Report on the ices international intercalibration of mercury in seawater

Results are described of an intercalibration between fourteen laboratories on mercury in seawater. Two samples were used for the intercalibration, one of acidified natural seawater of low concentration and the other acidified seawater spiked with mercuric chloride. Agreement was poor on the natural sample but better on the spiked sample.     

An intercalibration for trace metals in seawater

This paper describes the results of a recent international intercalibration experiment for trace metals in seawater. The results show that considerable improvements in analytical ability have been made for several metals at concentrations close to those found in offshore marine waters. Some systematic differences between the results of analyses of frozen and acidified samples are evident for most metals. The application of two different analytical techniques, atomic absorption spectrophotometry and anodic stripping voltammetry, appears to introduce relatively small differences in the results for copper and cadmium at near-natural concentrations.Two factors limiting our ability to conduct trace metal intercalibrations at the levels found in the deep ocean are the extent of tolerable contamination, especially for zinc and lead, and the degree to which the overall homogeneity of a group of intercalibration samples can be assured. Some reexamination of methods of sample preparation needs to be carried out before such low level calibrations can be effectively conducted.     

Determination of mercury in seawater at sub-nanogram per liter levels

Using a refined two-stage gold amalgamation preconcentration technique in conjunction with cold vapor atomic absorption (CVAA) detection, mercury in various Pacific Northwest coastal waters has been determined. Through the use of extremely clean reagents and special handling techniques, the blank has been reduced and reproducibility greatly improved for both “easily reducible mercury” and “total mercury”.Samples of unfiltered water, taken at the surface, as well as at 10 m, 100 m, and 175 m have all yielded “easily reducible mercury” at between 0.1 and 0.5 ng l^?1, and “total mercury” between 0.2 and 1.0 ng l^?1. Variability within replicate field samples is less than 10%, although variations of greater than 100% may be seen over a period of hours in Sequim Bay.     

An in situ pump sampler for trace materials in seawater

A battery-powdered in situ pump sampler, capable of operation at full oceanic depths, has been designed, built and tested. The pump motor, sealed in a pressure-proof housing, is coupled magnetically to the pump head. Water is drawn through a filter then passed successively through an adsorption column and the pumphead and finally through a flowmeter. The unit as a whole is pivoted and vaned so that the intake is always pointed into the current. The selectivity of the sampler depends upon the choice of packing material used in the adsorption column or columns. Column materials specific for different classes of compounds have been tested in the laboratory and at sea. Preliminary results in the collection and measurement of Cu and Cd suggest that the pumping system is free from shipboard contamination.     

Mercury distribution in seawater in Minamata Bay and the origin of particulate mercury

Horizontal and vertical distributions of mercury were determined in seawater in Minamata Bay and Yatsushiro Sea. The concentrations of total and particulate mercury respectively ranged from 56–285 ng l^–1 and 2.1–506 ng l^–1. They were both highest in the inner most part of Minamata Bay where the sediment was most heavily polluted, and decreased with increasing distance from there. Vertically, an evident increase in concentration was observed near the bottom. A good agreement was found between the mercury content in suspended matter and that in fine sediment particles.These facts show that the particulate mercury is supplied mostly from the bottom sediment, and that it is spreading offshore with dispersing in seawater. This process would eventually result in the pollution of sediment in Yatsushiro Sea.     

An improved method for the determination of ammonia in seawater

The Liddicoat, Tibbitts and Butler method for the determination of ammonia in seawater has been noticeably improved by replacing the UV lamp with a 300 VA tungsten lamp. The characteristics so achieved, a low blank value (0.012 ± 0.002a.u.), a good molar absorptivity (23.90 × 10⁻³) and a reasonably short reaction time (1 h), are such that this modified procedure is more convenient for routine work in chemical oceanography.     

An improvement in the small-volume technique for determining thorium-234 in seawater

The recently developed 2–5 L small-volume MnO₂ coprecipitation method for determining ²³⁴Th in seawater has provided a new way to substantially increase, both temporally and spatially, the sampling resolution of ²³⁴Th and ²³⁴Th-based particulate organic carbon export estimates in the upper ocean. In this study, we further optimize the technique by reducing reagent quantities and the use of an additional water bath heating step. This optimization allows the filtration of the MnO₂ precipitate onto a 25-mm diameter, 1.0 μm pore size QMA filter to be completed within < 30 min for unfiltered waters from the South China Sea. In addition, we have modified the MnO₂ purification procedure to allow for alpha spectrometric measurements of ²³⁴Th recoveries. Results from recovery experiments suggest that reagent amounts can be reduced to 0.0375 mg KMnO₄ and 0.1 mg MnCl₂·4H₂O per liter of sample, while still maintaining high ²³⁴Th recovery. This study further confirms that the addition of a yield monitor is necessary for the application of the small-volume method.     

An intercomparison of small- and large-volume techniques for thorium-234 in seawater

In this paper, an intercomparison of methods for the determination of ²³⁴Th in seawater is discussed. Samples were collected either from a shore-based 600 m water source, or from standard bottle casts in deep waters off Hawaii and the Southern Ocean. We compared large-volume techniques, which rely upon Mn cartridges for the collection of dissolved ²³⁴Th and its detection via gamma counting (>200-l samples), with small volume methods that employed either direct beta counting, or beta counting after radiochemical purification (2–20-l samples). Unique to this study is the presentation of small volume (2 and 5 l) ²³⁴Th methods. This new technique is an adaptation of 20-l methods that are based on the coprecipitation of thorium with Mn oxides followed by direct beta counting of the precipitate. The small volume Mn coprecipitation methods were found to be superior to other methods due to ease of sample collection, processing and low overall analytical uncertainties.     

The association of copper,mercury and lead with surface-active organic matter in coastal seawater

Significant fractions of copper (6–30%), lead (3–12%) and mercury (4–50%) were associated with surface-active dissolved or colloidal organic matter isolated from the water column of a controlled experimental ecosystem. Changes in the concentration of surface-active forms of these metals were attributed to release of surface-active metal-binding organic matter during the collapse of a phytoplankton bloom and early diagenesis of settled detritus. Naturally occurring surface-active organic matter can be expected to significantly influence the chemical speciation and transfer of metals across air-water and solid-water interfaces in the marine environment.     

An ion-exchange technique for concentrating ammonia from small volumes of seawater

An ion-exchange technique was used to concentrate ammonia from seawater. Precipitation of bivalent cations, prior to concentration, reduced the variability of ammonia recovery from the ion-exchange column and lengthened column life. Analysis of the eluate was by the phenolhypochlorite reaction. The method was designed for use with small seawater samples (>10ml) and has the advantage that it can be adapted for use with small volumes of seawater with no loss in sensitivity. In a 10 ml seawater example NH₃-N concentrations of 10–150 μg l^?1 were accurately determined.     

Marine sampling by scuba diving. 3. Sampling procedures for measurement of mercury concentrations in estuarine waters and seawater

A method for sampling estuarine waters and seawater by scuba diving is described. Samples taken from the Krka River Estuary on the eastern Adriatic coast were analysed for mercury content using cold vapour atomic absorption spectrometry (CVAAS). Mercury levels as low as 1 ng dm⁻³ were determined and these are among the lowest values reported to date. A simple contamination free procedure for the preconditioning of sample containers in the sea was applied. Sampling of water by divers is particularly useful when the intermediate layer in an estuary has to be identified and sampled, as well as for obtaining detailed vertical concentration profiles for mercury and other trace elements.     

基于图相似性匹配的海水溶解氧辅助评价方法

海水溶解氧评价的正确性对于维护海洋生态系统的稳定性具有重要意义。海水溶解氧数据属于非线性时间序列,可视图方法(时间序列转化为图)是分析这类数据较有效的方法,但仍存在未同时考虑数据的时间演变特性以及变量间相互影响的问题。针对现存问题,本研究提出基于图相似性匹配的海水溶解氧辅助评价方法。首先,同时考虑海温、盐度对海水溶解氧的影响以及数据的不可逆性,提出溶氧温盐-转移概率(dissolved oxygen temperature salinity-transition probability, DOTS-TP)有向可视图方法,实现了多变量时间序列到单变量溶解氧图的转化;然后,在将墨西哥湾溶解氧图作为评价参照的基础上,综合利用图的多层次信息,提出子图节点全局(subtree node global, SNG)图相似性匹配方法,通过计算SNG评价指数实现对海水溶解氧的辅助评价。实验结果表明DOTS-TP有向可视图方法能更准确地表达海水溶解氧信息,SNG图相似性匹配方法适用于所有海水溶解氧数据,并能得出正确的辅助评价结果。     

An improved method for the hydrolysis and MBTH analysis of dissolved and particulate carbohydrates in seawater

An improved method for the bulk characterization of particulate and dissolved combined carbohydrates in seawater is presented. Seawater samples are dried under vacuum and the residue is hydrolyzed with 12 M H₂SO₄. Carbohydrates in the hydrolysate are subsequently quantified spectrophotometrically by MBTH (3-methyl-2-benzothiazolinone hydrazone hydrochloride) analysis. Sulfuric acid was more effective in hydrolyzing high-molecular-weight structural carbohydrates (cellulose, chitin, and alginic acid) than weak (0.09 N) hydrochloric acid. Sulfuric acid hydrolysis of dissolved and particulate samples yielded estimates of carbohydrates equal to or greater (two- to four-fold) than those resulting from dilute (0.09 N) HCl hydrolysis. In samples from the Gulf of Mexico, total dissolved carbohydrates determined after sulfuric acid hydrolysis varied from 20 to 56 μM C and from 10 to 28% of the dissolved organic carbon determined by high-temperature catalytic oxidation.     

An investigation of the suitability of amberlite XAD-1 resin for studying trace metal speciation in seawater

Amberlite XAD-1 resin was examined to test its suitability for extracting organic complexes of copper, zinc and iron from seawater. At low flow rates and at loading capacities far below theoretical values, the adsorption of these metals is not reproducible and the results are reminiscent of the behaviour observed when the adsorption capacity is being exceeded or flow rates are too high. It is suggested that the resin also adsorbs small but significant amounts of inorganic ions from seawater and that this effect makes the resin unsuitable for quantitative measurements of trace metal speciation.     

Evidence for organic complexation of iron in seawater

Iron occurs at very low concentrations in seawater of oceanic origin and its low abundance is thought to limit primary production in offshore waters (Martin and Fitzwater, 1988). A new electrochemical method, cathodic stripping voltammetry (CSV), is used here to determine the speciation of iron in seawater originating from the Western Mediterranean taking advantage of ligand competition of an added electroactive ligand with the natural organic complexing matter to evaluate whether iron is organically complexed. The measurements indicate that iron occurs 99% (or 99.9% depending on which value is selected for α_Fe′) complexed by organic complexing ligands throughout the water column of the Western Mediterranean and by analogy probably also in other oceanic waters. The composition of the organic complexing ligands is as yet unknown, but the data indicate a major source from microorganisms (bacteria or phytoplankton) in and immediately below the fluorescence maximum in the upper water column. The organic complexes are apparently reversible releasing iron when the competing ligand is added and binding more iron when its concentration is increased. The organic complexing ligands occur at concentrations well above those of iron ensuring full complexation of this biologically essential element, and buffer the free iron concentration at a very low level against fluctuations as a result of removal by primary producers or inputs from atmospheric sources. The new data indicate that a re-evaluation of the concept of the bioavailable fraction of iron is required.     

An in-situ analysis and measurement of thorium-234, uranium isotopes in seawater

A low-level β counting-α spectrum apparatus used for the detection of the β radiation of thorium-234 and α spectrum of thorium-228 simultaneously and an easy operating procedure for the enrichment, radiochemical separating, sampling and measurement of thorium-234 and uranium-238 in the ocean are developed based on the requirements of analyzing export productivity in the euphotic zone via thorium-234-uranium-238 radioactivity disequilibrium. The detecting efficiency both of β and α radiation is higher than 20%. The background of β radiation is less than 0.5 min-1, and the energy resolution of α detector is better than 1%. Total recycle ratio of thorium-234 is about 75%. Using the above apparatus and procedure, the radioactivities of dissolved thorium-234 and uranium-238, particle thorium-234 and uranium-238 in seawater of the South China Sea, the Southern Ocean and the Arctic Ocean were analyzed.     

An improved specific interaction model for seawater at 25°C and 1 atmosphere total pressure

Equations recently developed by Pitzer are applied to a seawater model on the assumption that only interactions between ions of opposite charge sign are significant. This model is shown to give good agreement with available experimental data for the osmotic coefficient and for the mean-ion activity coefficients of the major electrolyte components. The results are compared with the predictions of the ion-association model and with a number of models using the concept of specific ionic interaction. The relative merits of the various models are discussed. Pitzer's original equations are modified to enable the calculation of conventional single-ion activity coefficients. These values are compared with those obtained on the basis of alternative conventions and an encouraging measure of agreement is observed. It is concluded that the new model introduces a fluency into the specific interaction approach that should greatly facilitate the incorporation of trace components into future models.     

Thallium concentrations in seawater

Measurements of the concentration of thallium in seawater collected from numerous ocean locations ranged from 12 to 16 ng kg⁻¹. Variations between the Atlantic and Pacific Oceans, between the northern and southern hemispheres of the Pacific Ocean, and between surface and deep waters of both the Pacific and Atlantic oceans were comparable with the precision of the analyses. This relatively constant distribution indicates that the element's cycle in seawater may be similar to those of the alkali metals which are its principal biogeochemical analogues.