Hawaii Ocean Time-series (HOT) Carbon Dioxide

All data in this product are from Station ALOHA (22°45'N, 158°00'W). HOT program background information and details of sampling strategy may be found in Karl and Lukas (1996) and Karl et al. (2001). Results from the HOT CO2 measurement program may be found in Winn et al. (1994, 1998) and Dore et al. (2003, 2009, 2014).

The data in this product represent mean surface (0-30 dbar) values for each HOT cruise except where otherwise indicated.

Temperature and salinity data are from a calibrated Seabird CTD (Lukas and Santiago-Mandujano 1996); the individual values to be averaged were determined during the upcast at the time of the bottle trip except where otherwise noted. All seawater samples were collected using a rosette package lowered on a conducting hydrowire except as otherwise noted. Phosphate and silicate were measured colorimetrically via autoanalyzer using standard HOT methodology (Karl and Lukas 1996; Dore et al. 1996). Phosphate and silicate were assumed to be 0.07 and 1.04 μmol kg-1 (approximate mean values), respectively, where data were not available.

Dissolved inorganic carbon (DIC; equivalent to total CO2) and total alkalinity (TA) were measured by coulometry and open cell acid titration, respectively. Since 1993, a semi-automated SOMMA system (Johnson and Wallace 1992) was used for the coulometric DIC measurements. Certified seawater CO2 reference materials (Dickson 2001) were used to validate DIC and TA calibrations. The spectrophotometric m-cresol purple method was used for direct pH measurements at 25 °C (Clayton & Byrne 1993). The reagent grade dye was sourced commercially and NOT further purified. It has been suggested that the absolute calibration of the m-cresol purple pH method may be slightly biased such that a +0.0047 correction to measured pH values is necessary (DelValls & Dickson 1998); we did NOT apply this correction factor to our pH measurements. All pH measurements since 2003 have been made with the same lot of dye (Aldrich Chemicals product# 21176-1, Lot# 01920CX). We have determined the value of 434Aimp according to Douglas and Byrne (2017) to be 0.003253 but have NOT applied a dye impurity correction to these data.

Sampling, sample storage and analytical protocols for all CO2 system measurements followed internationally accepted recommendations (Dickson et al. 2007) with only minor modifications. Salinity-normalized values for DIC and TA were calculated as: nDIC = 35*DIC/salinity and nTA = 35*TA/salinity. CO2 system calculations were carried out with the CO2SYS software (Excel version 2.1; Pierrot et al 2006), using the Mehrbach et al. (1973) formulations of the apparent dissociation constants of H2CO3 in seawater, as refit by Dickson and Millero (1987), and the total boron formulation of Uppström (1974). For all calculations made at in situ conditions a pressure of 10 dbar was assumed. Calcium carbonate saturation state is calculated as (Pierrot et al. 2006): Ω = [CO32-]*[Ca2+] / Ksp where the calcium ion concentration is estimated from salinity and Ksp is the solubility product for the given carbonate mineral (aragonite or calcite).

A few missing DIC and TA data points during the early years of the time-series have been filled with values from C. D. Keeling’s parallel Station ALOHA CO2 measurement program (Keeling and Guenther 1994; Keeling et al. 2004), as described in Dore et al. (2003, 2009) and noted in the data product file. The HOT and Keeling measurements have been carefully intercompared (Dore et al. 2001). The Keeling dataset (through November 2016) is maintained by R. F. Keeling and may be found at:

  https://scrippsco2.ucsd.edu/data/seawater_carbon/ocean_time_series.html

When using this data product please cite as adapted from:

  Dore, J.E., R. Lukas, D.W. Sadler, M.J. Church, and D.M. Karl. 2009. Physical and biogeochemical modulation of ocean acidification in the central North Pacific. Proc Natl Acad Sci USA 106:12235-12240.

Please direct questions or comments regarding this data product to John Dore: [e-mail]