Abstract
Measurement of stable oxygen (delta (super 18) O) and hydrogen (delta (super 2) H) ratios in water has been an important tool for understanding both modern and past climate and environmental conditions, particularly in terrestrial hydrologic systems. delta (super 18) O in the ocean is controlled by Rayleigh fractionation during evaporation and precipitation that is strongly governed by temperature, making delta (super 18) O a conservative interior water mass tracer. Despite this significance, global seawater delta (super 18) O and delta (super 2) H data are relatively sparse, especially from the interior ocean. Previously, it has been time and cost prohibitive to measure seawater isotopes from most ocean observing cruises using traditional isotope ratio mass spectrometry (IRMS) due to the large number of water samples collected. The assumption that seawater isotope variability throughout the global oceans is negligible has been another mitigating factor in minimizing data collection. The development and improvement of laser-based gas analyzers based on off-axis integrated cavity output spectroscopy (OA-ICOS) and cavity ring-down spectroscopy (CRDS) over recent decades has provided an opportunity for relatively low-cost, high-precision delta (super 18) O and delta (super 2) H measurements of natural waters to be analyzed without the time, expense and overhead of traditional IRMS. Previous studies have compared the precision and reproducibility of laser-based spectroscopy with IRMS techniques and found measurements of fresh waters using OA-ICOS and CRDS to be comparable to IRMS with a higher throughput and lower cost. This has drastically increased the data available for modern and paleoclimate water cycle interpretations. However, the feasibility of using laser absorption spectroscopy, particularly using OA-ICOS, for seawater samples has remained understudied. Here we present seawater data from the recent CROCCA-2s Indian Ocean cruise and North Pacific Ocean GEOTRACES (GP-15) transect and show that, using a systematic sample analysis, cleaning routine, and data normalization, measurement accuracies of + or - 0.1 ppm for delta (super 18) O and + or - 0.7 ppm for delta (super 2) H are achievable using OA-ICOS.