Performance Demonstration Statement Pro-Oceanus Systems Inc. PSI CO2-Pro.

Abstract

Alliance for Coastal Technology (ACT) demonstration projects are designed to characterize performance of relatively new and promising instruments for applications in coastal science, coastal resource management and ocean observing. ACT has evaluated four commercial pCO2 instruments that are capable of being moored for weeks to months. This document is termed a “Demonstration Statement” and provides a summary of the results for the Pro-Oceanus Systems Inc. PSI CO2-Pro. Briefly, test instruments were mounted on surface moorings in a temperate stratified estuary (Twanoh Buoy, Hood Canal Washington; August-September 2009; http://orca.ocean.washington.edu/mooringDesign.html;) and a coral reef (Kaneohe Bay Hawaii; October-November 2009; http://www.pmel.noaa.gov/co2/coastal/HI/). The sites were chosen based on existing moorings and the expected rapid changes in seawater temperature and pCO2. Water samples were collected to determine pH and Total Alkalinity (TA) for calculation of pCO2 (CO2Sys; Pierrot et.al. 2006) and direct measurements of pCO2 using a flow-through pCO2 analyzer (Oregon State University; gas equilibration and infrared gas detection). In situ pCO2 measurements are compared to both of these references and estimates of analytical and environmental variability are reported. Quality Assurance (QA) and oversight of the demonstration process was accomplished by the ACT QA specialists, who conducted technical, protocol and data quality audits. There were no test results for the Pro-Oceanus CO2-Pro for the Hood Canal, Washington deployment due to programming error in the time delay before the instrument was to initiate logging after deployment. The error was only discovered after retrieval. At the NOAA Crimp 2 buoy, Kaneohe Bay, temperature varied from 23.24 to 28.27 oC and salinity varied from 34.1 – 35.2 over the deployment. Measured pCO2 values of reference samples varied from 314 to 608 µatm, while the continuous hourly measured values for the CO2Pro varied from about 280 to 840 µatm, demonstrating a more complete assessment of the variability in the ecosystem. Due to a delayed start in the CO2-Pro deployment and a later maintenance problem with the Flow Analyzer, there were no data to compare differences for individual CO2-Pro values and the Flow Analyzer measurements. The mean and standard deviation of the difference for individual CO2-Pro determinations and the pCO2Sys reference measurements were +9 ± 14 µatm (n=29; CO2-Pro - pCO2Sys). The time-series provided by the instrument during the Hawaii field test (n=342) revealed strong diel patterns in pCO2 and captured a significantly greater dynamic range and temporal resolution than could be obtained from discrete reference samples (n=45). Accounting for data gaps during calibration cycles, nearly 100 percent of the data was retrieved and used for the Hawaii time-series. There were no systematic changes in the differences between instrument and reference measurements over time, indicating that biofouling and instrument drift did not affect measurement performance over the duration of the test