1. Introduction

Various gases present in the Earth's atmosphere, such as water vapor (H₂O), carbon dioxide (CO₂), ozone (O₃), nitrous oxide (N₂O),a methane (CH₄), and chlorofluorocarbons (CFCs), absorb thermal infrared (IR) radiation, creating the phenomenon commonly called the "greenhouse effect." Human industrial and agricultural activities have led to a rapid increase in concentrations of these "greenhouse gases," raising concerns over potential climate change. Of the greenhouse gases, CO₂ is the most important in terms of future global warming. Only about half of the anthropogenic CO₂ emitted remains in the atmosphere; the remainder is absorbed by the ocean and terrestrial biosphere. An excellent summary of the current state of knowledge on the greenhouse gases and projected increases can be found in the reports of the Intergovernmental Panel on Climate Change (IPCC) (Houghton et al. 1995, 2001). Predicting possible global climate change caused by CO₂ emissions requires forecasts of atmospheric CO₂ growth. This in turn necessitates obtaining temporal and spatial data from oceans and land that can be used to model the sequestration and storage of CO₂ in the oceans and terrestrial biosphere. Analysis of existing CO₂ measurements is fundamental to understanding the various uptake and storage processes (sinks) for CO₂ that have been observed to change on seasonal to decadal time scales. Future decisions on regulating emissions of greenhouse gases should be based on accurate models that have been adequately tested against accurate measurements.

During the 1990s, measurements of the oceanic inorganic carbon system, which are composed of total dissolved inorganic carbon (DIC), fugacity of CO₂ (ƒCO₂),1 total alkalinity (TAlk), and pH, were taken on the World Hydrographic Program (WHP) cruises of the World Ocean Circulation Experiment (WOCE) and those of the Ocean-Atmosphere Carbon Exchange Study (OACES) of the National Oceanic and Atmospheric Administration (NOAA) (Fig. 1). These measurements have provided a benchmark of unsurpassed accuracy for the ocean inventory of CO₂ and other properties. The inorganic carbon measurements performed by U.S. investigators were cosponsored by NOAA and the U.S. Department of Energy (DOE) as part of the U.S. Joint Global Ocean Flux Study (US JGOFS) Program. In addition to the U.S. cruises, the Atlantic synthesis included a significant number of cruises sponsored by the science agencies of the foreign nations. This report addresses the consistency of oceanic inorganic carbon system parameter measurements taken from 1990 to 1998 in the northern and southern Atlantic Ocean and lists adjustments to some of the DIC and TAlk measurements based on careful analysis of the full data set.

The analysis of the large-scale data quality of inorganic carbon system parameters for the Atlantic syntheses data set followed the procedures outlined in Lamb et al. (2001) and Feely et al. (1999) with the objective of determining the consistency of inorganic carbon data among the different cruise data. The focus was on the DIC and TAlk state variables used in the calculation of the anthropogenic CO₂ inventory and for studies of biogeochemical carbon cycling. Four approaches were followed to determine whether there were systematic offsets in the cruise data sets.

A. Inorganic carbon system values in deep water were compared where cruise tracks cross, hereafter referred to as "crossover analyses."

B. Multiple-parameter linear regressions (MLRs) of DIC or TAlk with potential temperature, salinity, oxygen, silicate, and nitrate were created from data of cruises that a similar cruise track. The calculated values were then compared with the measured parameters for each of the cruises.

C. On cruises where more than two carbon system parameters were measured, the internal consistency between parameters was evaluated by using known thermodynamic relationships between the paramters.

D. Finally, regional MLR regressions of DIC or TAlk with potential temperature, salinity, oxygen, silicate, and nitrate were created from all data in a particular region; date deemed good based on the previous checkswere used. These fits were used together with hydrographic data from individual cruises to investigate differences between the calculated DIC or TAlk and the measured values.

The cruise lines used are shown in Fig. 1, and the distribution of data vs time and latitude are presented in Fig. 2 and Fig. 3, respectively. Observations decreased in 1995 and 1996 because the WOCE/WHP and OACES programs were focusing on the Indian Ocean during those years. The observations vs latitude show a reasonably uniform coverage for the carbon system parameters. Of note is the absence of alkalinity data in the Atlantic sector of the Southern Ocean.

In the crossover analyses, the four inorganic carbon system parameters (DIC, ƒCO₂, TAlk, and pH) were compared in density space referenced to 4000 dB (σ₄) at 53 locations where cruises overlapped throughout the Atlantic Ocean (Fig. 1, Table 1). Such comparisons have been made for oceanic carbon parameters in the Indian Ocean (Johnson et al. 1998, Millero et al. 1998, Sabine et al. 1999) and the Pacific Ocean (Lamb et al. 2001). Similar comparisons are under way for nutrient data (Gordon et al. 1998) and CFC data (Smethie, personal communication).

The analyses presented in this report are the basis for recommending adjustments to the data sets to form a consistent basin-wide unified data set. The corrected working data set, which includes the original carbon data as well, is provided as two large datafiles at the following Web site: http://www.aoml.noaa.gov/ocd/gcc/atlantic_synthesis.html. The Atlantic data set in combination with those from the Indian and Pacific oceans will provide the first comprehensive global data set of DIC and TAlk to the research community. The Ocean Carbon Data System (OCADS) Web site will house all three corrected data sets. Data from the individual cruises can be found on the CDIAC site as well.