PM10 source apportionment in two sites of southern Spain by Positive Matrix Factorization. Evaluation of the relevance of sampling site altitude to the PM10 fingerprint

The evaluation of the sources of particulate matter (PM) is one of the most important topics in environmental science. Both natural and anthropogenic sources are involved in the overall PM pollution in both urban and rural areas. Mathematical methods, as Positive Matrix Factorization (PMF), applied to chemical data are the most powerful tools for the discrimination of PM sources. In the present work, the results obtained from a three-year sampling campaign (between 2017 and 2019) are presented. 700 PM10 filters were collected in the framework of FRESA Project (Impact of dust-laden African air masses and of stratospheric air masses in the Iberian Peninsula. Role of the Atlas Mountains) from two sites in Andalusia, southern Spain: the first one is in the city of Granada, while the second one is in Sierra Nevada. Filters were analyzed by ion chromatography and Particle-Induced X-ray Emission (PIXE) for elemental analysis. The two stations are relatively close to each other (around 20 km). However, the Sierra Nevada station is located at an altitude of 2550 m a.s.l, while the Granada station is at 738 m a.s.l. This altitude difference of almost 2000 m makes the two sites very different from a PM-sources point of view, as highlighted by the two parallel PMF models applied to chemical data. Indeed, Sierra Nevada samples showed the impact of frequent mineral dust intrusions from Sahara Desert, that greatly affected the overall PM composition; in Granada site, instead, samples showed the typical urban fingerprint, with lower evidences of Saharan dust intrusions, due to the different circulation as a function of height.