Source apportionment of sediment PAHs in Lake Calumet, Chicago: application of factor analysis with nonnegative constraints.

Abstract

A factor analysis model with nonnegative constraints (FA) was used to apportion the sources of PAHs found in sediments of Lake Calumet and surrounding wetlands in southeast Chicago. Source profiles and contributions, with uncertainties, are determined with no prior knowledge of sources. The model includes scaling and backscaling of data with average PAH concentrations without sample normalization. This work is a follow-up to a study that used a chemical mass balance (CMB8.2) model to apportion sources to the same data set. Literature source profiles, modified based on gas/particle partitioning of individual PAHs, from eight PAH sources were considered for comparison. FA results for a two-source solution indicate coke oven (45%) and traffic (55%) are the primary PAH sources to Lake Calumet sediments. A six-source FA solution indicates that coke oven (47%) and traffic (45%) related sources are major PAH sources and wood burning-coal residential (2.3%) is a minor PAH source. From the six-source solution, two coke oven profiles are observed, a standard coke oven profile (33%), and a degraded or second coke oven profile (14%) low in phenanthrene and pyrene. Observed traffic related sources include gasoline engine (36%) exhaust and traffic tunnel air (9.3%). This work supports the previous study of Lake Calumet PAHs by CMB model. In addition, FA provides new insights since wood burning and secondary coke oven profiles were not recognized in the CMB model.