Source apportionment of fine particulate matter by clustering single-particle data: Tests of receptor model accuracy

The source apportionment accuracy of a neural network algorithm (ART-2a) Is tested on the basis of its application to synthetic single-particle data g enerated by a source-oriented aerosol processes trajectory model that simul ates particle emission, transport, and chemical reactions in the atmosphere . ART-2a successfully groups particles from the majority of sources actuall y present, when given complete data on ambient particle composition at moni toring sites located near the emission sources. As particles age in the atm osphere, accumulation of gas-to-particle conversion products can act to dis guise the source of the primary care of the particles. When ART-2a is appli ed to synthetic single-particle data that are modified to simulate the bias es in aerosol time-of-flight mass spectrometry (P;TOFMS) measurements, best results are obtained using the ATOFMS dual ion operating mode that simulta neously yields both positive and negative ion mass spectra. The results of this study suggest that the use of continuous single-particle measurements coupled with neural network algorithms can significantly improve the time r esolution of particulate matter source apportionment.