Attribution of aerosol light absorption to black carbon and volatile aerosols

Abstract

We investigated the contribution of volatile aerosols in light-absorption measurement by three filter-based optical instruments [aethalometer, continuous light-absorption photometer (CLAP), and continuous soot monitoring system (COSMOS)] at Gosan Climate Observatory (GCO) from February to June 2012. The aerosol absorption coefficient (σ _abs) and the equivalent black carbon (BC) mass concentration (M _BC) measured by the aethalometer and CLAP showed good agreement with a difference of 9 %, which is likely due to the instrumental uncertainty. However, σ _abs and M _BC measured by the COSMOS with a heated inlet were found to be approximately 44 and 49 % lower than those measured by the aethalometer and CLAP under ambient conditions, respectively. This difference can be attributed to the light absorption by the volatile aerosols coexisting with the BC. Even considering inherent observational uncertainty, it suggests that approximately 35–40 % difference in the σ _abs and M _BC can be contributed by volatile aerosols. Increase in the difference of M _BC measured by the aethalometer and COSMOS with the increasing thermal organic carbon (OC) measured by Sunset OC/EC analyzer further suggests that the filter-based optical instruments without the use of a heater are likely to enhance the value of σ _abs and M _BC, because this sample air may contain both BC and volatile aerosols.

Appendices

Appendix 1

Principle of the aethalometer

The principle of the aethalometer is given below as suggested by Hansen and Schnell (2005). The “optical attenuation” ATN is expressed as

$$ \mathrm{ATN}=100 \ln \left(\frac{I_o}{I}\right) $$

(1)

where, I _o is the intensity of light transmitted through the reference portion of the filter, and I is the intensity of light transmitted through the sample spot.

Attenuation coefficient (σ _ATN) is given as:

$$ {\sigma}_{\mathrm{ATN}}=\frac{A}{100V}\frac{\varDelta \mathrm{ATN}}{\varDelta t} $$

(2)

where A is the filter spot area, V is the volumetric flow rate, and ΔATN is the change in the attenuation during the time interval Δt. Equivalent BC mass concentration (BC_ATN), micrograms per cubic meter is calculated using the following formula:

$$ {\mathrm{BC}}_{\mathrm{ATN}}=\frac{\sigma \mathrm{ATN}}{\alpha \mathrm{ATN}} $$

(3)

where, $\sigma \mathrm{ATN}=\frac{14625}{\lambda }$ is spectral mass specific attenuation cross-section.

Aethalometer correction

In this study, multiscattering error and filter-loading error that arise in the aethalometer is corrected by using corrected equation as proposed by Schmid et al. (2006),

$$ {\sigma}_{\mathrm{abs}\_\mathrm{correct}}=\frac{\sigma \mathrm{ATN}}{\mathrm{Cref}\left[\left(\frac{1}{\mathrm{f}}-1\right)\frac{ \ln \mathrm{ATN}- \ln 10}{ \ln 50- \ln 10}+1\right]} $$

(4)

where, σ _{abs _ correct} is corrected absorption coefficient, Cref is the correction factor for multiple light scattering effects within the filter (Weingartner et al. 2003) and R(ATN) is correction for shadowing effect due to filter loading,

$$ R\left(\mathrm{ATN}\right)=\left[\left(\frac{1}{f}-1\right)\frac{ \ln \mathrm{ATN}- \ln 10}{ \ln 50- \ln 10}+1\right] $$

(5)

where, f is shadowing factor equal to 1.2.

Corrected mass concentration of BC is then obtained as:

$$ M=\frac{\sigma_{\mathrm{abs}\_\mathrm{correct}}}{\sigma_{\mathrm{BC}}} $$

(6)

where, σ _BC is the mass absorption coefficient equal to 10 m² g⁻¹ (experimental value) as suggested by Miyazaki et al. (2008).

Appendix 2

Interpolation method

In order to get the aerosol absorption coefficients at the same wavelength, it needs to be interpolated or extrapolated to the matching wavelength with the aid of the Ångström exponent (Backman et al. 2010). If the two parameters are measured at different wavelengths λ₁ and λ₂, Ångström exponent,

$$ {\alpha}_{12}=-\frac{ \log \left(\frac{\sigma_1}{\sigma_2}\right)}{ \log \left(\frac{\lambda_1}{\lambda_2}\right)} $$

(7)

describes the wavelength dependency of scattering or absorption. The Ångström exponent can be used to relate the data to the same wavelength (λ _x ) with

$$ {\sigma}_x={\sigma}_1\left(\frac{\lambda_1}{\lambda_2}\right){\alpha}_{12} $$

(8)

This requires the assumption that the spectral dependence of the Ångström exponent is constant. In this study, the light absorption data from the aethalometer were interpolated to the same wavelength as the COSMOS measured the data in.

Appendix 3

Principle of the COSMOS

The absorption coefficient (σ _abs) is determined using the following equation (Miyazaki et al. 2008):

$$ {\sigma}_{\mathrm{abs}}=\left(\frac{A}{V}\right)\times \ln \left(\frac{I_i}{I_{i+1}}\right) $$

(9)

where, A (m²) is the area of sample spot, V (m³) is air sample volume during the given time period, and I _i and I _i + 1 are the average values of the filter transmittances during the prior and current time periods, respectively. Procedure detail can be found in Miyazaki et al. (2008).

COSMOS correction

Aerosol absorption coefficient measured by the COSMOS is corrected using the Bond et al. (1999) correction method. σ _{abs _ correct i} is given by the following equation

$$ {\sigma}_{\mathrm{abs}\_\mathrm{correct}\ i}={\sigma}_{\mathrm{abs}\ i}\times F\left(\mathrm{Tri}\right)/K $$

(10)

$$ F\left({T}_{\mathrm{ri}}\right)=\frac{1}{2\times \left(0.5398\times {T}_{\mathrm{ri}}+0.355\right)} $$

(11)

where, ${T}_{ri}=\frac{I_i}{I_0}$ is the filter transmission and K = 1.22 (Bond et al. 1999). Procedure detail is explained in Bond et al. (1999). Mass concentration of BC is then obtained by (6).