Dinitrogen (N₂) production from denitrification and anaerobic ammonium oxidation represents a loss of reactive nitrogen (N) from terrestrial and aquatic ecosystems to the atmosphere. The large ¹⁵N additions required to detect ¹⁵N₂ production against the high atmospheric background precludes the use of the ¹⁵N tracer technique in natural terrestrial ecosystems. We present an isotope ratio mass spectrometry technique that dramatically improves the precision of ¹⁵N₂ measurements. The approach uses gas chromatography to remove oxygen and gas purification techniques to remove water vapor and trace gases that can interfere with ¹⁵N₂ analysis. The analytical precision for manual gas sample injection was 0.018‰ δ¹⁵N; this translates to a minimum detectable N₂ flux of 0.12 ng-N g⁻¹ h⁻¹ over a 24 h incubation using our experimental parameters. We measured denitrification-derived N₂ production rates of 0.67 ± 0.04 ng N g⁻¹ dry soil h⁻¹ following the addition of 0.1 μg 98 atom % ¹⁵N–NO₃⁻ g⁻¹ dry soil to anaerobically-incubated soils; rates were significantly higher with the addition of 1.0 μg 98 atom % ¹⁵N–NO₃⁻ g⁻¹ dry soil (p < 0.001, n = 5), averaging 1.3 ± 0.03 ng N g⁻¹ dry soil h⁻¹. The N₂:N₂O ratio at 24 h after ¹⁵N addition was 48 ± 12 and 5.4 ± 1.9 for the 0.1 μg ¹⁵N g⁻¹ and 1.0 μg ¹⁵N g⁻¹ treatments, respectively. The decrease in the N₂:N₂O with the addition of only 1 μg ¹⁵N g⁻¹ underscores the need to use very low rates of ¹⁵N addition to accurately characterize denitrification dynamics. This analytical advance will allow us to better estimate the N₂:N₂O ratio of denitrification, constrain ecosystem N budgets, and explore the mechanisms of and controls on N₂ production.