Continuous measurements of dissolved oxygen isotopes can provide insight into how oceanic primary production varies over time and space. For example, ¹⁷Δ, the deviation from the expected mass-dependent isotopic fractionation, is a tracer of gross primary production. This thesis focuses on techniques for continuously measuring dissolved oxygen isotopes, and showcases measurements from the Scripps Institution of Oceanography pier. We developed a counterflow-type equilibrator with a time constant of 7-8 minutes for oxygen. When interfaced to a mass spectrometer, this equilibrator allows for a sampling flow rate of 3 mL min⁻¹. Using a model of O₂, N₂, and Ar, the behavior of major gases in an equilibrator is explored, and the corrections needed to account for incomplete equilibration are determined. We also quantify possible sources of interference to the measurement of oxygen isotopes, and find that CO₂ and N₂ contribute to the interference, while the interferences from water vapor and DMS are negligible. In addition, we describe a technique for keeping the O₂/N ₂ ratio constant, to reduce the interference from N₂.

Dissolved oxygen isotopes were measured near the surface ocean at the Scripps Institution of Oceanography pier for five weeks. The data show diurnal cycles in O₂ and δ¹⁸O, with amplitudes of 19 mmol m⁻³ and 1.1 per mil, respectively. The diurnal cycles are well described by a box model that includes photosynthesis, respiration, air-sea gas exchange, and mixing. The timing of the cycle can be explained using a photosynthesis rate proportional to photosynthetically active radiation. The maximum daily photosynthesis rate is 4.7 mmol O₂ m⁻³ hr⁻¹ (40.3 mgC m⁻³ hr ⁻¹ using a photosynthetic quotient of 1.4). This is in agreement with the production estimated from the chlorophyll concentration. Although the ¹⁷Δ data did not have a resolvable diurnal cycle, modeled ¹⁷Δ shows a diurnal cycle with an amplitude of 11 per meg. The oxygen isotope data also show variability over longer timescales, suggesting a change in the production rate over time. In the future, these techniques could be used around the world to improve our understanding of variability in oceanic primary production.