Characterization and Source Apportionment of Ion and Metals in PM10 in an Urbanized Valley in the American Tropics Using Principal Component Analysis and Positive Matrix Factorization

Santiago A. Franco C.1

Mauricio A. Correa1,Email

David Aguiar G.1

Luisa M. Gómez1

Henry A. Colorado2,Email

1Grupo de Investigación y Laboratorio de Monitoreo Ambiental—GLIMA, Universidad de Antioquia UdeA, Calle 70 N_. 52-21, Medellin 050010, Colombia.
2CCComposites Laboratory, Universidad de Antioquia UdeA, Calle 70 N_. 52-21, Medellín 050010, Colombia.

Abstract

The Metropolitan Region of the Aburrá Valley (MRAV) is an urbanized valley located in the central mountain range of the Andes of Colombia, with Medellin as the most important city of the region. Due to the geography, economy trends, industrialization, and demography, air pollution has become one of the main problems in terms of health and environment. The characterization of the ionic and metallic fraction of Particulate Matter (PM10) was carried out following the standard procedures, using Ionic chromatography for the anions (F-, Cl-, NO2-, NO3- ,PO4-3 and SO4-2) and for cations (K+, Na+, Li+, Mg+2, Ca+2 and NH4+), as well as Atomic absorption spectroscopy for the metals (Ni, Ba, Mn, Cu, Al, Ti, Cr, Zn, Fe). A total of 94 samples were collected in 10 different stations of the air quality network of the MRAV between December 2018 and October 2021. The results were statistically analyzed using receptor models: Principal Component Analysis (PCA) and the EPA Positive Matrix Factorization (PMF) with the aim of identifying the main source of the pollutant. It was found that the average PM10 concentration of the collected samples was 40.65 µg/m3, 43% of the data was over the standard stablished by the WHO (45µg/m3). The contribution of ions was approximately 26% of the PM10 mass, with SO4-2 occupying the largest fraction (approximately 12%), and Na+, having the highest concentration among the cations. The metals corresponded to 3% of the total mass of PM10, with Fe having the highest concentration. Through the statistical analysis, it was found that the results of both models have important similarities, identifying that the main emission sources of the pollutant in the MRAV are secondary aerosols (SA), biomass/waste burning (BB), soil/road dust (SD), vehicular emissions (VE), industrial emissions (IE), and mixed type contributions. The methods and results used in this case study for the particular geography of the valley are an important tool for understanding the effects of PM pollution in the local context and many regions worldwide.

Characterization and Source Apportionment of Ion and Metals in PM10 in an Urbanized Valley in the American Tropics Using Principal Component Analysis and Positive Matrix Factorization