Carbon monoxide (CO) is produced photochemically in the surface ocean and emitted to the atmosphere. To assess the magnitude of this ocean‐atmosphere flux, seawater and atmospheric CO mole fractions were measured on six cruises throughout the Pacific Ocean from 1987 to 1994. The results showed consistent regional and seasonal variations in surface seawater CO concentrations with daily averaged concentrations ranging from 0.1 to 4.7 nM. Based on the concentration fields, the data were divided into four seasons and 10 latitude zones from 75°S to 75°N. Using monthly Comprehensive Ocean‐Atmosphere Data Set wind and surface seawater temperature data and the Wanninkhof [1992] wind speed/transfer velocity relationship, the calculated zonal average fluxes ranged from 0.25 to 13 μmol/m2/d. The combined seasonal and zonal fluxes result in a total global flux of 0.46 Tmol CO/y with 2/3 of this flux in the southern hemisphere. The estimated uncertainty in this number is approximately a factor of 2.

Seawater and atmospheric methane (CH4) mixing ratios were measured on five cruises throughout the Pacific Ocean from 1987 to 1994 to assess the magnitude of the ocean‐atmosphere flux. The results showed consistent regional and seasonal variations with surface seawater concentrations ranging from 1.6 to 3.6 nM and saturation ratios ranging from 0.95 to 1.17. The equatorial Pacific Ocean was supersaturated with respect to atmospheric CH4 partial pressures, while areas outside the tropics often were undersaturated during fall and winter. Although atmospheric CH4 mixing ratios over the North Pacific during April increased 3.4% from 1988 to 1993, the saturation ratios remained constant. Based on the concentration fields, the data were divided into two seasons and 10 latitude zones from 75°S to 75°N. Using monthly Comprehensive Ocean‐Atmosphere Data Set (COADS) wind and surface seawater temperature data and the Wanninkhof [1992] wind speed/transfer velocity relationship, the calculated zonal average fluxes ranged from −0.1 to 0.4 μmol m−2 d−1. The combined seasonal and zonal fluxes result in a total global ocean‐to‐atmosphere flux of 25 Gmol yr−1 (0.4 Tg CH4 yr−1), which is an order of magnitude less than previous estimates [Intergovernmental Panel on Climate Change (IPCC), 1994]. The estimated uncertainty in this number is approximately a factor of 2.

A major purpose of the third joint Soviet‐American Gases and Aerosols (SAGA 3) oceanographic cruise was to examine remote tropical marine O3 and photochemical cycles in detail. On leg 1, which took place between Hilo, Hawaii, and Pago‐Pago, American Samoa, in February and March 1990, shipboard measurements were made of O3, CO, CH4, nonmethane hydrocarbons (NMHC), NO, dimethyl sulfide (DMS), H2S, H2O2, organic peroxides, and total column O3. Postcruise analysis was performed for alkyl nitrates and a second set of nonmethane hydrocarbons. A latitudinal gradient in O3 was observed on SAGA 3, with O3 north of the intertropical convergence zone (ITCZ) at 15–20 parts per billion by volume (ppbv) and less than 12 ppbv south of the ITCZ but never ≤3 ppbv as observed on some previous equatorial Pacific cruises (Piotrowicz et al., 1986; Johnson et al., 1990). Total column O3 (230–250 Dobson units (DU)) measured from the Akademik Korolev was within 8% of the corresponding total ozone mapping spectrometer (TOMS) satellite observations and confirmed the equatorial Pacific as a low O3 region. In terms of number of constituents measured, SAGA 3 may be the most photochemically complete at‐sea experiment to date. A one‐dimensional photochemical model gives a self‐consistent picture of O3‐NO‐CO‐hydrocarbon interactions taking place during SAGA 3. At typical equatorial conditions, mean O3 is 10 ppbv with a 10–15% diurnal variation and maximum near sunrise. Measurements of O3, CO, CH4, NMHC, and H2O constrain model‐calculated OH to 9 × 105 cm−3 for 10 ppbv O3 at the equator. For DMS (300–400 parts per trillion by volume (pptv)) this OH abundance requires a sea‐to‐air flux of 6–8 × 109 cm−2 s−1, which is within the uncertainty range of the flux deduced from SAGA 3 measurements of DMS in seawater (Bates et al., this issue). The concentrations of alkyl nitrates on SAGA 3 (5–15 pptv total alkyl nitrates) were up to 6 times higher than expected from currently accepted kinetics, suggesting a largely continental source for these species. However, maxima in isopropyl nitrate and bromoform near the equator (Atlas et al., this issue) as well as for nitric oxide (Torres and Thompson, this issue) may signify photochemical and biological sources of these species.

Measurements of CO 2 concentrations in the atmosphere and in the surface waters of the South Pacific Ocean were made by NOAA scientists between 1984 and 1989. These basin-wide measurements were all taken during austral autumn and provide data for evaluation of the seasonal flux of CO 2 from this region. The sensitivity of this flux to the uncertainty in the CO 2 gas exchange coefficient was evaluated using four different wind data sets and two formulations for the wind dependence of gas transfer velocity. The integrated net flux of CO 2 to the atmosphere during austral autumn (February to May) ranges from -0.03 (ocean influx) to +0.09 (ocean effiux) GT of carbon depending on the combination of wind field and wind-dependent exchange coefficient used. INTRODUCTION Gas flux across the air-sea interface is dependent on twoMajor spatial and temporal gaps in the measured distribution factors: the difference in air and sea concentrations (ApCO2), of CO 2 in the surface waters of the South Pacific Ocean have and the rate of gas transfer across the boundary. The former hindered an assessment of this ocean as a source or sink for determines the potential for gas flux and the sign of the flux, CO 2. The first annually averaged, global ApCO 2 map was made by Keeling [1968] using relatively sparse data. The large-scale features for the South Pacific are equatorial oversaturation, undersaturation in the subtropical gyre regions, and oversaturation in the polar region. Takahashi et al. [1986] i.e., flux into the ocean (negative) or flux out of the ocean (positive). The other component of flux, gas transfer rate, determines to what extent gas transfer will actually occur. It is not measured directly but is calculated from wind speed using an empirical, independently determined relationship subsequently filled in some of the data gaps and extrapolated between wind speed and gas transfer velocity. As more where necessary to create an annually averaged, zonal map of seasonal measurements of CO 2 become available, the limiting ApCO 2. This map confirms the equatorial oversaturation found factor in accurately assessing fluxes becomes that of assessing by Keeling [1968] but shows undersaturation in both the the gas transfer rate. Current formulations of gas transfer subtropical and polar regions. Additional measurements of velocity depend on wind speed, which is poorly known in the southern hemisphere. Wind speeds from direct observations ApCO 2 have permitted separation into seasonal maps [Tans et al., 1990], but large data gaps in the South Pacific still exist. are available sporadically, and very infrequently, in the South Since the role of this ocean is poorly understood, it is one of Pacific. Wind speed can also vary enormously over short time the major uncertainties in estimating global fluxes and scales, and in consequence, a large number of measurements assessing the role of the global ocean in absorbing atmospheric CO 2 [Tans et al., 1990]. Recently, general circulation models have been used to infer oceanic sources and sinks...

The equatorial Pacific Ocean is a source of both sulfur and carbon to the atmosphere. In February and March 1990, as part of the Soviet‐American Gases and Aerosols (SAGA 3) expedition, dimethysulfide (DMS), methane (CH4), carbon monoxide (CO), and carbon dioxide (CO2) partial pressures were determined in both surface seawater and the overlying atmosphere of the central equatorial Pacific Ocean (15°N to 10°S, 145°W to 165°W). The partial pressures were used to calculate the net flux of these gases from the ocean to the atmosphere. The average regional DMS and CO fluxes were similar, 7.1 and 4.2 μmol/m2/d, respectively. The mixing ratio of CH4 in surface seawater was close to equilibrium with the overlying atmosphere and hence the average flux was only 0.39 μmol/m2/d. The flux of CO2 clearly dominated the air‐sea carbon exchange with an average regional flux of 5.4 mmol/m2/d.