Fig. 2: Comparison of the observed and modeled σabs at 660 nm.

a Time series of the observed absorption coefficient (\({{\rm{\sigma }}}_{{\rm{abs}}}^{{\rm{obs}}}\)), the modeled absorption coefficient using the external mixing (\({{\rm{\sigma }}}_{{\rm{abs}}}^{\mathrm{mod}-{\rm{ext}}}\)), the modeled absorption coefficient using the core-shell morphology (\({{\rm{\sigma }}}_{{\rm{abs}}}^{\mathrm{mod}-{\rm{CS}}}\)) and modeled absorption coefficient using the coated-aggregate morphology (\({{\rm{\sigma }}}_{{\rm{abs}}}^{\mathrm{mod}-{\rm{CA}}}\)). The grey area in a and b corresponds to the uncertainty in the AE33 when compared to reference instruments like multi-angle absorption photometer (MAAP); details can be seen in Methods. b Diurnal variation of the modeled and observed absorption coefficient. c Relative error in the modeled light absorption as a function of the fraction of BC mass (fBC).