Radioactivity in the Marine Environment (1971)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Marine Sediments and Radioactivity 153 changes are reduction of Fe+3 to Fe+2, which is more soluble even in the presence of H2S, and the possible pre- cipitation of various metals, such as Zn, as sulfides. (See Chapter 5, for further discussion of this point.) Marine sediments in many areas have an oxygenated sur- face layer a few millimeters to a few centimeters thick, which probably is formed by O2 diffusion, burrowing of marine organisms, and mixing due to currents and waves. Below this surface zone, dissolved oxygen in interstitial wa- ter is depleted. Sediment in the lower zone is thought to represent the sediment that is below the level of reworking by currents and by physical and biological phenomena, while the upper zone is essentially in transit (Hayes, 1964; Rhoads, 1967). Under reducing conditions, the distribution coefficients are lower (Table 2) for 106Ru, 59Fe, 147Pm, 54Mn, 95Zr/ Nb, and 144Ce. Consequently, these radionuclides dissolve in the reduced zone and then move into the overlying oxy- genated zone; other radionuclides (90Sr, 65Zn, and 60Co) are less soluble under reducing conditions, and a reverse transport may be expected. Results similar to these for stable Fe, Mn, Co, and Zn have been reported by Sevastyanov and Volkov (1967), Brooks et al. (1968), and Presley and Kaplan (1968), based on the distribution of trace elements in interstitial waters from marine sediments. In many reactions, the different chemical behaviors under reducing conditions result from formation of different complexes rather than from direct reduction of the element. VERTICAL MOVEMENT OF RADIONUCLIDES IN MARINE SEDIMENTS Problems Sediment deposits contact the overlying seawater through the sediment-water interface, a two-dimensional surface that does not facilitate efficient uptake by sediments of radionuclides from seawater. Consequently, radionuclides are added to sediment by the deposition of particles with associated radionuclides or, in some circumstances, by movement of radionuclides through the interface into (or out of) the sediments. Thus, it is necessary to consider movements of radionuclides within the sediments. Large distribution coefficients for common radionuclides associated with sediments and sedimentary particles indicate that migration of radionuclides in these deposits cannot be very rapid. Rates of nuclide movement are at least partially determined by whether nuclides sorbed by sedimentary par- ticles can be released to the interstitial seawater or whether the nuclide remains associated with the particles. Diffusion of radionuclides into particles themselves further decreases the probability of subsequent radionuclide migration. How- ever, although the rate of radionuclide loss from particles is slower than the rate of uptake (Figure 5; Pomeroy et al., 1966), radionuclides associated with particles on or very near the water-sediment interface may redissolve in the overlying water as well as diffuse deeper in the sediment. Diffusion Two types of movement of radionuclides on the ocean bottom may be distinguished. First, and perhaps simplest, is diffusion of radionuclides into (or out of) the sediment owing to concentration gradients in the interstitial waters or overlying seawater, in combination with sorption and desorption on the sediment particles. Second, and more complicated, are the diffusive processes where the medium itself may be moving. Laboratory and field experiments were made of the diffusion processes, using 36C1 ions that were not sorbed by the sediment particles (Duursma and Bosch, 1970). These results, combined with observed distri- bution coefficients of sorption, were used to calculate dif- fusion coefficients for the same sediments and radionuclides discussed earlier. The results obtained for 36C1 agree with those (about 10~6 cm2/sec) obtained by Shiskina (1966) using a different approach and in situ measurements. The 90Sr values are in satisfactory agreement with those (about 10-8 cm2/sec) obtained for 226Ra (Koczy and Bourret, 1958). Chemical composition of interstitial waters (Home, 1969, p. 395-400) also plays an important role in controlling either radionuclide bonding to sediment particles or forma- tion of soluble ligands. The results of these experiments and calculations indicate that diffusion in sediments is extremely slow for most radio- nuclides. This can be seen from Figure 7, which shows the time required for the movement of a 10 percent concentra- tion front from a constant source. Note also that the se- quence of diffusion coefficients is the reverse of the se- quence of the distribution coefficients (page 148). In other words, the stronger the binding of a particular radionuclide to a sediment, the smaller the diffusion coefficient and the slower the diffusion process. Current and Wave Action Currents and wave action cause substantial mixing of sedi- ment, both vertically within a given sediment mass and horizontally, so that sediment-associated radionuclides may go into suspension from an initial depositional site and be moved substantial distances before final deposition. The upper few centimeters of cores often contain radionuclide

154 Radioactivity in the Marine Environment 10'd 10Jr (I,YEAR) (10 VEARsKv;fl CENTURY), (i'6 CENTUR1ES1 - 10' • 10' 10' TIME (DAYS) FIGURE 7 Depth penetration of a 10 per- cent concentration front into a sediment from an overlying water mass with a con- stant radionuclide concentration, as a func- tion of time, for various diffusion coeffi- cients. The range of diffusion coefficients found for various elements and marine sedi- ments is shown at the right. concentrations that are aberrant with respect to those at greater depths in the cores (Barnes and Gross, 1966; Jefferies, 1968). This may be caused by physical processes, by biological activity, or possibly by disturbance of the water-sediment interface during the sampling operation, if special care has not been taken to avoid such disturbance. Biological Activity Activity of burrowing organisms probably causes much ver- tical movement of radionuclides in shallow-water sediments. Burrowing by benthic organisms in subtidal muds causes the reworking of the upper 2 or 3 cm several times annually; the upper few millimeters is probably cycled daily. In sub- tidal sands, burrowing occurs to a depth of 30 cm (Rhoads, 1967). Barnes and Gross (1966) suggested that biological activity could account for the apparent rapid mixing of radionuclides in the upper 2 cm of several cores taken near the Columbia River mouth. Mixing due to biological activity can be treated as a dif- fusive process, and a biological diffusion coefficient can be calculated. Using data from Haven and Morales-Alamo (1966), who studied the depth distribution in sediment of fluorescent particles, a diffusion coefficient of about 10~6 cm2/sec can be calculated, a result of diffusion plus motion of the medium. In the Irish Sea near the Windscale discharge outlet (Templeton and Preston, 1966), mixing by benthic orga- nisms may be the cause of 106Ru penetration in the sedi- ments to depths of 25 cm below the interface (Figure 8). Assuming that the concentration in the sediment is the inte- gral of continuous diffusion processes from instantaneous sources (deposits on top of the sediment) and taking into regard the decay of 106Ru, the depth of penetration indi- cates an apparent diffusion coefficient of about 10~7 cm2/ sec, which is much closer to an effect predicted from bio- logical activity than to diffusion, where the coefficient for 106Ru should be about 10-11 to 10~12 cm2/sec (Duursma and Bosch, 1969). Interstitial Water Movement Upward movement of water through the water-sediment interface is still little known. However, sediment accumula- tion does cause compaction with a resulting upward move- ment of interstitial water. In the case of diffusion, this effect will be less than that of burial by sediment accumulation (see below). Probably a more important mechanism is the movement of water through the sediment owing to ground- water discharge from submarine aquifers on the continental margins. Evidence for submarine discharge from both fresh and brackish waters has been found as much as 120 km off the Florida coast, as well as elsewhere on the continental shelf and slope (Manheim, 1967). Such submarine discharges are especially likely off land areas with high rainfall that are underlain by suitably permeable aquifers (Kohout, 1967). Considering the slow rates of diffusion, it seems probable that such upward movement of water through the sediment-

Marine Sediments and Radioactivity 155 1000- I00- A ASL0PE • C0RE I •C0RE 2 'C0RE 3 10 I5 (CM) 20 25 0 DEPTH 200 400 tcn*] 600 FIGURE 8 Concentration of 106Ru in sediment cores from the Irish Sea (Templeton and Preston, 1966) (A) as a function of depth below the sediment-water interface and (B) replotted as a function of the square of the distance. The slope of the line in B indicates an apparent diffusion coefficient of 10"' cm2/sec. (Calculated accord- ing to Duursma and Hoede, 1967.) water interface could be an important process moving radio- nuclides from the sediment back into the overlying water. Another possibility in areas with high organic debris in their sediments is gas production (Reeburg, unpublished Ph.D. thesis, The Johns Hopkins University), which may cause partial mixing of interstitial water, resulting in redis- tribution of radionuclides. Burial by Sediment Accumulation If sediment accumulation is fairly rapid, a single layer of contaminated sediment will be mixed with or buried by later sediment deposits, thus inhibiting exchange of radio- nuclides across the sediment-water interface. Under such circumstances and where current, waves, and biological ac- tivities are almost absent, sediments can assimilate larger amounts of radionuclides than in areas where the rate of sediment accumulation is much lower. Figure 9 indicates the relative importance of diffusion of a 10 percent concen- tration front from a thin contaminated sediment layer and the competing process of burial by later sediment deposits. Since diffusion does not proceed uniformly, the answer to whether the sedimentation or the diffusion is faster depends on the time scale of the process. The diagram (Figure 9) gives a rough indication for periods of 1 to 107 years. For most of the deep-ocean floor where sediment accu- mulates at rates of about 1 mm per thousand years (Sverdrup et al., 1942) and for the 70 percent of the world's continental shelf not covered by sediment deposited in the past 3,000-5,000 years (Emery, 1968), diffusion is probably the dominant process. In these areas, loss and exchange of radionuclides from the sediments to the over- lying seawater will be primarily dependent on the reactions of the radionuclides. On the shallow continental shelf near major rivers, where sediment is accumulating relatively rapidly, the radioactive sediment may be deposited rapidly enough that it is not substantially disturbed by physical processes (currents and wave action) and before it can be extensively burrowed by benthic organisms. Under such conditions, thin contami- nated sediment layers may be covered. In parts of the northern Adriatic and Ligurian seas, such sediment layers containing fallout nuclides (90Sr, 144Ce, 147Pm, 155Eu) deposited during the period 1958-1965 were detected (Schreiber, 1966; Schreiber etal., 1968; Cerraie?a/.,1967; Albini et al., 1968), where depth below the interface is pro- portional to the time since deposition (Figure 10). Consid- ering that the diffusion coefficient of 90Sr (see Figure 7) is about 10~8 to 10~7 cm2/sec and that the diffusion coeffi- cient for the three rare-earth metals is 10~u to 10~10 cm2/sec, and further that the deposition rate is 2 cm/year in the North Adriatic, it can be concluded from Figure 9 that for 90Sr the diffusion should have been "faster" dur- ing the eight years of deposition of radioactive sediment but that the rare-earth radionuclides 144Ce, 147Pm, and 155 Eu have been covered up. Where low dissolved-oxygen concentrations occur sea- sonally in near-bottom waters, layers formed by individual phytoplankton blooms and varves formed by seasonal changes in sediment sources can be distinguished (see Gucluer and Gross, 1964, for an example of sediment ac- cumulation under such conditions). Consequently, one might expect to find sediment layers containing distinctive radionuclides from a single event, such as a period of unusu- ally large amounts of atmospheric fallout. In oxygenated waters such distinctive layers would be quickly mixed with other nonradioactive sediment by burrowing organisms. RADIONUCLIDE TRANSPORT BY SEDIMENTS Our consideration of radionuclide transport by sediments, though restricted to the continental shelf and near-shore waters, may have applications to deeper waters. Sediment- associated radionuclides are more likely to prove trouble- some in near-shore waters, either through direct contact with humans or through uptake by food organisms, especi- ally filter-feeding organisms. Among potential depositional sites are beaches, estuaries and their tidal flat areas, and open continental shelves.

156 Radioactivity in the Marine Environment GR0SS B RAD1OACT1VITY FALL0UT RAD10ACTIVITY (C0RE I0A) 10 0.I I.0 10 I0Z I03 DISTANCE 10% DIFFUSI0N FR0NT AND SEDIMENTATI0N RATE (cm/I000YEARS) FIGURE 9 Relation between the distance at which 10 percent of the concentration of a thin contami- nated sediment layer may be found and the diffusion coefficient, for different times after the instantaneous contamination of this layer. Examples follow: 1. If the sediment is covered by a layer of newly deposited sediment in a certain time (say 10 years at sedimentation rate 1 cm/10 years), then the diffusion is regarded as "faster" (based on 10 percent front dis- tance) than the sedimentation for radionuclides having a diffusion coefficient higher than 3.5 X10-10 cm2/sec. 2. For the same sediment accumulation rates, the diffusion coefficients for 1 and 100 years are 3.5 X 10-H cm2/secand 3.5 X10•9 cm2/sec, respectively. Note the agreement for the Po river mouth (see Figure 10), where the sedimentation rate is 2 cm/year; for periods of 1 to 10 years sedimentation dominated be- cause the diffusion coefficients of the radionuclides involved were smaller than 3 X10~" cm2/sec (except for 90Sr). I965 1964 1963 1962 cr I96I 1960 I959 I958 100 200 300 400 (mC1/Km', M0NTHLY) 10 II 12 13 14 I5 (cpm/g DRY) FIGURE 10 Variation of gross beta activity from atmospheric fallout observed at Ispra, Euratom Centre, Italy, and gross beta activity in sediment from a core collected in the Adriatic Sea near the Po river mouth in 1966(Schreiberef J/., 1968). The apparent rate of sediment accumulation is approximately 2 cm/year. a major discharge of radioactivity (Templeton and Preston, 1966). The low level of activity of the radionuclides is prob- ably the result of the lack of fine-grained particles (see Table 1) and the continual abrasion of the grain surfaces. Sediment may be moved fairly rapidly in the area between the breakers and the beach when incoming waves strike the beach at an angle (Galvin, 1967). Longshore currents trans- port substantial quantities of sediment, and near discharges of radioactive effluent, this may be an important route of nuclide movement (Templeton and Preston, 1966). Estuaries Beaches Sediment deposited on beaches is more or less continually stirred by wave action. As a result, the sediments typically are well-sorted sands with grain sizes of 100 to 200 ju (Postma, 1967a and b; Jefferies, 1968). Although there is often a pronounced landward movement of water along the bottom (Morse et al., 1968), there is commonly relatively little artificial radioactivity associated with sands, even near Estuaries act as sediment traps (Postma, 1967a and b). A rel- atively small fraction of riverborne sediment or sediment moving along the coast escapes in a short time over most of the world's continental shelves (Emery, 1968). Instead, large quantities of fine-grained sediment (< 100 n) accumu- late, at least temporarily, in the estuaries or on tidal-flat areas (Postma, 1967a and b), where the fine particles are passed in large quantities through filter-feeders (Verwey, 1952), some of them important as food resources. In an estuarine system, sediments are susceptible to resuspension

Marine Sediments and Radioactivity 157 and movement by tidal or river currents (Conomos, 1968). During major floods or periods of strong wave action, sedi- ments are washed out from the estuaries to be deposited on the continental shelf. Where there are river deltas, the fine- grained sediment (<50^) can remain suspended near the point where the near-bottom flow changes from a net-down- stream direction in the river to a net-upstream flow in the estuary (Postma, 1967a and b). Klingeman and Kaufman (1965) demonstrated deposition of riverborne sediment-associated radionuclides in San Fran- cisco Bay. Radionuclides bound to river sediments may be desorbed when they enter the sea. De Groot (1966) and Eisma (1968) showed that Fe in suspended matter was de- sorbed at low salinities and later was precipitated in seawater. Continental Shelves Information about sediment movement on continental shelves comes from interpretation of the distribution of various sedimentary characteristics. Studies of radioactive sediment dispersal on the continental shelf provide some of the best information about actual sediment movement in this area. Large amounts of radioactive materials discharged from the Windscale plant on the Irish Sea (Templeton and Preston, 1966) and the Columbia River discharge from the Hanford plutonium-producing reactors (Gross, 1966) pro- vide enough tracer materials to permit sediment movements to be followed over substantial distances. In both areas, the radioactive sediment moved generally parallel to the coast- line. Near-bottom currents on the continental shelf are little known, but they probably play a major role in transporting sediment-associated radionuclides. From the few studies of near-bottom currents available, it appears that near- bottom currents tend to parallel the adjacent coast except near major estuarine systems, where near-bottom currents flow toward the estuary (Bumpus, 1965; Templeton and Preston, 1966;Rehrerera/., 1967;Morse era/., 1968). On the Atlantic coast of the United States, near-bottom cur- rents are rather similar to the surface currents, setting essen- tially southward at about 0.5 km/day (Bumpus, 1965). On the Pacific coast of the northwestern United States, the near-bottom currents generally set toward the north, al- though the variable surface currents set southward for about six months a year (Morse et al., 1968). Reported speeds of seabed drifters, devices used to study near-bottom currents, are about 1-2 km/day. Although such studies indicate probable directions of sediment movements, they do not indicate the speed. Sediment-associated radionu- clides, however, may be used to determine apparent speeds of movement of radioactive particles (Gross and Nelson, 1967) under certain conditions. Submarine Canyons Submarine canyons appear to be the major routes of sedi- ment movement across the continental shelf (Shepard, 1965). Where rivers bring large amounts of sediment to the ocean, frequent turbidity currents seem to empty the nearby canyons and carry large amounts of sediment onto the deep-ocean floor. Where continental shelves are wide, sub- marine canyons appear to be inactive at present. Where shelves are narrow and the canyons head close to shore, they appear to have sediment continually moving through them by creep, sand flows, and slumps (Shepard, 1965). Except for a report of 65Zn and 51Cr in sediment in a submarine canyon near the Columbia River (Osterberg et al, 1963), there have been no reports of deep-ocean accumula- tion of sediment-associated nuclides from the various coastal discharges of radioactivity. Arctic Areas Movement of sediment-associated radionuclides in high lati- tudes, where transport by ice may be involved, has not been included in our discussion largely because of the lack of available information about the behavior of sediments under those conditions. However, it is known that glacially derived clays in seawater strongly sorb many common radionuclides (Duursma and Eisma, unpublished). SCAVENGING ABILITY OF SEDIMENTS In areas where sediments are resuspended by tidal currents, wave activity, and river currents, there will be high concen- trations of particles in the water. The sorption capacity of these particles for dissolved radionuclides may cause scaveng- ing and deposition of the radionuclides on the bottom. Re- moval of radionuclides from the water will depend on the distribution coefficients of the radionuclides and particles involved, the rates of sorption, and the settling velocities of the sediment particles. Because of their settling velocities, sediment particles will have a relatively short time of contact with the water for sorption of soluble nuclides to occur. Thus, complete sorption equilibrium between particles and water may not be achieved. This, however, is probably not too significant, because the first step in the sorption process is relatively rapid, and for most of the radionuclides at least 80 percent of the equilibrium is achieved in several hours. Using data on distribution coefficients presented in this chapter, it is possible to calculate the effect of scavenging for a given concentration of sediment settling through a

Radioactivity in the Marine Environment 05 o 55000 £1000 _225- z LU 100: 5 50 UJ b h- Od O 01 S E EFFECT OF REPEATED SCAVENGING FOR K=I04 •= Ix x=2x 0= 3x = 10' QC LL) >- ='04| tx. o 103 E cr LU Q. CO z O 10 20 30 40 50 60 70 80 90 100 PERCENTAGE LEFT IN SOLUTION FIGURE 1 2 The effect of five successive scavengings by equal amounts of sediment with a density (dry) of 2.27 g/ml for a radio- nuclide with a distribution coefficient of 10 . 0 10 20 30 40 50 60 70 80 90 100 PERCENTAGE LEFT IN S0LUT10N FIGURE 11 Predicted relationship between the fraction of the original radionuclide concentration remaining in solution as a func- tion of the amount of sediment that has settled out from suspension, for different distribution coefficients. The assumed dry sediment density is 2.27 g/ml. [%RAOI0NUCLIDE [LEFT IN S0LUT10N 10 I02 K3! 10* 10* I06 AM0UNT 0F SEDIMENT F0R IX SCAVENGING [T0NS/km*(20m LAYER)] FIGURE 13 The relation between the amount of sediment necessary to obtain the same effect of scav- enging, if divided into four parts or if used at once, for different distribution coefficients (density of dry sediment 2.27 g/ml). Abscissa indicates the amount of sediment if used only once; ordinate, the total amount of four successive treatments. Note that scavenging dividing up into subsequently smaller amounts has less effect when the quantities are too small (where the full lines approach the 45° line).

Marine Sediments and Radioactivity 159 contaminated layer of seawater. Figure 11 is a model in which the amount of radionuclide remaining in seawater after settling of the particles is plotted as a function of the concentration of suspended sediment for different distribu- tion coefficients. For example, 225 mg (dry weight) of sedi- ment per liter of seawater with a distribution coefficient of 104 for the radionuclide involved will reduce the concentra- tion of the radionuclide by 50 percent. Repeated suspension and settling of smaller amounts of sediment will remove more of such radionuclides than a single release of an equiv- alent total amount of sediment (shown by the dotted lines in Figure 12; see also Figure 13). It must be remembered that the model is based on pre- liminary results using a relatively small number of labora- tory experiments; the postulated removals have not been tested experimentally under natural conditions. Scavenging is known to occur, however, under natural conditions-rough weather in the Irish Sea near the Windscale pipeline outlet causes more rapid removal of radioactivity from the water and deposition in the bottom sediments (Templeton, per- sonal communication). This model is presented as an example of possible uses of sediments or other solids to remove waterborne pollutants such as radionuclides, causing them to be deposited on the bottom. Whether it is desirable to do this rather than de- pending on natural dilution and burial processes depends on local conditions and other factors. The results of these cal- culations suggest that such applications of sediments would be restricted to dealing with accidents or other releases in coastal areas such as bays, lagoons, harbors, rivers, or lakes (Duursma, 1969). SUMMARY Sediments and sedimentary particles have a substantial ca- pacity to remove radionuclides from seawater. The particles may be deposited on the ocean bottom or picked up by filter-feeding organisms that may concentrate the radio- nuclides and pass them into man's food supply. Certain radionuclides are more strongly sorbed on sedi- ment particles than others. Mineral-related reactions, each with a degree of specificity, appear to be controlled by chemical species of the radionuclides and the physico- chemical conditions of the sediments and sedimentary par- ticles. In some laboratory experiments, the amount of radionuclides taken up by sediments and sedimentary par- ticles depends on the history of the sediment-water system prior to the introduction of the radionuclide; this makes it difficult to predict radionuclide behavior in sediment-laden water. Radionuclides may move through the sediment after de- position. Among the processes by which radionuclides move are diffusion through the interstitial water, movement with the interstitial water expelled from the sediment by com- paction, and groundwater discharge through the sea bottom. Where sediment accumulation is sufficiently rapid, the ra- dioactive particles may be covered by later sedimentary deposits before these relatively slow movements can move the nuclides through the sediment-water interface. On a large part of the continental shelf and on much of the deep- ocean bottom, rates of sediment accumulation are slow enough that diffusion processes may be dominant. Sediment particles and their associated radionuclides may be mixed or moved by burrowing activity of benthic organisms, near-bottom currents, wave action, turbidity currents, longshore currents, or by ice. Sediments or sedimentary particles can remove radio- nuclides from seawater and deposit them on the sea bottom. Calculations have been given to indicate the effect of such applications of sediments, and criteria have been presented to guide the selection of materials to be used in such appli- cations. REFERENCES Albini, A., C. Giacoletto, M. G. Mezzadri, and C. Triulzi. 1968. Radiochemical studies for radioactivity determinations on marine samples: 90Sr, 144Ce, 147Pm, and 155Eu. CISE-108. CISE- Documentation Serv., Segrato (Milan). 77 p. Anikouchine, W. A. 1967. Dissolved chemical substances in compact- ing marine sediments. J. Geophys. Res. 72:505-509. Barnes, C. A., and M. G. Gross. 1966. The distribution at sea of Columbia River water and its load of radionuclides, p. 291-302. In disposal of radioactive wastes into seas, oceans, and surface waters. IAEA, Vienna. Brooks, R. R., B. J. Presley, and I. R. Kaplan. 1968. Trace elements in the interstitial waters of marine sediments. Geochim. Cosmo- chim. Acta32:397-414. Bumpus, D. F. 1965. Residual drift along the bottom on the Con- tinental Shelf in the Middle Atlantic Bight. Limnol. Oceanogr. 10 (Suppl.):R50-R53. Carrol, D. 1958. Role of clay minerals in the transport of iron. Geochim. Cosmochim. Acta 14:1-27. Cerrai, E., B. Schreiber, and C. Triulzi. 1967. Vertical distribution of 90Sr, 144Ce, 147Pm, and 155Eu in coastal marine sediments. Energ. Nucl. 14:586-592. Chave, K. E. 1965. Carbonates: Association with organic matter in surface sea water. Science 148:1723-1724. Conomos, T. J. 1968. Processes affecting suspended particulate mat- ter in Columbia River effluent system, summer, 1965, 1966. Ph.D. Thesis. Univ. Washington, Seattle. Duursma, E. K. 1969. Chemistry of sediments, p. 92-97. In Annual report of the International Laboratory of Marine Radioactivity, Monaco, IAEA Laboratory Activities, Sixth Report, Tech. Rep. Ser. No. 98. Duursma, E. K., and C. J. Bosch. 1970. Theoretical, experimental and field studies concerning diffusion of radioisotopes in sedi- ments and suspended solid particles of the sea. Part B: Methods and experiments. Neth. J. Sea Res. 4(4):395-469. Duursma, E. K., and D. Eisma. In preparation. Theoretical, experi- mental and field studies concerning diffusion of radioisotopes in sediments and suspended solid particles of the sea. Part C: Field studies. Neth. J. Sea Res. Duursma, E. K., and C. Hoede. 1967. Theoretical, experimental and

160 Radioactivity in the Marine Environment field studies concerning diffusion of radioisotopes in sediments and suspended solid particles of the sea. Part A: Theories and mathematical calculations. Neth. J. Sea Res. 3:423-457. Eism.a, D. 1968. Composition, origin and distribution of Dutch coastal sands between Hoek van Holland and the island Vlieland. Neth. J. Sea Res. 4:123-267. Emery, K. O. 1968. Relict sediments on continental shelves of world. Amer. Ass. Petrol. Geol., Bull. 52:445-464. Galvin, C. J., Jr. 1967. Longshore current velocity: A review of theory and data. Rev. Geophys. 5:287-304. Garder, K., and O. Skulberg. 1964. Sorption phenomena of radio- nuclides to clay particles in river water. Int. J. Air Water Pollut. 8:229-241. Groot, A. J. de. 1966. Mobility of trace elements in deltas. Trans. Comm. II and IV. Int. Soc. Soil Sci., Aberdeen, 267-279. Gross, M. G. 1966. Distribution of radioactive marine sediment derived from the Columbia River. J. Geophys. Res. 71:2017- 2021. Gross, M. G., and J. L. Nelson. 1966. Sediment movement on the continental shelf near Washington and Oregon. Science 154:879- 885. Gucluer, S. M., and M. G. Gross. 1964. Recent marine sediments in Saanich Inlet; A stagnant marine basin. Limnol. Oceanogr. 9:359-376. Hamaguchi, H. 1962. Study of radionuclides sorbed on marine sedi- ments. Progress Rep. Contract No. 88/Rl/Rb IV. IAEA, Vienna. 10 p. Haven, D. S., and R. Morales-Alamo. 1966. Use of fluorescent par- ticles to trace biodeposits in marine sediments. J. Cons. Cons. Perm. Explor. Mer 30:267-269. Hayes, F. R. 1964. The mud-water interface. Annu. Rev. Oceanogr. Mar. Biol. 2:121-145. Holeman, J. N. 1968. The sediment yield of major rivers of the world. Water Resourc. Res. 4:737-747. Home, R. A. 1969. Marine chemistry. Wiley (Interscience), New York. 568 p. Jefferies, D. F. 1968. Fission-product radionuclides in sediments from the North-East Irish Sea. Helgolander Wiss. Meeresunters. 17:280-290. Jenne, E. A., and J. S. Wahlberg. 1968. Role of certain stream- sediment components in radio-ion sorption. U. S. Geol. Surv. Prof. Pap. 433-F. 16 p. Jones, R. F. 1960. The accumulation of nitrosyl ruthenium by fine particles and marine organisms. Limnol. Oceanogr. 5:312-325. Kautsky, H. 1966. Possible accumulation of discrete radioactive elements in river mouths, p. 163-174. In Disposal of radioactive wastes into seas, oceans and surface waters. IAEA, Vienna. Klingeman, P. C., and W. J. Kaufman. 1965. Transport of radio- nuclides with suspended sediment in estuarine systems. Sanitary Engineering Res. Lab. Rep. No. 65-15. Univ. California, Berkeley, Calif. 222 p. Koczy, F. F., and R. Bourret. 1958. The distribution of radium in deep-sea sediments. Progr. Rep. Nat. Sci. Found. Grant No. G-3995.V. M. Acct. 8891. 18 p. Koczy, F. F., and J. N. Rosholt. 1962. Radioactivity in oceanog- raphy, p. 18-46. In H. Israel and A. Krebs [ed.] Nuclear radia- tion in geophysics. Springer-Verlag, Berlin. Kohout, F. A. 1967. Ground water flow and the geothermal regime of the Floridian Plateau. Trans. Gulf Coast Ass. Geol. Soc. 17: 339-354. Manheim, F. T. 1967. Evidence for submarine discharge of water on the Atlantic continental slope of the southern United States and suggestions for further search. N.Y. Acad. Sci., Ser. II, 29:839- 853. Morse, B. A., M. G. Gross, and C. A. Barnes. 1968. Movement of seabed drifters near the Columbia River. J. Waterways Harbor Div., Proc. Amer. Soc. Civil Eng. 94 (WW1):93-103. Noshkin, V. E., and V. T. Bowen. 1965. Vertical distribution of in- dividual radioisotopes from fallout in marine cores. NYO-2174- 34. In Symposium on American chemistry forefronts in terrestrial geochemistry, Atlantic City, New Jersey. (Abstr.) Osterberg, C. L., L. D. Kulm, and J. V. Byrne. 1963. Gamma emit- ters in marine sediment near the Columbia River. Science 139: 916-917. Pomeroy, L. R., E. P. Odum, R. E. Johannes, and B. Roffman. 1966. Flux of 32P and 65Zn through a salt-marsh ecosystem, p. 177- 188. In Disposal of radioactive wastes into sea, oceans and sur- face waters. IAEA, Vienna. Postma, H. 1967a. Marine pollution and sedimentology, p. 225- 233. In T. A. Olson and F. J. Burgess [ed.] Pollution and marine ecology. Wiley, New York. Postma, H. 1967b. Sediment transport and sedimentation in the estuarine environment. In G. H. Lauff [ed.] Estuaries. Amer. Ass. Advan. Sci. Publ. 83. Washington, D.C. Presley, B. J., and I. R. Kaplan. 1968. Changes in dissolved sulfate, calcium and carbonate from interstitial water of near-shore sedi- ments. Geochim. Cosmochim. Acta 32:1037-1048. Rehrer, R., A. C. Jones, and M. A. Roessler. 1967. Bottom water drift on the Tortugas grounds. Bull. Mar. Sci. Gulf Caribbean 17:562-575. Rhoads, D. C. 1967. Biogenic reworking of intertidal and subtidal sediments in Barnstable Harbor and Buzzard's Bay. J. Geol. 75:461-476. Richards, F. A. 1957. Oxygen in the ocean, p. 185-238. In J. W. Hedgpeth [ed.] Treatise on marine ecology and paleoecology. Vol. 1. Ecology. Geol. Soc. Amer., Mem. 67. Robinson, B. P. 1962. Ion-exchange minerals and disposal of radio- active wastes-a survey of literature. U. S. Geol. Surv. Water- Suppl. Pap. 1616. 132 p. Schreiber, B. 1966. Radionuclides in marine plankton and coastal sediments, p. 75 3-770. In Proceedings of the International Symposium: Radioecological concentration processes, Stockholm. Pergamon Press, New York. Schreiber, B., L. Tassi Pelati, M. G. Mezzadri, and G. Motta. 1968. Gross beta radioactivity in sediments of the North Adriatic Sea: A possibility of evaluating the sedimentation rate. Arch. Oceanogr. Limnol. 16:45-62. Sevastyanov, V. F., and 1.1. Volkov. 1967. Redistribution of chemi- cal elements at redox processes in the bottom sediments of the oxygen zone of the Black Sea [in Russian]. Trans. Inst. Oceanol. USSR, Acad. Sci. 83:115-134. Shepard, F. P. 1965. Importance of submarine valleys in funneling sediments to the deep sea, p. 321-332. In M. Sears [ed.] Progress in oceanography. Vol. 3. Pergamon Press, New York. Shiskina, O. V. 1966. On the determination of the intensity of ex- change of chemical elements between marine sediments and natural water, p. 26-34. In Chemical processes in sea and oceans. Acad. Sci. USSR, Comm. Oceanogr., Moscow. (In Russian.) Sverdrup, H. U., M. W. Johnson, and R. H. Fleming. 1942. The oceans. Their physics, chemistry and general biology. Prentice- Hall, Englewood Cliffs, New Jersey. 1087 p. Templeton, W. L., and A. Preston. 1966. Transport and distribution of radioactive effluents in coastal and estuarine waters of the United Kingdom, p. 267-289. In Disposal of radioactive wastes into seas, oceans and surface waters. IAEA, Vienna. Verwey, J. 1952. On the ecology of distribution of cockle and mussel in the Dutch Wadden Sea, their role in sedimentation and the source of their food supply. Arch. Neerl. Zool. 10:171-239.

Chapter Seven ACCUMULATION AND REDISTRIBUTION OF RADIONUCLIDES BY MARINE ORGANISMS F. G. Lowman, T. R. Rice, F. A. Richards The transport and distribution of radionuclides introduced into the sea are dependent upon the interactions of the water, organisms, and sediments with the added material and with each other. The organisms, detritus, and sediments accumulate several elements to levels many times greater than those in seawater and may influence the distribution of the corresponding radionuclides in areas of high biological activity or rapid sedimentation. In the open sea, the bottom sediments are far removed from most of the vertical water column, and the interaction of these sediments with surface-introduced material may be limited almost entirely to their constituting a floor for in- soluble particles carried down by gravity. In the surface wa- ters of these areas, the biota, too, may provide insignificant mass and exert only minor influences upon the transport of most trace elements. In regions of upwelling and in shallow near-shore areas, however, biomass is often relatively great. The influence of the organisms upon the distribution patterns of some trace elements or radionuclides added from land runoff or from direct fallout may be significant. In the shallow near-shore areas, sediments and suspended particulate matter may alter biological influences. The high degree of adsorption by these sediments may, however, be partly due to biological activity of the attached periphyton. In marine ecosystems, the plants and animals are in direct contact with trace elements in their environment, thus allowing direct transfer of elements as ions, colloids, or particulates at the phase boundaries between the hydro- sphere and biosphere. This close relationship may result in rapid and strong mutual interactions between the organisms and the environment when fluctuations of biomass or abun- dance of trace elements or radionuclides occur. Thus, when other factors are not limiting, the abundance of available nutrients controls the biomass. Large increases in biomass may, in turn, deplete some nutrient elements from localized areas. Not only may marine organisms alter the distribution of stable elements normally present in marine waters, but they may also influence the seasonal, horizontal, and verti- cal distributions of radionuclides and stable elements intro- duced into the sea by man. These interactions should be considered in studies of the effects of biological activity upon the distribution patterns of radionuclides and trace elements in the sea. The sources of radioactivity and the inorganic interac- tions of the radionuclides and associated stable elements with the sea have been reviewed in detail in Chapter 2 and will be discussed here only with respect to their influences upon the availability of radionuclides to marine organisms. Biological processes that result in the passage of radionu- clides through food webs to man will be considered, as well as the relative importance of physical and biological trans- port on the distribution patterns of radionuclides introduced into the marine environment. 161

162 Radioactivity in the Marine Environment INFLUENCE OF NONBIOLOGICAL FACTORS ON THE AVAILABILITY OF RADIONUCLIDES TO MARINE ORGANISMS Source of Radionuclides The availability of radionuclides to the marine biota de- pends largely upon their chemical and physical forms and the rates and modes of addition to the marine environment. Most of the nuclear detonations thus far have occurred under oxidizing conditions. Devices exploded on or near the surface of land incorporated large amounts of coral, water, device material, support structures, and air into the fireball, delivering large amounts of oxides. In marine areas receiving large amounts of fallout, precipitation of calcium and ferric hydroxides effectively scavenged some radio- nuclides to deeper water, out of reach of the plankton and pelagic animals but in direct contact with the benthic ani- mals. Tower or air bursts, in contrast to surface detonations, introduced only small amounts of contaminants into local- ized marine areas but contributed a majority of the radio- nuclides in oxidized form in worldwide fallout. The total weight of the stable material in this fallout was only 1,000 to 10,000 times the weight of the radionuclides, and this extremely small amount of material would not be significant in scavenging radionuclides from the surface waters. Thus, these radionuclides would be available to the organisms in the upper waters and would be subject only to limited transport to the bottom. Over long periods of time, the insoluble radionuclides associated with oxides of iron and manganese would sink, however, as a result of their being incorporated into fecal pellets by filter-feeding zooplankton. Manganese-54, whose stable counterpart occurs as the Mn+2 ion in seawater, has been reported by several investigators to exist in seawater mainly as particulate manganese dioxide (Slowey et al., 1965; Schelske et al., 1966). Filter-feeding sponges have been observed to accumulate 54Mn in preference to stable Mn in the sea, whereas benthic algae from the same area ac- cumulate their manganese mainly from solution (Lowman et al., 1967a). Ruthenium-103 and ruthenium-106 in world- wide fallout also is present in particulate forms at ratios 4 to 13 times those for stable ruthenium in seawater (Dixon era/., 1966). Nuclear explosions under the surface of the sea also occur under oxidizing conditions, partly because of dissolved oxy- gen in the water. Although the metal in the device and the support sources would provide less mass than the sodium chloride in the vaporized seawater, the centers of nucleation would be provided by the oxides of iron, calcium, and magnesium. Underwater detonations in shallow bodies of water may produce sufficient heat to convert interstitial water of the bottom sediments into steam, dislodging sediments that may be incorporated into the fireball. A detonation of this type on a calcium carbonate bottom of a lagoon resulted in little or no precipitation of 89Sr by the suspended calcium oxide and hydroxide. Although this radionuclide would be expected to be coprecipitated, it remained in the soluble fraction. During the time of calcium precipitation, the precursors of 89Sr (4.4 sec 89Br, 3.2 min 89Kr) constituted the major isotopes of fission chain 89, and only a small fraction of the 89Sr was available for coprecipitation. Thus, the chemical characteristics of the precursors of a long-lived fission product at the time of detonation influence the availability of the radionuclides to the biota. Underground nuclear explosions, such as those tested for excavating, differ in several ways from detonations in the air, on the surface of the ground, or in the water. Under- ground detonations (a) introduce large amounts of stable elements from the vaporized and pulverized rocks into the fallout, which could cause significant precipitation and coprecipitation of radionuclides as well as isotope dilution in seawater; (b) reduce the amounts of radionuclides reach- ing the sea, since a major fraction of the total radioactivity would be adsorbed onto the surfaces of the material that falls back into or near the lip of the excavation (the fallout and cloud would contain only 5 to 10 percent of the radio- activity); (c) produce a reducing environment as a result of the presence of vaporized metal from the device, support structure, and the casing of the drill hole, electrons from fission, hydrogen from fusion, and elementary carbon, hy- drogen, and native metals in the vaporized geological material. Several elements, 54Mn and 56Mn, for example, would descend in fallout from underground detonations in a partly or completely reduced state and would exhibit increased solubility when added to seawater, compared to the oxi- dized forms. The behavior of radioisotopes of tungsten also appears to be influenced by this type of detonation. Radio- tungsten produced under oxidizing conditions was com- plexed tightly to the waxy leaves of land plants; however, no foliar or root uptake into the plant could be detected (Lowman, 1960). In an underground detonation at the Nevada test site, Romney et al. (1967) observed a high degree of uptake of radiotungsten produced under reducing conditions through the roots but reported no chelation onto the leaves. The distribution in the sea of radiotungsten produced under oxidizing conditions suggested that a physi- cal or chemical change occurred after its introduction into the water. The radiotungsten constituted about 50 percent of the total radioactivity in the water and plankton of an open-sea area about 180 miles in diameter. About half of the tungsten in the plankton (25 percent of the total radio- activity in the area) was associated with silica, probably the skeletons of marine diatoms. During a second survey of the same body of water three weeks later, all of the radiotung- sten had dissociated from the plankton but was still present in the water (Lowman et al., 1959).

Accumulation and Redistribution of Radionuclides by Marine Organisms 163 Environmental Factors Environmental factors may vary geographically and season- ally. Biological productivity is usually higher near the land than in the open sea, and a greater fraction of the total radionuclides might be expected to be associated with the biota in the near-shore regions. However, other mechanisms may operate near the shore to reduce the availability of in- dividual radionuclides to the plants and animals. These in- clude fractionation of radionuclides during runoff from the watersheds, isotope dilution, chemical precipitation and coprecipitation, adsorption by suspended and bottom sedi- ments, and vertical and horizontal diffusion. Estuarine and other near-shore areas receive variable amounts of river runoff from the adjacent land mass. Frac- tionation of radionuclides of different elements in water draining from watersheds occurs because of differences in solubility, particle size, and adsorptive characteristics of the individual radionuclides and stable elements. Radionuclides that are easily dissolved or are not strongly adsorbed to the larger solid particles occur in the runoff in large amounts compared with the insoluble or strongly adsorbed elements. Radionuclides present in enhanced amounts in the runoff will, however, usually be accompanied by the corresponding stable element. In these cases, isotope dilution may reduce the specific activity* of the radionuclides in the marine or- ganisms near the outflows of rivers to levels below those occurring in the offshore animals and plants contaminated by direct fallout. Although the specific activity may be lower in the near-shore organisms, they may contain greater total amounts of element and thus more radionuclide (Duke, 1967). In these cases, the total ash content of plants and animals living in areas of heavy runoff from the land may be two to three times greater than that in the same species from other areas (Lowman et al., 1967a). Radionuclides added to the sea from river outflows are diluted by the corresponding stable elements normally present in seawater. In areas of upwelling, relatively large amounts of trace elements are supplied to the surface from the deeper layers. Phytoplankton, zooplankton, and omniv- orous fish collected in these areas often contain smaller amounts of fallout radionuclides than organisms of the same groups collected in other marine areas where upwelling does not occur (Avila, 1969). In areas of mixing of fresh water and seawater, radio- nuclides in ionic form, adsorbed onto suspended river sedi- ments, may be released into the estuaries as a result of the presence of electrolytes in seawater (Schutz and Turekian, 1965 a). Colloids of iron and aluminum may simultaneously 'Specific activity is the ratio between the amount of radioactive isotope present and the total amount of all other isotopes of that same element, both radioactive and stable. Most commonly, it is given in microcuries of radioisotope per gram of total element. be precipitated and act as scavengers for radionuclides, in- cluding 55Fe, 59Fe, 65Zn, 54Mn, 95Zr, 144Ce, 147Pm, 106Ru, and 110Ag. In areas of extensive scouring and resus- pension of bottom sediments by tidal currents, several radio- nuclides introduced by river waters may be adsorbed onto sediment surfaces. Some of the adsorption may be due to inorganic processes, but significant amounts of added radio- nuclides and trace elements may be bound to sediment sur- faces by the periphyton (Pomeroy et al., 1965). Radionuclides introduced into the surface waters of the open sea become distributed in the upper mixed layer. Be- tween 50°N and 45°S, the oceans usually consist of an upper "warm-water sphere" separated by a vertical transi- tion zone of rapid temperature and density change from a deeper "cold-water sphere" that reaches to the bottom of the sea (Dietrich, 1963). The warm upper waters vary in depth but are often about 100 m thick and are rapidly mixed by winds. In this layer the temperature, salinity, and density are nearly uniform. A shear zone often exists just below the mixed water in the layer of rapid change in den- sity. This layer, the pycnocline, constitutes a barrier for the downward movements of soluble material from the upper mixed layer or for the upward movement of dissolved ele- ments in the deeper waters (Revelle, 1956; Lowman, 1960). Radionuclides introduced into the sea exist as solutes, colloids, or particulates, depending upon their chemical characteristics. Soluble and colloidal forms tend to follow the water masses into which they are placed unless they be- come associated with particles through physicochemical or biological activity. On the average, the bottom is at a depth of about 3,800 m, and the accumulation of radionuclides by the benthic biota from surface introduction occurs mainly through activity of the plankton and the sinking of organic debris and inorganic particulates from the surface waters. BIOLOGICAL FACTORS THAT INFLUENCE THE AVAILABILITY OF RADIONUCLIDES TO MARINE ORGANISMS Concentration Factors Biological concentration factors for nonconservative ele- ments are generally greater than those for the major, or conservative, elements, which are present in amounts di- rectly related to the salinity or chlorinity of the water and are effectively transported and distributed in the sea by currents. Nonconservative elements do not follow these patterns, however, and marine biologists have often postu- lated that marine plants and animals participate in the trans- port of these elements. This concept is based mainly upon the observations that marine organisms concentrate several

164 Radioactivity in the Marine Environment elements to levels many times those in seawater and that the organisms often are capable of moving relatively quickly in directions different from those of the water masses in which they live. This was noted originally for phosphorus and silicon. Several elements occur in marine organisms in amounts exceeding those in the water. The term "concentration fac- tor" has been used in this relationship and may be defined as the ratio of the concentration of an element or radio- nuclide in an organism or its tissues to that concentration directly available from the organism's environment under equilibrium or steady-state conditions. This definition of a concentration factor is complicated by the fact that aquatic organisms do not normally derive all of their nutrients or radionuclides from one source but instead accumulate them from a variety of sources, including food, water, and sus- pended or deposited sediment. Concentration factors are indicators only and are not absolute. They may be altered by biological and environmental factors. Polikarpov (1966), in his often-used definition of "con- centration factor," stated that this factor is "the ratio of the concentration of a radionuclide (or corresponding stable element) in the organism and in the aqueous solution." The restrictive character of this definition is illustrated by his added statement that "the capacity of an organism to accu- mulate radioactive substances is expressed by the ratio of its radioactivity to that of the aqueous medium or the preceding food link* in which the radionuclide was con- centrated." In many marine organisms, the concentration factor is based on the preceding food link. The limited prac- tical utility of a concentration factor concept based entirely on water becomes even more apparent when one considers some infaunal marine organisms that do not live in water and do not eat "normal" food but rather live in bottom sediments and feed on detritus. Because many of these organisms ingest large amounts of sediment along with the organic detritus, one is faced with the problem of selecting the environmental basis for comparing the amounts of ele- ments or radionuclides in the organisms. Comparisons of the amounts of trace elements in an infaunal organism with the amounts in seawater are not representative, since, for many elements, even the concentrations of dissolved ele- ments in the interstitial water of sediments differ signifi- cantly from those in "normal" seawater (Brooks et al., 1968). Comparisons of elemental concentration in orga- nisms with the total amount of element per unit weight of sediment also introduce large errors due to the variable amounts of certain elements that cannot be metabolized by the organisms because they are incorporated in mineral matrices that are not broken down by biological activity. Concentration factors based on comparisons of elemental content of marine organisms with amounts of the element *Emphasis added. associated with organic detritus in the near-shore areas of the sea may be subject to large errors. Here active precipita- tion and adsorption may occur for several elements intro- duced by rivers in soluble or colloidal form. Not only do these elements coprecipitate and adsorb onto the surfaces of organic detritus, but they often adsorb onto the surfaces of clay and mineral particles that may present a greater total surface area than organic detritus. Sediment is ingested along with the food, and only a small fraction of the total ele- ments accumulated by the organisms may have come from the food. The removal of adsorbed stable or radioactive nuclides from the surfaces of organic detritus and sediment during digestive processes of marine infaunal invertebrates is strongly influenced by pH, total electrolyte content, enzymic digestion of biological substrates, and other fac- tors that determine the degree of association between ad- sorbed ions or colloids and the exposed surfaces of ingested material. Not all trace elements associated with organic de- tritus in estuarine areas are adsorbed onto the surface but may have been incorporated into the living material. Some elements do flocculate rapidly from solution or colloidal suspension when river water is mixed with seawater in estua- ries. For these elements, the amounts adsorbed onto the surfaces of organic detritus may exceed by factors of several hundred the amounts biologically incorporated into the detritus (Phelps, personal communication). The problem of defining "concentration factor" in ben- thic organisms is related to the complex relationships of these organisms to the total available trace elements or radionuclides that surround them. Many benthic organisms are of importance from the viewpoint of human food sources and may comprise an important fraction of the total biomass in marine areas of economic importance (Sanders, 1956, 1958, 1960; Sanders et al., 1962; Phelps, 1967). In contrast to infaunal and benthic organisms, concen- tration factors for plankton and nekton may be based di- rectly upon concentrations of the elements in seawater. On the basis of the total oceans, plankton and nekton provide most of the biomass, and, of these, the phytoplankton con- stitute the largest group in terms of both volume and ex- posed surface area. Phytoplankton accumulate nutrient ele- ments directly from water, and their concentrations of stable or radioactive nuclides may be compared directly with the amounts in water for calculating concentration fac- tors without the complication of intervening trophic levels. Water is also the ultimate source of nuclides for the zoo- plankton and nekton, although the radionuclides may be preconcentrated or discriminated against as they pass through one or more trophic levels before being incorpo- rated into the individuals of any given step in a food web or chain. Even for those animals constituting the higher trophic levels, an equilibrium may be established between the ani- mals, their food, and the amount of the element or radio- nuclide in the water. Because the intervening food orga-

Accumulation and Redistribution of Radionuclides by Marine Organisms 165 nisms approach a given equilibrium with the water and are in turn eaten by the animals of the higher trophic level, so too will the organisms higher in the food web approach an equilibrium with the amount of the material in the water. Turnover Rates In the natural environment, the capacity of an organism to achieve equilibrium with a radionuclide will be directly re- lated to the biological turnover rate of the corresponding stable element in the organism and inversely related to the rate of environmental dilution. The length of time required for an organism to exchange half of its total content of a given element is referred to as the biological half-life of the element. The shorter the biological half-life, the faster the organism may achieve equilibrium with the water. Whether equilibrium has become established may be determined by measuring the specific activity of the radionuclide in the organism and in the water (or food, if equilibrium between organism and food is of interest). When the specific activity of the organism is equal to that in the source, empirical equilibrium has been achieved.* In larger animals, the speci- fic activities within various organs often differ from each other and from that in the water for periods of time mea- sured in weeks or months following the introduction of the radionuclides, even in restricted environments. In many organisms, the total amount of a given element or radio- nuclide may consist of several pools, each with a different biological half-life varying from hours to weeks (Figure 1). Since the amount of a given radionuclide in the water from a single introduction usually decreases with time because of turbulent diffusion or other causes, the different body pools of the radionuclide in an organism living in the area will fol- low the decrease in environmental activity at different rates, each dependent upon a particular turnover rate. Under these conditions, the organism is unable to arrive at equilibrium with the environment. The smaller marine organisms, includ- ing the phytoplankton and some zooplanktons, frequently have turnover rates for trace elements with half-times mea- sured in hours. In these organisms, "equilibration" with the changing levels of a radionuclide in the water may be achieved. Although all organisms may approach equilibrium with the environment with respect to a given element, most ma- rine organisms in higher trophic levels probably are not ca- pable of achieving true equilibrium with introduced radio- nuclides of medium or short physical half-life. Failure to attain equilibrium is often caused by relatively rapid changes 'The use of specific activity to determine if equilibrium has been achieved is valid only if the radionuclide and the corresponding stable element are present in the same physicochemical form. in environmental levels of radionuclides in marine areas sur- rounding restricted sites of introduction. It may also be caused by changes in vertical and horizontal distribution of radionuclides. The changes in radionuclide content in a given area of the open sea above the thermocline or in estu- arine areas often occur rapidly in comparison with biologi- cal turnover rates,* especially for the higher trophic levels. However, if a chronic disposal situation exists for many years, organisms that grow up in that environment may be at equilibrium even with radionuclides of medium physical half-lives. Mauchline and Taylor (1964) have discussed equilibrium conditions for the long half-lived 137Cs and 90Sr under these conditions. Surface Adsorption and Incorporation of Nuclides by Organisms The degree to which an element and its radionuclides are concentrated by an organism depends upon the physico- chemical interactions of the element with the environment and with the organism as well as upon the requirements of the organism. Goldberg (1957) showed that, in general, the ability of marine organisms to concentrate metals from the sea paralleled the order of stability of metal ligand com- plexes, and Szabo (1967) reported that the levels of accu- mulation of alkaline earths by mixed plankton occurred in the same sequence as that for cation exchangers such as Dowex-50 in which the order of association is radium > barium > strontium > calcium > magnesium. Grim (1953) proposed that the control of the alkali metals sodium, po- *The reciprocal exchange of atoms of an element between an orga- nism and the environment is referred to as turnover. Turnover rate refers to the fraction of atoms exchanged per unit weight of an or- ganism in a given time period. The net movement of atoms in or out of an organism may change as a result of growth or other changes in the physiological condition of the organism or as a consequence of environmental changes. Not all atoms of an element that move in or out of an organism are necessarily involved in turnover. The net movement of atoms in or out of an organism may be explained as follows: U = Uptake of element by organism L = Loss of element by organism When U = L (net movement = zero, all movement is the result of turnover) When U > L (turnover occurring but net movement = the increase in amount of element in the organism) When U < L (turnover occurring but net movement = the decrease in amount of element in the organism) Turnover rates of an element in an organism have been obtained by measuring the amount of a radioisotope of the element retained by the organism when it is placed in seawater not containing the radio- nuclide. Many investigators have used this method to follow the loss of a radionuclide from an organism with time.

166 Radioactivity in the Marine £nyironment 100 ao 60 40 20 o R..M.-0.1W4l,.46,-0.00«. 20 30 40 DAYS 50 60 70 FIGURE 1 Retention of 95Ni by the Atlantic croaker, Mlcropogon undulatus, showing separation of the curve into two components (A and B). Solid circles represent biological retention (Kg, corrected for physical decay); open circles represent effective retention (R/ . not corrected for decay). (Baptist etal., 1970; reproduced from Health Physics 18:141-148,1970, by permission of the Health Physics Society.) tassium, rubidium, and cesium in the sea involved ion- exchange reactions with the clay minerals of the bottom sediments, with retention on the clay increasing with the ionic radius of the element. Similar ion-exchange reactions for the alkali metals may occur at the surfaces of marine organisms. Several investigators have postulated that surface sorp- tion, including ion exchange, may play an important role in the incorporation of trace elements into food webs by phytoplankton. For the heavier alkaline earths and alkali metals, at least, the major uptake mechanism probably is provided by ion-exchange mechanisms, although some other elements may be adsorbed mainly by chelating processes. Fukai (1966), in a study of the binding of radionuclides on ion-exchange resins, observed that radionuclides of chro- mium, manganese, iron, cobalt, zinc, and cerium rapidly dissociated from Dowex-50 when the loaded resins were placed in seawater, but that more than 90 percent of all of these elements remained chelated to Chelex-100 after elu- tion with seawater. The resin showed an especify high affinity for transition elements. Although ion-exchange resins will not remove most me- tals from seawater, the resins are capable of extracting sev- eral elements from seawater if they are first converted to chelating resins. Carritt (1965) reported that by absorbing the chelating agent dithizone onto Dowex-l-12x, 100 g of the resulting chelating resin could concentrate cobalt, cop- per, nickel, zinc, cadmium, and lead from 2,000 liters of seawater before saturation occurred (see also Boni, 1966, and Callahan et al, 1966). Goya and Lai (1967) studied the adsorption of 37 elements from water by Chelex-100 and observed that more than 95 percent of manganese, iron, cobalt, chromium, nickel, zinc, cadmium, lead, bismuth, and several of the lanthanides and actinides were chelated onto the resin from seawater at the natural pH. More than 50 percent of the mercury, tin, and ruthenium and more than 20 percent of the barium, silver, gallium, strontium, calcium, scandium, titanium, and technetium were chelated by the resin. Less than 0.6 percent of the sodium and mo- lybdenum were taken up, although 1.9 percent of the cesium was adsorbed. If the average concentration factors of the marine phy- toplankton, zooplankton, and attached algae for elements in seawater are plotted against the percentage of the cor- responding elements sorbed from the water by Chelex-100, uptake by organisms and resin in general are positively re- lated (Figure 2). The dashed lines in the figure represent the subjective relationship between the two variables. Five of the elements-titanium, scandium, silver, gallium, and zirconium-appear to be concentrated by the organisms to values higher than would be expected from their uptake by Chelex-100. Titanium, scandium, gallium, and zirconium share a common characteristic of being easily hydrolyzed and forming more or less insoluble hydroxides and ion com- plexes at the pH of seawater. Also included in those that form insoluble hydroxides are aluminum, yttrium, iron, the lanthanides, and plutonium. All of these elements are con- centrated by plankton and seaweeds with average concen- tration factors as follows: aluminum, 70,000; iron, 25,000; zirconium, 29,000; titanium, 10,000; gallium, 5,500; plu- tonium, 2,200; scandium, 1,700; cerium, 51,100; and yttrium, 500. RESULTS OF BIOLOGICAL ACTIVITY Accumulation and Concentration of Elements by Benthic Algae, Phytoplankton, Zooplankton, Molluscs, Crustacea, and Fish The average concentration factors for benthic algae, phyto- plankton, zooplankton, and the muscle tissue of molluscs, crustacea, and fish are shown in Table 1. The concentration