Positive Matrix Factorization and Data Quality Assessment of EPA's PM_(2.5) Chemical Speciation Network Derived from Six Collocated CSN Sites for the Period 2010-2013

摘要

1. RCM Results 1.1 Underestimates measured mass (70-87%) 1.2 No attempt to determine organic material (OM) from OC 1.3 Not blank corrected 1.4 Primary and collocated RCM seem to agree well 1.5 Influenced primarily by 'major' constituents 1.6. Not a true material balance. 2. PMF Results 2.1 "Best" fits with 4-6 factors (robust mode PMF2 with no rotations) 2.2 Accounts for 90-98% of measured mass 2.3 Reasonably consistent results between primary and collocated samplers 2.3.1 Same factors 2.3.2 Dust for collocated sampler at Rubidoux and exception 2.4 Sources seem highly mixed, presenting challenges for PMF to separate them 2.4.1 Coal + oil roughly equal for Roxbury primary and collocated samplers, but for primary sampler coal is ~1/3 of combination while for the collocated sampler it is ~2/3 2.4.2 May be result of uncertainty in "tracer" elements in factor (e.g., Se for coal & V for oil) 2.4.3 Has implications for trying to optimize control strategy abatement costs.