Search the data

Extended Ellett Line Dissolved Oxygen Concentration for RRS Discovery cruise D312

Originator's Data Acquisition and Analysis

Dissolved oxygen concentrations were determined using the Winkler whole bottle titration method with the aim of providing 1) accurate measurements of in-situ dissolved oxygen concentrations, 2) a calibration tool for the oxygen sensor deployed on the CTD and 3) and an extension to the time series of oxygen measurements obtained from previous Ellett Line cruises.

All seawater samples were drawn through short pieces of silicone tubing from the Niskin bottles into clear, pre-calibrated borosilicate glass bottles (approx 100 ml in volume). The temperature of each sample was measured with a handheld digital thermometer prior to the addition of 1 ml manganous chloride and 1 ml of alkaline iodide (reagents made following Dickson 1994). Glass stoppers were then inserted taking care to avoid bubbles and/or headspaces in each bottle and the bottle was shaken vigorously for 30 seconds. All Niskin bottles were sampled from every CTD cast. Samples were taken back to the chemistry lab and left for at least 1 hour before analysis. If time permitted, bottles were left for 1-2 hours before being given a second shaking and then left allowing the precipitate to settle.

Analysis of all samples used the Winkler whole bottle titration method with spectrophotometric end-point and started with the addition of 1 ml of dilute sulphuric acid (280 ml concentrated H2SO4 made up to 1 L with Milli-Q water) to each sample bottle. A small magnetic bar was added to each bottle to facilitate the dissolution of the precipitate, which was aided by placing each bottle onto a magnetic stirring plate. Each bottle was then titrated with sodium thiosulphate until clear. The titration volume (ml) was recorded and used with the initial fixing temperature in the calculations of Dickson (1994) to calculate the dissolved oxygen concentration.

One batch of sodium thiosulphate was made up during the cruise (25g L-1) and was tested daily for stability. This was achieved by titrating the sodium thiosulphate against 5 ml of certified potassium iodate standard. The daily mean (n = 5) volume of thiosulphate required to titrate the iodate standard was then used in the calculation of dissolved oxygen concentration for all samples collected on that day.

At each station a number of replicates were taken. Replicate reproducibility improved during the cruise with experience and the cruise mean percentage difference between replicate samples was 1 ± 1%.

References Cited

Dickson, A.G. 1994. Determination of dissolved oxygen in seawater by Winkler titration. WOCE operations manual, WOCE Report 68/91 Revision 1 November 1994

BODC processing

The data were received in 64 pstar files which were loaded into the BODC database using established BODC data banking procedures without any changes.

The mapping between the originator's channels and BODC parameter codes is detailed in the table below.

Data Quality Report

There was a slight drift in the data that occurs between stations 16143 and 16148, the problem seemed to lie in the CTD sensor calibration.