The total PAH associated to the airborne particulate matter (PM10) was apportioned by one receptor model based on positive matrix factorization (PMF) in an urban environment (Zaragoza city, Spain) during February 2010-January 2011. Four sources associated with coal combustion, gasoline, vehicular and stationary emissions were identified, allowing a good modelling of the total PAH (R(2) = 0.99). A seasonal behaviour of the four factors was obtained with higher concentrations in the cold season. The NE direction was one of the predominant directions showing the negative impact of industrial parks, a paper factory and a highway located in that direction. Samples were classified according to hierarchical cluster analysis obtaining that, episodes with the most negative impact on human health (the highest lifetime cancer risk concentrations), were produced by a higher contribution of stationary and vehicular emissions in winter season favoured by high relative humidity, low temperature and low wind speed.