We applied the positive matrix factorization model to two large data sets collected during two intensive measurement campaigns (summer 2011 and winter 2012) at a suburban site in Beirut, Lebanon, in order to identify NMHC (non-methane hydrocarbons) sources and quantify their contribution to ambient levels. Six factors were identified in winter and five factors in summer. PMF-resolved source profiles were consistent with source profiles established by near-field measurements. The major sources were traffic-related emissions (combustion and gasoline evaporation) in winter and in summer accounting for 51 and 74 wt %, respectively, in agreement with the national emission inventory. The gasoline evaporation related to traffic source had a significant contribution regardless of the season (22 wt % in winter and 30 wt % in summer). The NMHC emissions from road transport are estimated from observations and PMF results, and compared to local and global emission inventories. The PMF analysis finds reasonable differences on emission rates, of 20-39 % higher than the national road transport inventory. However, global inventories (ACCMIP, EDGAR, MACC-ity) underestimate the emissions up to a factor of 10 for the transportation sector. When combining emission inventory to our results, there is strong evidence that control measures in Lebanon should be targeted on mitigating the NMHC emissions from the traffic-related sources. From a global perspective , an assessment of VOC (volatile organic compounds) an-thropogenic emission inventories for the Middle East region as a whole seems necessary as these emissions could be much higher than expected at least from the road transport sector.