AFSC/ABL: Ocean Acidification in Southeast Alaska

Alaska, Gulf of Alaska, Southeast Alaska, Prince William Sound, Kodiak

Note: Dataset migrated by Dan Woodrich (AFSC data management coordinator) on 12/15/2021. Contact: Daniel.woodrich@noaa.gov

Water samples were collected at each location and depth with Niskin bottles and stored as described by Dickson et al. (2009) except they were not poisoned with mercuric chloride. Water was filtered through 53 m nylon sleeves placed around the exit hose on the Niskin bottle as it was collected from bottom to top with 0.5 to 1 volume overflow in pre-cleaned BOD (biological oxygen demand) bottles (after rinsing). The glass-stoppered BOD bottles were sealed with Apiezon L grease, refrigerated, and analyzed within 2 w of collection. Observation parameters did not change measurably within the 2 week measurement period. Additional water was collected unfiltered for chlorophyll measurement in brown Nalgene bottles (250 ml). Chlorophyll samples were refrigerated, filtered within 24 h of collection (0.7 m glass microfiber Whatman GF/F), placed in cryovials and stored at -80C pending analysis.

Water samples were analyzed for pH, total alkalinity (TA), total dissolved inorganic carbon (TC) and salinity with the methods of Dickson et al. (2009). Quality assurance measures were included daily with every group of samples. These measures included sample replicates for precision and for accuracy, certified reference material (CRM) for TA, TC, and salinity (Oceanic CO2 Quality Control - Marine Physical Laboratory, Scripps Institution of Oceanography, CA).

pH was measured in 10 cm cells with the indicator dye m-cresol purple in a UV/visible spectrophotometer (Model 8453, Agilent Technologies). A tris buffer solution in synthetic seawater served as the CRM (Oceanic CO2 Quality Control - Marine Physical Laboratory, Scripps Institution of Oceanography, CA). Typical precision, measured as the coefficient of variation, was 0.08%. Typical accuracy was 0.06%.

Total alkalinity was initially measured by the open cell titration method with a Dosimat autopipette (Metrohm) through August 2011 and thereafter with a QC Titration Module a TitraSip System (Mantech Associates Inc.). A series of pH buffer solutions was used to calibrate the Ross pH probe (Thermo Scientific) and the probe was subsequently soaked in filtered seawater for 1 h before sample measurement. A 0.7 M NaCl solution was used as a rinse between seawater sample measurements. Conditioning soaks were not used with the manual TA analysis approach. Typical precision was 3.6% with the Dosimat and 0.5% with the Mantech. Typical accuracy was 1.3% with the Dosimat and 0.7% with the Mantech.

Total dissolved inorganic carbon was measured with a CO2 Coulometer (CM5014, UIC Inc.) and Acidification Module (CM5130, UIC Inc.; Dickson et al. 2009). Each 20 g water sample was acidified with 6 ml of 1.5 M phosphoric acid to release the inorganic carbon as CO2 gas. An electrochemical cell with cathode/anode solutions and potassium iodide salt was used to measure the amount of CO2 (g C). Potassium hydroxide (45%) was used as a pre-scrubber and acidified potassium iodide (50% to pH 3) as a post-scrubber on the acidification module. High and low concentrations of sodium carbonate were used to calibrate the Coulometer, and a slope calibration factor was applied to each inorganic carbon value (Goyet and Hacker 1992). Typical precision and accuracy were 0.9 and 0.5%, respectively.

Chlorophyll a was measured as described by (Parsons, Maita et al. 1984) in a fluorometer (TD 700, Turner Designs). Each sample was measured twice and results averaged. Salinity was measured with an AutoSal (Model 8400B, Guildline Instruments Limited). Each sample was measured in triplicate and results averaged. Typical precision and accuracy were 0.7 and 7%, respectively.

Calcite and aragonite saturation levels were calculated from pH and TC measurements with the CO2sys (Lewis and Wallace). Alternative calculations as a function of TA and TC were also completed.