Source apportionment of the carbonaceous aerosol Quantitative estimates

Source apportionment of the carbonaceous aerosol – Quantitative estimates based on 14 C- and organic tracer analysis Bordeaux 23 - 25 April 2008 KE Yttri 1, D Simpson 2, H. Puxbaum 3, K Stenström 4, T Svendby 1 1. 2. 3. 4. Norwegian Institute for Air Research Norwegian Meteorological Institute Technical University of Vienna Lund University

SORGA - Main objectives • Quantify the contribution of biogenic and anthropogenic carbonaceous matter to PM in the Nordic urban and rural environment ü Biogenic/anthropogenic fraction in urban and rural areas ü Separate the biogenic and the anthropogenic fraction into primary and secondary carbonaceous matter ü Size-distribution (PM 1 and PM 10) of biogenic and anthropogenic carboanceous matter • Improve current knowledge of what are the concentrations of VOCs crucial for SOA-formation • Improve the aerosol modules of the following models: MAPS - EPISODE - EMEP

SORGA - Measurements sites Measurement campaigns Summer period: 19 June - 5 July 2006 Winter period: 1 - 8 Mars 2007 Oslo (Urban background) Oslo Hurdal (Rural Background)

SORGA - Aerosol parameters measured Table 1: Input parameters for source apportionment of the particulate carbonaceous fraction Parameter Size fraction Time resolution Objective Mass concentration PM 1, PM 10 12 hours PM concentration EC, OCp, TCp (QBQ) PM 1, PM 10 12 hours Carbonaceous PM loading Levoglucosan PM 10 12 hours Wood burning Sugars and Sugar-alcohols PM 1, PM 10 12 hours Fungal spores (PBAP) Cellulose 1 PM 1, PM 10 Grab sample Plant debris (PBAP) f. M (14 C-analysis)2 PM 1, PM 10 Grab sample day Separates between modern Grab sample night Carbon and fossil Carbon 1. Cellulose analysis performed at the Technical University of Vienna 2. 14 C-analysis performed at the University of Lund Yttri et al. in progress

SORGA - Sources of carbonaceous matter OCbb OC from residential wood burning ECbb EC from residential wood burning OCff OC from combustion of fossil fuel ECff EC from combustion of fossil fuel OCpbs OC from fungal spores OCpbc OC from plant debris OCbsoa OC from biogenic sec. org. aerosols OCasoa OC from anthropogenic sec. org. aerosols

SORGA - Equations and uncertainty estimates Equations to calculate carbonaceous subfractions Table 2: Low, central and high factors used to estimate the carbonaceous subfractions using LHS Confounding factors OCbsoa: OCnf OCbsoa, OCmeat cooking, condensation of SVOC from biomass(? ), PBAP not accounted for by sugars and sugar-alcohols

SORGA - Source apportionment of TCp in PM 10 Summer 2 Ocbsoa in PM 10 µg C m-3 OCbsoa: Comparison between sites 1, 5 1. 5 µg C m-3 1 1. 2 µg C m-3 0, 5 Hurdal (RB) PM 10 TCp = 2. 9 ± 1. 2 µg C m-3 0 Natural: 72% Anthropogenic: 28% Oslo (UB) PM 10 TCp = 3. 7 ± 1. 3 µg C m-3 HURDAL OSLO Natural: 46% Anthropogenic: 54%

SORGA - Source apportionment of TCp in PM 1, Summer 1, 5 OCbsoa PM 1 µg C m-3 OCbsoa: Comparison between sites 1 1. 1 µg C m-3 0, 5 Hurdal (RB) PM 1 TCp = 1. 7 ± 1. 1 µg C m-3 0 Natural: 66% HURDAL Anthropogenic: 34% 0. 77 µg C m-3 Oslo (UB) PM 1 TCp = 2. 3 ± 0. 8 µg C m-3 OSLO Natural: 36% Anthropogenic: 64%

SORGA - Source apportionment of TCp in PM 10 Winter 2 TCbb: Comparison between sites TCbb PM 10 µg C m-3 1, 5 1. 5 µg m-3 1 0, 5 PM Hurdal (RB) 10 TCp = 1. 2 ± 0. 5 µg C m-3 0 Natural: 8% Anthropogenic: 92% 0. 58 µg m-3 HURDAL Oslo (UB) PM 10 TCp = 3. 2 ± 1. 5 µg C m-3 OSLO Natural: 5% Anthropogenic: 95%

SORGA - Source apportionment of TCp in PM 10 Summer Day/night variation TCbb PM 10 µg C m-3 1 TCbb: Comparison between day and night 0. 77 µg C m-3 0, 5 Oslo(UB) PM 10 DAY 0. 42 µg C m-3 -3 TCp = 3. 8 ± 1. 2 µg C m 0 Natural: 52% Anthropogenic: 48% DAY Oslo (UB) PM 10 NIGHT TCp = 3. 6 ± 1. 4 µg C m-3 NIGHT Natural: 40% Anthropogenic: 60%

SORGA - Relative cont. of carb. matter to PM 10 (Summer) Hurdal (RB) PM 10 = 10. 4 µg m-3 Oslo (UB) PM 10 = 15. 6 µg m-3 PCM/PM 10 = 49% Natural PCM/PM 10 = 35% PCM/PM 10 = 38% Natural PCM/PM 10 = 19% Conversion factors: OC = 1. 8; OC = 1. 6; OC = 2. 0; EC and EC = 1. 1; OC = 1. 3; OC = 1. 8

SORGA - Relative cont. of carb. matter to PM 10 (Winter) Hurdal (RB) PM 10 = 4. 2 µg m-3 Oslo (UB) PM 10 = 9. 5 µg m-3 PCM/PM 10 = 45% Natural PCM/PM 10 = 3% PCM/PM 10 = 54% Natural PCM/PM 10 = 3% Conversion factors: OC = 1. 8; OC = 1. 6; OC = 2. 0; EC and EC = 1. 1; OC = 1. 3; OC = 1. 8

SORGA - Measured vs modelled conc. of OCbsoa (Oslo, summer) OCp (PM 10) 3. 2 ± 1. 1 µg C m-3 OCp (PM 1) 1. 8 ± 0. 7 µg C m-3 OCbsoa 0. 8 -1. 2 µg C m-3 OCbsoa modeled 0. 2 ± 0. 3 µg C m-3

SORGA - Measured vs modelled conc. of OCbsoa (Hurdal, winter) OCp (PM 10) 3. 0 ± 1. 2 µg C m-3 OCp (PM 1) 1. 4 ± 1. 0 µg C m-3 OCbsoa 1. 0 -1. 5 µg C m-3 OCbsoa modeled 0. 3 ± 0. 3 µg m-3

SORGA - Summary of findings (1) • The combined effort of 14 C, TOA, and organic tracer analysis is a powerful tool to explore various sources of carbonaceous matter • OCbsoa was the major carbonaceous fraction in summer regardless of site and size fraction • OCbb was the major carbonaceous fraction in winter regardsless of site and size fraction • ECff is the major contributor to EC regardless of season and size fraction Secondary organic aerosols vs primary carbonaceous aerosols Summer: Winter: Rural background site: SOA > PCA Urban background site: SOA ≤ PCA Rural and urban site: SOA << PCA

SORGA - Summary of findings (2) Anthropogenic vs natural sources of carbonaceous matter Urban background site: Summer: Natural >> Anthropogenic Winter: Natural << Anthropogenic Rural background site: Summer: Natural < Anthropogenic Winter: Natural << Anthropogenic