Co-occurrence of elemental carbon and mineral dust in thermal-optical analysis: reducing the bias and approximating mineral dust

Abstract

The quantification of elemental carbon via thermal-optical analysis (TOA) is continuously discussed, as many parameters influence the split between organic and elemental carbon. Here, we focus on the bias introduced by the co-occurrence of mineral dust in PM10 and snow samples and present an approach to estimate the mineral dust mass directly from TOA data. Monitoring light-absorbing aerosols, including mineral dust and elemental carbon, is of high interest due to their direct and indirect climate forcing. During mineral dust events, particulate matter concentration in air is increased and altered elemental composition of PM is observed, as mineral dust introduces high levels of certain elements, e.g., iron and calcium. Especially for remote environments like glaciers, the quantification of light-absorbing aerosols is crucial, as these compounds reduce the albedo and promote melting of the snow cover after deposition. In TOA, which is the reference method for the quantification of organic and elemental carbon for ambient air filters (DIN EN 16909:2017) and which is widely used for the quantification of these compounds in snow samples after filtration, mineral dust leads to a bias in the assignment to organic and elemental carbon. Using routine protocols for evaluation, the elemental carbon loading on the filters is underestimated or total carbon is erroneously even fully allocated to organic carbon. Still, the interference can be used to assess iron loadings attributed to mineral dust (Kau et al, 2022). We review TOA data of PM10 filters collected at the Global Atmosphere Watch station Sonnblick Observatory at an elevation above 3100 m a.s.l. located in the Austrian Alps. The study period covers 6 years, 2019 to 2024, and is based on weekly filters. We observe influence of mineral dust reaching the site after long-range transport on organic and elemental carbon loadings in 16 filters during this period, up to 9 % of filters annually. We correct for the bias introduced by mineral dust using a linear approach and quantify the changes in organic and elemental carbon for single filters and the yearly average values. For single filters, increase of elemental carbon may range up to a factor of 2.5, while the effect on the yearly average values is markedly lower.

After deposition, elemental carbon and mineral dust are stored in the snow cover, which is building up during the accumulation period. The snow cover of a glacier in proximity to Sonnblick Observatory was sampled between 2019 and 2024 in increments of 20 cm and the water insoluble particles were analysed via TOA. We apply the identification, correction and evaluation of changes in organic and elemental carbon to single samples and the entire snow covers. The routinely measured and corrected elemental carbon concentrations in the snow cover collected in 2020 are shown in Figure 1. Between 2019 and 2024, 16 filters required a correction due to the occurrence of mineral dust, up to 38 % of samples per year. Figure 1. Elemental carbon concentrations in the layers of the snow cover collected in 2020. The dark grey bars mark the increase in elemental carbon due to the consideration of mineral dust.

Using the elemental composition of mineral dust obtained from surface snow after strong mineral dust events collected between 2020 and 2024, we calculate the contribution of iron to the mineral dust mass. Using this ratio and the iron loading deduced directly from TOA data, we estimate the mineral dust masses in air and snow for the affected periods.

Kau, D., Greiliniger, M., Kirchsteiger, B., Göndör, A., Herzig, C., Limbeck, A., Eitenberger, E. and Kasper-Giebl, A. (2020) Atmos. Meas. Tech. 15, 5207-5217.