Practice and applications of 17-O-excess measurements of water using novel laser spectroscopy

Wednesday, 17 December 2014
Eric J. Steig1, Gkinis Vasileios2, Andrew Schauer1, Spruce W Schoenemann1, John Hoffnagle3 and Kate J Dennis3, (1)University of Washington, Seattle, WA, United States, (2)Center for Ice and Climate, Copenhagen, Denmark, (3)Picarro, Inc., Santa Clara, CA, United States
17O-excess, defined as the deviation from the Global Meteoric Water Line (GMWL) in a plot of ln18O+1) vs. ln17O+1), is an evolving tool for understanding the modern water cycle and reconstructing past climate regimes. Because of competing effects between equilibrium and kinetic fractionation small variations in 17O-excess can be used, for example, to (i) infer changes in temperature and sea ice across glacial-interglacial cycles in Antarctica (Schoenemann et al., 2014), (ii) study the role of rain re-evaporation during convective events thereby improving the incorporation of isotopes into GCMs (Landais et al., 2010), and (iii) assess the role of stratospheric water vapor intrusions at high altitudes or in polar regions (Winkler et al., 2013).

In natural waters, variability in 17O-excess is very small (on the order of tens of per meg, where 1 per meg is 0.001‰). Until recently, only measurements made via Isotope Ratio Mass Spectrometry (IRMS) could achieve the required precision, following time-consuming front-end chemistry that converted H2O into O2 for analysis of m/z+ 32, 33 and 34. Recent improvements in laser-based spectroscopy, e.g., Cavity Ring-Down Spectroscopy (CRDS), are enabling quicker and easier measurement of 17O-excess in water (Steig et al., 2013; 2014). The Picarro L2140-i is certified with a precision of ≤ 0.015‰; however implementation of best practices can result in an achievable precision of ≤ 0.008‰, thereby demonstrating comparable performance to IRMS. We will review our recommendations for achieving high-precision measurements of 17O-excess on the Picarro L2140-i, including how to calibrate the system, the frequency of standards analysis, the number of replicate injections and vials required, and approaches to dealing with sample-to-sample memory. We will also compare the external accuracy achieved by three distinct Picarro L2140-i analyzers for multiple waters with distinct isotopic composition.