Dataset: Arabian Gulf Carbonate System Chemistry

Sampling
Water column sampling using rosette mounted 10-L PVC Niskin bottles was conducted on December 5, 2018 and May 18, 2019 in the Arabian Gulf on the R/V Janan (Figure 1). Surface water samples and hydrographic data were collected at seven stations (stations 1C, 2C, 3C, 4C, 5C, 6B, 6C) along a transect from the central east coast of Qatar across the Qatari Exclusive Economic Zone (EEZ). Stations were chosen to be nearly  perpendicular to the major axis of the Gulf to capture main regional hydrographic features across the EEZ. The transect provides a reasonable representation of hydrographic distributions across the wider part of the Gulf. Samples were collected in triplicate at each station. Vertical profiles with one surface sample, one bottom sample, and 1 to 3 mid-depth samples were collected at stations 2C, 4C, 6B, and 6C.

Samples for DIC and total alkalinity were collected in 300 mL Wheaton BOD glass bottles with ground glass stoppers. Samples were sealed immediately after collection to prevent loss of CO₂. Samples were poisoned with 50 µL HgCl₂ (0.05% by volume) to prevent biological activity, then covered with aluminum  foil to eliminate light and biological growth. 

Analysis
The best approach for understanding ocean acidification is to measure the primary capacity factors of the carbonate system chemistry which are dissolved inorganic carbon (DIC) and total alkalinity (TAlk). DIC and alkalinity are the best set of parameters use for calculating pH, pCO₂ and carbonate ion, because they are conservative properties during water mass mixing.

After collection, samples were shipped to the University of Washington for DIC and TAlk analyses in Dr. Alex Gagnon’s laboratory. Carbonate system measurements follow the methods of Dickson et al. (2007). Briefly, TAlk (μmol kg^-1) was determined through open-cell automated titration (876 Dosimat plus, Metrohm AG) with a solution of 0.1M hydrochloric acid (HCl)+0.6M sodium chloride (NaCl). Total DIC (μmol/kg) was obtained through coulometric determination (VINDTA 3D, Marianda with UIC coulometer). Certified reference materials for TAlk and DIC obtained from Andrew Dickson (Scripps Institution of Oceanography) were run in conjunction with seawater samples as a calibration standard and to monitor precision. Long-term precision for DIC and TAlk in this lab, based on repeated measurements of CRM materials, was ± 3.7 μmol kg^-1 (2σ std. dev.) and ± 4.3 μmol kg^-1 (2σ std. dev.), respectively (Bolden et al., 2019).

 pCO₂ and pH were calculated from DIC and Alk using CO2Calc using the total pH scale with carbonate equilibrium constants refit from Mehrbach et al. (1973) by Dickson and Millero (1987); borate alkalinity was calculated using the boron/chlorinity (salinity) relationship provided by Lee et al. (2010) and equilibrium constants from Dickson (1990).  Where necessary, NTAlk, and NDIC values used in subsequent calculations were salinity-normalized to a mean salinity value of 40.0. Data analysis was executed using Microsoft Excel. Certain data properties were calculated using CO₂Calc. This includes parameters such as pH, CO₃^2-, and pCO₂.

During both cruises, samples were also collected and analyzed for: Hydrographic properties using a SeaBird Electronics, SBE 911 on a CTD SeaBird rosette (T, S, O₂, pH, % Transmission and fluorescence). The pH sensor used in our project is the SBE 27 pH/O.R.P (Redox) Sensor.

Dissolved oxygen was measured using the SBE 43 Dissolved Oxygen sensor. Additional discrete samples were analyzed for dissolved oxygen within a few hours of collection using the titrimetric method (Winkler, 1888). Nutrients (NO₃, NO₂, NH₄, PO₄, and Si) and chlorophyll were analyzed on filtered samples using classical techniques (Parsons et al., 1984).

Suspended particulate matter for particulate Ca analyses was sampled using 47 mm filter holders and 0.45 µM mesh size Nuclepore filters. Filtration volumes of 2 L, provided sufficient samples for analyses.  Particulate samples were acid digested in a clean lab on hot plates using trace metal grade concentrated HF (16.5M), HCl (6M) and, HNO₃ (16M) acids (Yigiterhan and Murray, 2008; Yigiterhan et al., 2011, Yigiterhan et al., 2018). H₂O₂ was added for complete removal of the organic material. The elemental analyses were performed using ICP-OES.