UC San Diego

In situ autonomous chemical sensors, combined with the right deployment platforms provide novel, powerful tools for oceanographers to observe biogeochemical processes on unprecedented spatial and temporal scales. However, many aspects of chemical sensor technology have not yet reached full maturity, preventing routine use by the community at large. This dissertation aims to fill this critical need in ocean observing technology, with a focus on Ion Sensitive Field Effect Transistor (ISFET) pH sensors for profiling float applications. Following a brief introduction to the current status of marine chemical sensor technology, the four chapters address the various steps involved in sensor development: sensor characterization, calibration, data quality control (QC), and a modeling effort using sensor data. Chapter 2 introduces a simple QC protocol for profiling float oxygen data by comparison to a monthly climatology. This protocol can constrain O₂ at the surface to better than 3%, and detect sensor drift with high confidence. Similar approaches can be taken to QC other chemical sensors data from profiling floats. Chapter 3 characterizes the response of the ISFET pH sensor and the Chloride-Ion Selective Electrode by comparison to the hydrogen electrode and the silver-silver chloride electrode, respectively. Both electrodes showed near-Nernstian response, thus the error in pH due to non-theoretical behavior of the electrodes is negligible over the oceanic range of pH and salinity. Chapter 4 quantifies the effect of pressure on the pH of certified tris buffer prepared in synthetic seawater. Assignment of pH values to certified buffer solutions is essential for sensor calibration. As the number of pH sensors deployed under high pressures is expected to increase, this chapter will fill a critical need in sensor validation and traceability. Chapter 5 presents habitat-specific ocean acidification projections between 2012 and 2100 for 4 habitats in the upper 100 m of the Southern California Bight. The projections were generated by combining high frequency pH sensor data, a regional empirical relationship of the CO₂ system, and hydrographic data to characterize the properties of upwelled waters. Habitat specific acidification signals were predicted, and implications for future ocean acidification research are discussed