The N stable isotopes from particulate ammonium nitrate: understanding of the atmospheric processes and the sources of PM10 in France

Aerosol nitrate is an important fraction of the mass of the fine particles (PM) in atmosphere. Moreover, aerosol nitrogen species are predominantly anthropogenic over Europe (traffic, industry and farming). In the view of the increasingly drastic constraints of the European air quality standard for PM, the understanding of these sources has become crucial in France. Indeed, during the months of March and April, several high concentration of PM are observed on a large scale throughout the national territory. During these specific events, the ammonium nitrate (NH4NO3) is a predominant component of the airborne PM. There are two main ways that produces NH4NO3 in air with many precursors as NH3 from specific primary sources (agricultural activity). The presence of these species in PM is the result of complex primary and secondary processes which are always investigated. In literature, many studies propose the nitrogen stable isotope ratio (?15N) of NOx as an indicator for NOx source apportionment ((Walters, Goodwin, et Michalski 2015; Felix, Elliott, et Shaw 2012) but there is a lack of information regarding the ?15N ratio of NH4. Thus, these last two years, the main objectives of the INAC and INCAS-2 programs were to collect various samples from ambient rural (2 sites) and urban (5 sites) background sites in order to quantify the nitrogen isotope ratio of nitrate and ammonium in PM. Then, subsequently the characterization of emissions of the different types of samples (traffic, biomass burning, and agriculture activity) was performed to provide clear information between the nitrogen isotope ratio and the sources. Currently, this study represents the most important French database (long-term and source data) on ?15N composition of particulate ammonium nitrate in PM. The first results show that ?15N-NOx has a high seasonal variability which is relatively difficult to interpret as the impact of the specific sources. Whereas, the ?15N of NH4 seems to contain information about the pollution sources. Indeed, this isotope composition appears to be negatively correlated with particulate NH4NO3 events on several sites as described on Figure 1 (urban site, Grenoble). We observed that the values of ?15N-NH4 ranged from 10‰ to 25‰ in large part of years and decreased significantly toward negative values during these periods of peak agricultural activity. Based on these first results, source apportionments of PM10 collected over one years at each sampling sites were conducted with Positive Matrix Factorization (US EPA PMF v5.0) and Multilinear-Engine (ME-2 SoFI from PSI). In this work, the interest to include such isotope measurements as input data for PMF will be discussed for the first time, focusing on their application to identify specific source of ammonium nitrate in PM