Comparison of different methodologies used for the estimation of secondary organic carbon (SOC)

A significant fraction of organic carbon (OC) is of secondary origin. The estimate of secondary organic carbon (SOC) is necessary for the implementation of air quality policy effectiveness. However, its evaluation still remains uncertain despite intensive efforts undertaken during the last decades to characterize and understand the chemistry leading to the formation of secondary organic aerosol (SOA). Though, there is still no direct way to separate secondary from primary organic aerosol fractions. Recent researches have used several methodologies to achieve this evaluation though the need of studies is still required to determine their associated variability and the consistency between the different methodologies. The objective of this work is to estimate the contribution of SOC fraction to OA content by using various methodologies, and to evaluate and discuss the differences accounted in the apportionment and time evolution of SOC estimates. Measurements were conducted at the SIRTA atmospheric research observatory, representing the suburban background air quality conditions of the Paris area (about 25 km SW from Paris city center). PM10 samples were collected every 4 hours over a period of intensive PM pollution events (PM > 50 μg m-3 over several days) from 6 to 24, March 2015, concomitantly with online measurements, carried out using ACSM, Aethalometer 7, TEOMFDMS, PTR-MS, NOx and O3 analyzers. Following offline sampled filter analyses, SOC were estimated using EC-, WSOC-, and SOA tracer-based methods, as well as using online data using positive matrix factorization (PMF) applied to ACSM data (Kleindienst et al. 2007; Sun et al. 2011; Turpin and Huntzicker 1995; Weber et al. 2007). SOC estimates ranged from 19% to 51% of PM10 OC during the campaign. SOA tracer method, which relies on the use of secondary species to account for SOC contribution from individual precursors, provided the lowest SOC estimate. For the first time, estimation of SOC concentrations from phenolic compounds (methylnitrocatechols (Iinuma et al. 2010)), known to account for a major fraction of SOA biomass burning (Bruns et al. 2016), has been investigated, showing a contribution of 60% to total SOC during the first part of the PM pollution event which reflects the predominance of local emissions and notably residential heating. The comparison of SOC estimates showed a similar time evolution profile to a certain extent (Figure 1). After March 14th, high concentrations of secondary inorganic species such as NH4NO3 and (NH4)2SO4, SOC results presented different temporal behavior. These results highlight the problem associated with the methods to apportion SOC in a polluted environment. Discussion will further focus on the photochemical origin of SOA, and cause for the noticeable change observed in the temporal behavior of SOC estimates according the nature and origin of the PM during the pollution events. Information about the limitations and challenges associated to SOC estimations will also be discussed