Evaluation of organic markers for chemical mass balance source apportionment at the Fresno Supersite

Sources of PM2.5 at the Fresno Supersite during high PM2.5 episodes occurring from 15 December 2000-3 February 2001 were estimated with the Chemical Mass Bal- ance (CMB) receptor model. The ability of source profiles with organic markers to distinguish motor vehicle, residen- tial wood combustion (RWC), and cooking emissions was evaluated with simulated data. Organics improved the dis- tinction between gasoline and diesel vehicle emissions and allowed a more precise estimate of the cooking source con- tribution. Sensitivity tests using average ambient concen- trations showed that the gasoline vehicle contribution was not resolved without organics. Organics were not required to estimate hardwood contributions. The most important RWC marker was the water-soluble potassium ion. The es- timated cooking contribution did not depend on cholesterol because its concentrations were below the detection limit in most samples. Winter time source contributions were esti- mated by applying the CMB model to individual and average sample concentrations. RWC was the largest source, con- tributing 29-31% of measured PM2.5. Hardwood and soft- wood combustion accounted for 16-17% and 12-15%, re- spectively. Secondary ammonium nitrate and motor vehicle emissions accounted for 31-33% and 9-15%, respectively. The gasoline vehicle contribution (3-10%) was comparable to the diesel vehicle contribution (5-6%). The cooking con- tribution was 5-19% of PM2.5. Fresno source apportionment results were consistent with those estimated in previous stud- ies.