RATIONALE: Oxygen isotope ratio measurements of NO2- and NO3- by the azide method and denitrifier method are sensitive to the ?18 O value of the sample water. However, the influence of ?18 OH2O on those measurements has not been quantitatively evaluated and documented so far. Therefore, we investigated the influence of ?18 OH2O of a sample on the ?18 O analysis of NO2- and NO3- . METHODS: We prepared NO2- and NO3- standards (with known ?18 ONO2- and ?18 ONO3- values) dissolved in waters having different ?18 OH2O values (?18 OH2O ?=?-12.6, 25.9, 56.7, and 110.1‰). Nitrite and nitrate were converted into N2 O using the azide method and the denitrifier method, respectively. The isotope ratios of the generated N2 O were measured with a Sercon purge-and-trap gas chromatography/isotope ratio mass spectrometry (PT-GC/IRMS) system. The measured ?18 O values of the produced N2 O were plotted against known ?18 ONO2- and ?18 ONO3- values to evaluate the influence of exchange of an oxygen atom with H2 O during the conversion of NO2- into N2 O and NO3- into N2 O, respectively. RESULTS: The degree of oxygen isotope exchange was 10.8?±?0.3% in the azide method and 5.5?±?1.0% in the denitrifier method, indicating that the azide method is more susceptible to artifacts arising from differences in the ?18 OH2O value of water than the denitrifier method. Thus, the intercept of the standard calibration curve must be corrected to account for differences in ?18 OH2O . Abiotic NO2 -H2 O equilibrium isotope effect experiments yielded a rate constant of (1.13?±?007)?×?10-2 (h-1 ) and an equilibrium isotope effect of 11.9?±?0.1‰ under the condition of pH?=?7.5, 30°C, and 2.5% salinity. CONCLUSIONS: Oxygen isotope ratio measurements of NO2- by the azide method are highly sensitive to ?18 OH2O as a result of significant oxygen isotope exchange between NO2- and H2 O. Therefore, to obtain the most accurate measurements water with the same ?18 OH2O value as that of the sample must be used to make the NO2- and NO3- standards.