At depths where light penetrates, the ocean produces 50-70% of the oxygen in the atmosphere, but we cannot say with certainty which ocean regions are net producers and which are net consumers of oxygen. Ocean stratification, cloud cover, water depth, strong seasonality, and complicated air-sea exchange processes all confound the budgets of oxygen production and consumption. If the budget is changing, we would have difficulty detecting where and determining why. For this project, researchers at SRI International (SRI) and the University of Rhode Island - Graduate School of Oceanography (URI-GSO) developed new methods to estimate net ocean oxygen production using an underwater mass spectrometer and an undulating tow vehicle (Figure 1) to better capture the variability in primary productivity throughout the entire euphotic zone where photosynthesis can occur by measuring dissolved gas concentrations. A new calibration system from SRI was used to improve the precision and accuracy of these in situ measurements by allowing measurements of the water column to be periodically compared to measurements of a dissolved gas standard contained in a compressible bag at the same temperature and hydrostatic pressure as the ambient ocean. In particular, high spatial resolution in situ measurements of dissolved oxygen and argon were obtained in the upper 200 m of the water column over long distances in the Gulf of Mexico and Atlantic ocean during research cruises on URI?s R/V Endeavor using a Triaxus tow sled provided by the University of Minnesota, Duluth. By comparing the changes in argon concentration (which are due solely to physical processes in the ocean) to the variations in oxygen concentration (which are caused by both biological and physical processes) the net effect of changes due to the biological processes that produce and consume oxygen can be deduced. SRI?s has completed its support of the project by providing in situ measurements using underwater mass spectrometry techniques, developed at SRI, on two research cruises, and working with URI personnel to develop new data processing algorithms. URI will continue data analysis through 2018 to produce the final outcomes of the project. Last Modified: 12/20/2017 Submitted by: Robert T Short