; absorptiongCombining double low linegextinction minus scattering), can exceed 100g% for atmospherically relevant SSA values (>0.9). This increasing estimated uncertainty with increasing SSA yields AAE values that may be too uncertain for measurements in the range of atmospheric aerosol loadings. We recommend using the DM for measuring AAE values when the SSAg