Dataset: Seawater Trace Metal Concentrations from China incubation experiment April 2011

Seawater for incubation experiments was collected at two sites located in and near the Yangtze River plume in April 2011 aboard a local fishing boat. Site 1 was within the plume, and site 2 was further away from the river discharge point. Site 1 was nearer to the mouth of the Yangtze River, between two islands in an area with visible suspended sediment (30° 43′ 04″N, 122° 39′ 58″E). Site 2 was east of the Zhoushan Archipelago in waters influenced by the Kuroshio Current outside the area of high sediment resuspension from the Yangtze River, and had a salinity of 31.7ppt (30° 38′ 12.2″N, 122° 56′ 8.51″ E). Site 1, being more heavily influenced by the river discharge, had a salinity of 28.7ppt. Assuming river water has a salinity of zero and ocean water has a salinity of 33, the water we collected at site 1 was a mixture of 87% seawater and 13% river water, while site 2 was 96% seawater and 4% river water. Both of these sites are clearly affected by the Yangtze River outflow, and this is reflected in their salinity levels and nutrient characteristics.

Trace metal clean techniques were used to collect incubation water and during all handling and sample processing steps. Acid cleaning included a 1 week Citranox soaking step, followed by a 1week 10% hydrochloric acid soaking step, followed by a thorough milliQ rinse. All containers and equipment were stored double bagged until use after acid cleaning. For each experiment, surface seawater was collected using a peristaltic pump with acid cleaned C-Flex tubing, passed through 80 μm mesh into acid clean, sample rinsed 50L carboys, and transported to Shengsi Island in the dark. Transport took less than 2 h. In the lab, water was dispensed into 500 mL acid clean, sample rinsed, clear polycarbonate bottles. Nutrient additions included nitrogen (N, as 10 μM NaNO3 + 0.5 μM (NH4)2SO4), phosphorus (P, as 0.5 μM NaH2PO4), or N + P together, and these concentrations were designed to approximately double the background concentrations of these nutrients observed at site 1. Control bottles contained no nutrient additions.

Aerosol treatments included aerosols representing 5 different dates of collection (see below). To simulate 10 days of a moderately strong deposition event for this region (1 g m−2 d−1) over a 10 m mixed layer, 1 mg aerosol on the filter was added per L of seawater. The aerosol filter was cut to the correct size to deliver the appropriate amount of aerosol and then added directly to the bottles. No prior leaching step was included to increase the solubility of the aerosols. The annual average deposition flux for the region is 4.6–98 g m−2 y−1. The deposition rate assumed for the additions is thus an order of magnitude above average and represents strong wind or dust storm conditions. For aerosol NW37, concentrations of 0.2 mg L−1 (similar to average or typical loading at the site) and 5 mg L−1 (very high deposition, strong dust storm conditions) were also included to test phytoplankton responses to aerosol dosing. Bottles were incubated in a flowing seawater pool by pumping coastal seawater into the pool at a rate high enough to maintain ambient coastal seawater temperatures without allowing the pool to warm from sunlight. No attempt to artificially manipulate temperature was made. The pool was shaded with a neutral density shade cloth to attenuate sunlight 50% for both site 1 and site 2 experiments. Three bottles per treatment were sampled at “time zero” (collected immediately after nutrient or aerosol additions were made) and on the final day of the incubation 3 days later. The incubation time of 3 days was selected because it is long enough to observe changes in phytoplankton community structure based on prior studies, while avoiding major bottle effects that occur with longer-term incubation experiments.

Water samples for trace metal analyses from the incubation experiments were filtered (0.2μm PES) in a laminar flow hood following trace metal clean techniques. Sample bottles were stored in clean plastic bags and stored frozen.

To measure trace metals in the incubation seawater, samples from the incubation experiments and operational field blanks (MilliQ water, see below) were acidified to pH < 2 with concentrated trace metal grade nitric acid at least 48 h before processing. Trace metals were concentrated using Nobias Chelate-PA1 resin (HITACH, Japan) to remove the seawater matrix, and eluted in 5 mL nitric acid following (Sohrin et al., 2008). Concentrated samples were analyzed by ICP-MS using standards prepared in 5% nitric acid as described previously (Mackey et al., 2015). All sample preparation occurred within a class 1000 clean room in class 10 laminar flow hoods under HEPA filtration.

Operational field blanks of MilliQ water collected to determine if trace metal contamination was introduced during the course of the incubation. Incubation bottles were filled with 500 mL of MilliQ water and were incubated identically to the seawater samples. One bottle was collected every day of the experiment and these blanks were processed along with the seawater samples. The concentrations were 0.046 ± 0.028 nM Cd, 0.010 ± 0.0032 nM Pb, 3.2 ± 2.1 nM Al, 0.98 ± 0.28 nM Fe, 0.00024 ± 0.00 nM Co, 0.13 ± 0.088 nM Ni, 0.22 ± 0.12 nM Cu, and 1.8 ± 0.76 nM Zn (average ± standard deviation, n = 5). Separate MilliQ procedural blanks (including column chemistry through ICP analysis) were also performed. Reported values herein for seawater samples have been blank corrected using the procedural blank averages.