Dataset: Elemental mercury in the tropical Pacific Ocean

Surface water samples were obtained using water collected at 5-10 m depth from the ship’s underway sampling system. The water is continuously sparged with low Hg0 air in a water-gas separation device (gas equilibrator) with a reverse flow configuration where the water is added to an inner chamber at the top and air at the bottom of the chamber. The water then flows through an outer jacket of the device to maintain the water temperature in the inner chamber at that of the incoming sweater. The device is based on the equilibrator described in detail in Andersson et al. (2008). For the best performance the water flow should be high relative to that of the air, and in this study the air flow was maintained around 1.5 L min-1, somewhat higher than that required by the Tekran 2537B instrument that was used for Hg detection (1 L min-1). The water flow was somewhat variable but typically was 6-10 times that of the air. The incoming air is passed through spargers that create small gas bubbles to enhance equilibration and this is also enhanced by the mixing induced by the incoming water. Based on the relative flow rates, the response time of the instrument to changes in water concentration is <5 min (Andersson et al., 2008) so changes between a set of measurements could reflect changes in the water on the same timescale. Depending on the ship’s speed, this represents a spatial sampling resolution of 1-2 km for a 5 minute sample. Based on our experience, measurements while the ship is stationary are often higher and more variable, and so these measurements are not considered reliable and are not included in the database. The air is dried using a Teflon filter and a soda lime trap prior to the passing to the detector – a Tekran 2537B mercury analyzer.
The detection of Hg as elemental Hg in the air after sparging relied on another Tekran 2537B instrument with a sampling resolution of 5 min. As described above, the instrument is calibrated in two ways. Air is sampled continuously as there are two sampling gold traps lines within the instrument and while one sample is being analyzed, the other is being trapped, with the timing controlled by the instrument’s software. Air was sampled from the outside at a location sufficiently above the water level to prevent entrainment of water, and in a position to prevent contamination for the ships’ exhaust while underway. The air is dried using a Teflon filter and a soda lime trap prior to the detector.