Source apportionment of ambient volatile organic compounds (vocs) in gran La Plata, Argentina

Abstract

In many countries, mostly -but not exclusively- developing countries, only small and/or medium databases obtained by discrete monitoring are available to analyze some characteristics of atmospheric pollution. Ambient Volatile Organic Compounds are pollutants that affect public health, mainly in urban regions close to a petrochemical complex. In this context, the source apportionment is addressed across to different statistical tools oriented to receptors, applied to small and medium databases obtained by discrete monitoring, so that valuable information is acquired in spite of the technical constraint above mentioned. Through this application, it is possible to characterize the relevant emission sources, estimating the profiles of the sources and their contributions by principal component analysis with multiple linear regression (PCA - MRL). Finally, the most probable areas where the pollutants under study originated are estimated from the back trajectories of the contributions of the present sources by means of a Nonparametric Trajectory Analysis (NTA). As a case study, the analysis of ambient Volatile Organic Compounds (VOCs) levels in Gran La Plata (Argentina) is addressed. This region lacks continuous monitoring and is also characterized by the presence of a petrochemical area and high traffic. This study shows that the highest concentrations are near the petrochemical area but that the data collected at the receptor sites can be explained with two main sources, emission associated to the industry (70.6%) and the traffic (28.0%). Then, the integration of PCA-MRL and NTA allowed access to graphic and numerical information of simple interpretation, useful for stakeholders. Furthermore, the estimated lifetime cancer risk (LCR) in Gran La Plata shows that conglomerate inhabitants are not probably exposed to this risk.