Letter to T Mccraw, Subject: Enclosure of Draft Report on Iron-55 in Rongelap People

Page 1

Environmental Safety Branch Division of Operational Safety The draft of the report on iron-55 in Rongelap people Your comments would be appreciated. Tom Beasley and I feel that there should perhaps be some additional information, especially about current sources of iron-55 at Rongelap, before this is published. If we can obtain additional funds we will try to go to Rongelap with Bob Conard and coordinate that trip with With best regards,

Page 3

Since 1965, the distribution of 55Fe in the biosphere has been studied both in the United States and in the Scandinavian countries. Initially, 55Fe concentrations were determined in Alaskan Eskimos, residents of Richland, Washington, and in 55Ferepresentative foodstuffs of both (1). Subsequently, concentrations in environmental samples and in residents of Finland (2) and Sweden (3) were reported which generally confirmed the findings of the earlier study. Addi~ional research shows I that (i) marine organisms and people whose diet is largely seafood contain the highest concentrations of 55Fe (4); (ii) 55residents of the northern hemisphere have higher Fe body burdens than those of the southern hemisphere (5); and (iii) the 55Fe levels in people reached peak concentrations in 1966 and continue to decrease (6,7). We determined the 55 “ Fe body burdens of natives at Rongelap Atoll in the Marshall Islands. Not only is their diet high in seafoods (8), but the atoll received high-level fallout following the detonation of a thermonuclear device at Bikini Atoll in 1954 (9). We considered it probable that retention of 55Fe at the atoll from that event, coupled with world–wide fallout from largescale nut].ear device testing in 1961-62 could lead to unusual body burclens of this radionuclide in the Rongelapese. The method of separation of ’55Fe was identical to that previously described in this journal (1). The counting technique

Page 4

anticoincidenc”e with an umbrella of nine Geigerwas used to detect the 5.9-kev x-ray emitted in capture decay of . Pulses from the proportional recorded in a 512-channel multichannel analyzer. The detection syste”m was surrounded by four inches of lead .. shielding and the resultant background counting rate under the Fe photopeak was 1.7 counts min . Disintegration rates of electrodeposited from a standard solution obtained from the National Bureau of Standards. Correction was made for the self: absorption of the X-rays in the electrodeposited iron. Stable iron was determined calorimetrically, using o-phenanthralein as Body burdens were estimated by first measuring the 55Fe in a known volume of blood (4-26 ml). Total blood volume was estimated using body weight and average blood volumes of 82 ml blood per kg body weight weight for females. The . . 65 per cent of the total for males and 2’4ml blood per kg body iron content of blood was assumed to be ‘ body iron. This method of calculating total blood volume is that used by Persson (11) in his estimate of 5’Fe body burdens in T,apps of Northern Sweden. Previous estimates of body burdens from composite blood specimens (1,5) were macle assuming that the average total blood volumes of 5 liters and that 60 per cent of the total iron is in the blood.

Page 5

with the Rongelap datal t> = average Fe body burclcns agreecl within 15 per cent of those calculated using body Table 1 gives the average body burdens of 55 Fe in a selected group of Rongelapese sampled in March, 1970, and Figure 1 shows a frequency distribution of the body burdens of males and females. Iron-55 levels in the blood samples were sufficiently high to permit count rate measurements to + 5 per cent at the 95 per cent (2c) confidence level. Not..all donors were weighed in 1970, therefore body weights from previous years were used to compute total blood volumes. However, weights from previous years “ applied mostly to creases with age, and therefore the younger donors. Since weight generally insome individual estimates of Fe body burdens, ‘ averages shown in Table 1, are likely to be The maximum body burden in the males was 0.85 KCi, while three females had body burdens greater than this value. The maximum observed female body burden was 1.0 pCi, approximately l/100th of the maximum permissible body burden which has been the total body as the critical organ (18). Previous measurements le body burdens during a period of increasing Fe fallout generally showed that Fe body burdens of females were higher than those of males (4,5) . Presumably this is due to higher +!lS,I-/C-

Page 6

“ rat”es of iron in xemale S than in ma,las{ with the result that “ females are more nearly at equilibrium with their environment. “As environmental levels of 55 Fe decrease. females should, on the average, reflect this change by exhibiting lower Fe body burdens than those of males. Figure 1 shows that more female . body burdens tended toward burdens were more normally on”body burdens showed a significant Regression analysis of age. burdens. This observation is consistent with earlier data from Comparison of the 55Fe’body burdens of peoples of different . . countries (5) requires knowledge of the turnover rates of in the environment and in humans. Jennings (12) has shown that the ‘5Fe .pecific activities of salmon taken from the northeast Pacific Ocean decreased eightfold between 1964-1967. Assuming that a first eider reaction governed the removal of Fe from the mixed layer of the ocean (upper 100.m) he calculated the in cattle and rain waters show decreases,, but at lesser rates (6). Iron-55 body burdens of adult males in Richland, Washington, decreased approximately fourfold between 1967 and 1970 (7), corresponding to an effective Fe half-life of 1.5 years. If. . the 55Fe turnover rates of Richland, Washington, residents are

Page 7

6imilar to those of insular populations, we conclude that people from maritime cultures would exhibit similar and perhaps faster turnover rates of 55Fe because of the short “ecological half-life” (13) of this radionuclide in the marine environment. The highest 55Fe body burdens previously measured were female natives at Bethel, Alaska, during 1966 (5). The average “body burden of eighteen females was 1.1 vCi. During the same 55.- year, the average Fe body burden of females and males at TokaiMura, Japan, was 0.92 pCi and”-0.63 ~Ci respectively (5). If the effective half-life of 55Fe in Richland, Washington, males and in Tokai-Mura males is comparable, by 1970 the latter group would be exp”ected to hav~’body burdens of approximately 0.1 ~Ci, four times lower than those measured in Rongelapese males.. . Iron-55 body burdens of females at Tokai-Mura might be expected to be . comparable to those of males for reasons discussed earlier. All of the donors of the Rongelap study were subjected to external radiation during the accidental contamination of Rongelap Atoll in 1954. ‘Because of the high levels of radioactivity at the Atoll, the Rongelap native’s we’re moved to Majuro Atoll where they resided for three and one-half years. Following exposure in 1954, whole body counting and urinalyses disclosed measurable quantities of internally deposited fallout radionuclides. By 1957, however, the only radionuclides present in the Rongelapese (9) . No 55F“e analyses were perfornled at that time so bocly burclens

Page 8

of this radionucl icte are ])ot known. Howe VP ‘“f based on the levels small (9). The Ronge lapese returned to the atoll in mid-July, 1957 l Whole body counting and urinalysis measurements one year later disclosed significant body burdens of several radionuclides, the most notable being 137c~ It is probable that 55 burdens increased similarly. The concentrations of 55Fe in a selected species of fish and in soils are listed in ~able 2. Actiyity in the fish is based on wet weights for comparison with previous work. Pooled samples were used in the analyses to reduce the effect of withinsample variation. The Fe values for soil samples collected,. in 1963 at both Kabelle Islet and, Rongelap Islet are averages of samples collected at depth increments of O-1.3 C1l’1 and 1.3-2.5 cm. Specific activities are not given for soils since varying . amounts of pre-1954 plant detritus could significantly alter the stable iron content but not the Fe activity. In this instance, activity per unit ,weight of soil is a better index of changes which occur as” a result of input or loss than is specific The decrease in Fe’specific activities in Mulloidichthys, sp. (goatfish) at Rongelap between 1959-1961 corresponds to an ecological half-li~e of 11 months, identical to that observed by Jennings (11) . Higher specific activities may have occurred

Page 9

at earlier times, however, estimates based ..lly 011 expone~~tial loss would not take into account possible retention and cycling of “ Fe within the lagoon, or and maximum specific activity The increase in specific the time lag between deposition in the aquatic biota. activity of goatfish liver between . late 1961 and mid 1963 reflects the increased environmental concentrations of Fe resulting from testing nuclear devices in 1961-62. Introduction of this radionuclide to Rongelap Atoll can occur both by atmospheric-””fallout and by water transport of radioactivity from oceanic regimes. The westwar”d-rnoving North Equatorial Current is con~prised of waters from ‘northern latitudes where fallout from the 1961-62 tests were maximal (14). Speeds of surface currents in both the California and North Equatorial Currents are sufficiently large to account for transport of waters from 30-40° N latitude to Rongelap Atoll (15) in the time period 1961-63; similarly, maximum surface deposition of occurred in 1963 (14) and thus the data of Table 2 probably reflect contributions from both sources. (16). .- Our measurements of ’55Fe in soils collected from the same sites between 1959-1963 do not clarl’fy either of the input sources mentioned +ove; the change in concentrations are greater than can be accounted for by physical decay. Natural processes which remove Fe from the upper 2.54 cm of soil may preclude its use as a precise collecior, canaltherefore the results are.

Page 10

useful only to indicate order of magnitude values of Fe soil :concentrations present at the collection time. Unfortunatelyit is not possible to offer a clear argument in explanation of the 55 Fe body burdens of the Rongelapese presented here at this time. Samples from 1963 through 1969 would have shed light on the problem, but none are available for analysis. The possibility of Rongelap lagoon acting as a intriguing. Removal and retention of both stable Fe and “from the North Equatorial Current could lead to high specific activities of the radionuclide in species important ,ln the Rongelapese diet. The fact that livers from mature goatfish contain between 2-3 times as mucfi stable iron per unit wet weight as do livers of mature salmon (4) is consistent with this argument. Marsh’allese in general show a tendency toward anemia (9) and thus may absorb more iron from their diet than do non-anemic individuals. The 55Fe body burdens of the Rongelapese may there- “. fore only reflect more complete uptake of iron rather than uptake .of iron of high specific activity. It is clear that further measurements of the specific activities of Fe in the diets of the Rongcl.apese and the effective half-life they display for this radionucllde will be needed to clarify these possibilities.

Page 11

REFERENCES AND NOTES 1. H.E. Palmer and T.M. Beasley, Science 149, 431 (1965). 2. T. Jaakkola, in Radioecological Concentration Processes, 3. B. Persson, in Radioecological Concentration Processes, 4. “H.E. Palmer and T.I!. Beasley, Nature 211, 1253 “(1966). H.E. Palmer and T.M. Beasley, Health Physics 13, 889 (1967) . I-I.E.Pa~mer, et. al., Radiological Health Data Reports 9, ‘.— C.E. Jenkins and J.C. Langford, Pacific Northwest Lab. Rep. (T-446) (1967) . This report contains a brief review of past medical surveys on Rongelapese natives and a bibliography listing medical reports, radioecological studies, and graduate theses which deal with work conducted at E.B. scmdcll, in Calorimetric Determination of Traces of

Page 12

E.E. Held, Pacif. Sci. 14, 18 (1960). (Univ. of Wash. Press, 1965), ppi 68-94. The higher specific activity noted in goatfish liver in waters at Rongelap Atoll than’ does the sample of March 1963. Livers from immature goatfish were used in the August specimen; - while the Ma”rch sample was obtained from mature fish. Recent measurements of the specific activity of immature and mature goatfish liver collected at Johnston Atoll in 1968 showed that immature fish livers contained less stable iron 55Fe per unit wet weight than did livers from mature We assume that the immature fish are in forming their iron stores and therefore. the environmental specific activities than do the mature fish whose iron stores are already formed. In the latter case, exchange rates may be slow.

Page 15

soil samples from Rongclap Atoll. Sample numbersanti in parentheses; stable Fe and 55Fe are expressedare . kilogram of wet tissue or-per kilogram dry soil.per * Sample counting times weue arranged to determine Fe in biological samples to f 5 per cent at the 95 per cent confidence level (2T) ; soil samples to + 5-20 per cent at the 67 per cent confidence level (16) . Stable Fe determination, * 10 per cent at the 67 per cent cohficlence level (1~) . Activities are corrected to