17 O-excess and d-excess of atmospheric water vapor measured by cavity ring-down spectrometry: Evidence of a matrix effect and implication for the calibration procedure.

RATIONALE Producing robust high frequency time series of raw atmospheric water vapor isotope data by laser spectrometry requires accurate calibration. Especially, the chemical composition of the analyzed sample gas can cause isotope bias. This study assesses the matrix effect on calibrated δ17 O, δ18 O, δ2 H, 17 O-excess, and d-excess values of atmospheric water vapor. METHODS A Picarro L2140-i cavity ring-down spectrometer combined with an autosampler and a vaporizer is used to analyze δ17 O, δ18 O, δ2 H, 17 O-excess and d-excess of two water standards. Isotope data obtained using synthetic air and dry ambient air as carrier gas at water mixing ratios ranging from 2000 to 30000 ppmv are compared. Based on the results, atmospheric water vapor measurements are calibrated. The expected precision is estimated by Monte Carlo simulation. RESULTS The dry air source strongly impacts raw isotope values of the two water standards, but has no effect on the mixing ratio dependency functions. When using synthetic air, δ17 O, δ18 O and 17 O-excess of calibrated atmospheric water vapor are overestimated by 0.6 ‰, 0.7 ‰, and 217 per meg, respectively, while δ2 H and d-excess are underestimated by 1.5 ‰ and 7.3 ‰. Optimum precisions for the calibrated δ17 O, δ18 O, δ2 H, 17 O-excess and d-excess values and 12-min integration time are 0.02 ‰, 0.03 ‰, 0.4 ‰, 14 per meg and 0.4 ‰, respectively. CONCLUSIONS In light of the obtained results, recommendations for the calibration of atmospheric water vapor isotope measurements are presented. The necessity to use dry ambient air as dry air source when running the standards for calibration is pointed out as a pre-requisite for accurate atmospheric water vapor 17 O-excess and d-excess measurements.