Several excellent and thoughtful reviews on the chemical composition of seawater were published in the early 1980s. In particular, Bruland [1983] discussed speciation of various elements, Broecker and Peng [1982] focused on the dynamic aspects, and Quinby-Hunt and Turekian [1983] presented a way to predict oceanic concentrations. Since then, however, new data continuously became available based on the advanced methods of sampling and analysis, making it clear that the concentration levels and oceanic distributions of many trace metals had to be revised. Needless to say, it is difficult to determine extremely low levels of trace chemical constituents, and none of the laboratories can determine all of the elements in seawater. Thus, we are not yet capable of obtaining every data set from the same location, even through intensive collaboration.
Figure 1 shows the profile data obtained by many workers from various locations of the North Pacific that are summarized according to the form of periodic table.
The water column concentrations of trace metals, such as Al, Ti, Ga, Zr, Ru, Ag, In, Te, Ir, Pt, Au, Hg, and Bi, are more than one order of magnitude lower than the values stated in the early 1980s. Although there are fewer elements today whose concentrations are not yet known, we still lack knowledge about Nb, Ta, and Ru. The data for Hf, Os, and Sn are few and probably unreliable. Furthermore, there are many trace metals whose distributions were obtained only through study at one or a few locations, including Sc, Ti, Zr, In, platinum group elements, Au, Hg, and Th. Obviously, these data must be confirmed by others. For elements other than those described above, the concentrations and distributions in seawater are well established.
The distribution patterns are classified into the following four categories: conservative type, nutrient type, scavenged type, and redox-controlled type. The conservative elements, such as halogens, alkali, and alkaline Earth elements, are present in seawater at relatively high concentration levels in constant proportion to salinity. They are homogenized in the ocean by water circulation on the time scale of 103 years within their relatively long mean oceanic residence times (>>105 years). On the other hand, the nutrient elements, such as phosphorus, nitrogen, and silicon, are depleted in surface water due to biological uptake and are enriched in deep water by regeneration from particulate matter. As a consequence of ocean circulation and the biogeochemical cycle, North Pacific and Indian deep waters have higher concentrations of nutrients than North Atlantic water. We now know that many trace metals, such as Ni, Zn, Cd, Ba, and Ge, follow this type of distribution.
In contrast, some heavy metals like Al, Mn, Co, Ce, Pb, Bi, and Th are highly reactive in seawater and are scavenged by particulate matter. They have short mean oceanic residence times (e.g., <<102–103 years) and occur in seawater at extremely low concentrations. Owing to the influence of terrestrial input to the surface ocean largely through the atmosphere, surface waters generally have higher concentrations of those metals than deep waters. Also, different strengths of this terrestrial source to the ocean surface tend to result in higher concentrations in the Atlantic than in the Pacific.
Some elements like Cr, As, Se, I, Te, and Pu exist in seawater at more than one oxidation state. Their oceanic behavior and distribution patterns are variable. The thermodynamically unstable reduced species of those elements are probably formed through biological mediation, and once in seawater of normal open ocean, they tend to be oxidized to the higher valency state. Anoxic basins, such as the Black Sea, Cariaco Trench, and some fjords, are exceptions where reduced species are stable in the waters.
Except for the conservative elements whose distributions are almost vertically constant, some elements fall into more than one category, for example, nutrient + scavenged type for Be, Cu, Ga, Zr, rare Earth elements etc., nutrient + redox-controlled type like Se, and redox-controlled + scavenged type like Te [Li , 1991] . Also, the elements Pb, Pu, Am, and possibly some others have been delivered to the ocean by human activities and therefore their vertical profiles are changing with time. The factors affecting those distributions are discussed in more detail in the original papers and previous reviews.—Yoshiyuki Nozaki, Ocean Research Institute, University of Tokyo, Japan
An electronic supplement to this article that includes this article, the figure, and a table, listing average element concentrations, with an accompanying reference list, may be obtained on the World Wide Web at http://www.agu.org/eos_elec as 97025e.html.
Broecker, W. S., and T. H. Peng, Tracers in the Sea , Eldigio Press, 690 pp., Palisades, New York, 1982.
Bruland, K., Trace elements in seawater, in Chemical Oceanography, 2nd Edition, Vol. 8 , edited by J. P. Riley and R. Chester, pp. 147-220, Academic Press, London, 1983.
Quinby-Hunt, M. S., and K. K. Turekian, Distribution of elements in sea water, Eos, Trans. AGU, 64 , 130-131, 1983.
Li, Y. H., Distribution patterns of the elements in the ocean: A synthesis, Geochim. Cosmochim. Acta, 55 , 3223-3240, 1991.
Back to Eos Electronic Supplement
