(6)

The nitrate (NO3-) dual isotope approach was applied to snowmelt, tundra active layer pore waters, and underlying permafrost in Barrow, Alaska, USA, to distinguish between NO3- derived from atmospheric deposition versus that derived from microbial nitrification. Snowmelt had an atmospheric NO3- signal with N-15 averaging -4.81.0 (standard error of the mean) and O-18 averaging 70.21.7. In active layer pore waters, NO3- primarily occurred at concentrations suitable for isotopic analysis in the relatively dry and oxic centers of high-centered polygons. The average N-15 and O-18 of NO3- from high-centered polygons were 0.5 +/- 1.1 parts per thousand and -4.1 +/- 0.6 parts per thousand, respectively. When compared to the N-15 of reduced nitrogen (N) sources, and the O-18 of soil pore waters, it was evident that NO3- in high-centered polygons was primarily from microbial nitrification. Permafrost NO3- had N-15 ranging from approximately -6 parts per thousand to 10 parts per thousand, similar to atmospheric and microbial NO3-, and highly variable O-18 ranging from approximately -2 parts per thousand to 38 parts per thousand. Permafrost ice wedges contained a significant atmospheric component of NO3-, while permafrost textural ice contained a greater proportion of microbially derived NO3-. Large-scale permafrost thaw in this environment would release NO3- with a O-18 signature intermediate to that of atmospheric and microbial NO3. Consequently, while atmospheric and microbial sources can be readily distinguished by the NO3- dual isotope technique in tundra environments, attribution of NO3- from thawing permafrost will not be straightforward. The NO3- isotopic signature, however, appears useful in identifying NO3- sources in extant permafrost ice.

来源平台：JOURNAL OF GEOPHYSICAL RESEARCH-BIOGEOSCIENCES