The nitrogen (N) and oxygen (O) isotopic composition (δ 15 N and δ 18 O) of water soluble aerosol nitrate was measured in aerosol samples collected in Eilat, Israel, from August 2003 to November 2004. During this period δ 15 N values ranged from -6.9‰ to + 1.9‰ and δ 18 O from + 65.1‰ to + 84.9‰ and exhibited strong seasonal variability with higher average δ 15 N values observed in the summer and higher δ 18 O values in the winter. Nitrate isotopic composition was compared with bulk chemical composition and extractable ion and trace metals on co-collected samples linking nitrate isotopic composition to various sources of aerosols to this region. Atmospheric processes impacting the isotopic signatures of nitrate were also considered. Based on back trajectory analyses, the majority of NO 3 -came from air masses originating over the Mediterranean Sea (34%), Western Europe (20%) and the local Negev desert (19%), which contain a larger anthropogenic imprint compared to southern and eastern air masses which are dominated by mineral dust.
The potential role of reactive mineral dust aerosols as a regulator of NO 3 -isotopic composition is considered; however, based on factor analysis, neither δ 15 N nor δ 18 O were associated with mineral dust components (such as Fe or Al), but rather with anthropogenic indicators such as Cu, Cd, P and Pb. Seasonality in primary NO x cycling reactions driven by seasonal changes in solar radiation, relative humidity and temperature also influence the observed isotopic signatures. The isotope data, together with trace element analysis, suggests that seasonal variations in both δ 15 N NO3 and δ 18 O NO3 are related to both NO x source and transport processes as well as NO x chemical reactions in the atmosphere. The flux-weighted δ 15 N of aerosol NO 3 -in this area averaged -2.6‰ making aerosol deposition a substantial contributor of low δ 15 N nitrogen to the oligotrophic waters of the Gulf of Aqaba. Thus, while the flux of atmospheric N to oligotrophic marine systems is smaller than the upward flux of NO 3 -from deep water, it nonetheless represents an important source of new N having a low δ 15 N. Further, if this low δ 15 N signature is not considered, it could interfere with N-fixation estimates based on isotopic composition of dissolved nitrate or particulate organic nitrogen. Thus, atmospheric deposition should be constrained for accurate estimates of marine N-fixation when based on δ 15 N in the ocean. Indeed, in the Gulf of Aqaba, low upper water δ 15 N NO3 values could be related to inputs of atmospheric NO 3 -as well as N-fixation.