Evaluation of the Radionuclide Concentrations in Soil and Plants from the 1975 Terrestrial Survey of Bikini and Eneu Islands

Page 1

EVALUATION OF THE RADIONUCLIDE CONCENTRATIONS IN SOIL AND PLANTS FROM THE 1975 TERRESTRIAL SURVEY OF BIKINI AND ENEU ISLANDS Prepared for U.S. Energy Research & Development Administration under contract No. W-7405-Eng-48

Page 2

This report was prepared as an account of work sponsored by the United States Government. Neither the United States nor the United States Energy Research & Development Administration, nor any of their employees, nor any of their contractors, subcontractors, or their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for thc accuracy, completeness or uscfulness of any information, apparatus, product or process disclosed, or represents that its use would not infringe Reference t o a company or product name does not imply approval or recommendation of the product by the University of California or the US.Energy Research & Development Administration to the exclusion of others that may be suitable. Printed in the United States of America National Tecbmical lnformslion %race US. DeDartment of Commerce Price: Printed Copy $ PI Range hice ' A d d 12.50 for wch additional 100 page increment from 601 to 1,000 p y : ;

Page 3

LAWENCE LIVERMORE LABORATORY EVALUA TlON OF THE RADIONUCLIDE CONCENTRATIONS IN SOIL AND PLANTS FROM THE 1975 TERRESTRIAL SURVEY OF BIKINI AND ENEU ISLANDS

Page 5

Environmental Sampling and Laboratory Analysis ...........4 Results and Discussion ..........................12 Surface Soil Survey ........................ 12 Prediction of Plant Uptake .....................17 Summary and Conclusions .............,........... 31 Appendix A. Soil Profiles of Bikini and Eneu Islands .......... A-1 Appendix B. The Distribution of Radionuclides with Depth in Soil Profiles of Bikini and Eneu Islands ..........B - 1 Geographical Distribution of Radioactivity in the Concentrations in Soil and Vegetation on Bikini and Eneu Islands . . (microfiche in pocket on inside back c o v e )

Page 7

EVALUATION OF THE RADIONUCLIDE CONCENTRATIONS IN SOIL AND PLANTS FROM THE 1975 TERRESTRIAL SURVEY OF BIKINI AND ENEU ISLANDS In June 1975, personnel from LLL and from other laboratories and agencies conducted a radiological survey of the terrestrial environment of Bikini and Eneu Islands (Bikini Atoll) to evaluate the potential radiation dose to the returning Bikini population. In this report, we present measurements of the radionuclide concentration in soil profiles and in dominant species of edible and nonedible, indicator plants. We also describe the use of these data to derive relationships to predict the plant uptake of radionuclides from soil. Approximately 620 soil and vegetation samples from Bikini and Eneu Islands were analyzed by Ge(Li) gamma spectrometry and by wet chemistry. The predominant radionuclides in ral, the radionuclide concentrations in soil from Eneu Island and from the four areas of Bikini Island appear tions. The median surface-soil concentrations (pCi/g) of Eneu Island and 0.22 for 241Am) are ten times lower than those measured on Bikini Island (0.86 for 6oCo, 76 for 'OSr, 2.4 for 241Am). We found that radioactivity is unevenly distributed over the surface of these islands and that the distribution of activity with soil depth varies greatly in different parts of the islands. Concentrations in the soil of greater than concentrations of and 239'240Pu which, in turn, are greater than concentrations of 6oCo. To quantitatively evaluate the concentration factors as well as tion ratios for indicator and edible plant species from the same location. In general, the concentration factors for cs in terrestrial vegetation concentration factors for both of these nuclides exceed those for 2399240Pu by one to two orders of magnitude (10 to 100 times). For 43 for 137Cs, 3.0 for 239,240pu, and and 2339240Pu, nuclide uptake by fruit is less than that by mature leaves; however, the opposite is tribution of the individual plant

Page 8

varies with the nuclide under consi- concentration ratios to predict de ration. Thus, we also describe nuclide concentrations in fruit from the use of concentration factors and those observed in soil or leaves. Since the termination of nuclear testing on Bikini Atoll in 1958, periodic environmental surveys have been conducted to evaluate the radio - logical status of the atoll. The early surveys of Bikini Atoll as well as the recent survey and assessment a t nearby Enewetak Atoll indicate that concentrations of radio nuclides in certain terrestrial foods are relat iv ely high,suggesting that the terrestrial foodchain could bea major exposure pathway. I n June 1975, Bikini Atoll was resurveyed to deter - mine the residual.radioactivity in the terrestrial environments of Bikini and Eneu Islands,the two main islands of the atoll(Fig. 1). The 1975 survey included measurement of environmental gamma-ray exposure rates and the col - lect ion and analysis of samples of soil, ground water, cistern water, and vegetation for use in assessing the internal dose via various ingestion pathways. (The dose from external gamma eapeslrse apd the radio nuclide concentrations in cistern and ground water have been previously The longer - lived fission and activation products are the nuclides of primary concern at Bikini Atoll. Previous studies have shown that because of their long half. - lives and 239’240Pu contribute nearly all the population dose from the terrestrial tai ned for 6oCo and 241Am are also In this study,our major However, the results obincluded because 6oCo is widely d is - tribute d and is present in the marine pathway and because the concentration due to t h e decay 241pr. results of the soil and ve&e&&iSn studies of the 1975 survey. We also discuss the use of the data to derive relationships that predict the plant uptake of nuclides from soil wa Bikini Atoll. Geometric iruaon values Am surface - soil concentrations are developed for Eneu Island and for each of the four areas into which

Page 9

Bo kae tok tok I analyzed soil profiles to investigate the distribution of activity with and soil-plant regression equations, fruit concentration ratios, are 239'240Pu in edible and indicator plants as well as in soil from the associated sampling site. From our evaluation of these data, a method is developed for predicting the nuclide concentrations in edible plants at a given location from the determined nuclide concentrations in soil at the same location. The predicted nuclide concentrations in edible plants subsequently serve as input data to predict the internal dose from ingested terrestrial foods. This dose assessment is presented in a separate report.9 At Bikini Atoll, the uptake by plants of radioactive material from the soil is the principal source of foodchain contamination. Thus, ooil is both the convenient and the logical starting point for a prediction of

Page 10

radio nuclide concentration in terresfactors or soil - plant regression equat ions are commonly used for a quant it a - tive comparison of the capacity of different plant species and various plant organs to accumulate radionucli des through soil - root uptake. Prediction of radio nuclide uptake by edible plants is needed to convert the measured environmental soil concent rations into the potential dose to man from the soil - plant pathway. Where fruit samples are not available, correlations between the concentrations in leaves and fruit of a particular plant species or between con cent ra - t ions in leaves of indicator and edible plant species enable us t o predict plant uptake from soil. ENVIRONMENTAL SAMPLING AND The objective of the surface - soil survey was to define the distribution of radioactivity within the soil on Bikini and Eneu Islands.This survey was conducted ina manner similar to that used a t Enewetak Atoll. ’ number of soil samples collected on each island and within specific areas on the island was a function of the anticipated radioactivity levels, the various housing locations under consideration, and the expected living patterns of the future inhabitants. Thus, Eneu Island,because of it slow and homogeneous activity levels was sampled less densely than Bikini Island.Bikini Island has elevated and more variable activities and was divided into four distinct areas, each o f which could b e used for future village sites. Sampling sites were selected by superimposing a rectangular - grid network over an aerial photograph of each island and randomly choosing the grid squares to b e sampled within each specific area t ions for the islands are shown i n Fig.2. These samples were taken with a coring tool(a steel pipe, 30 15 cm. The surrounding soil was scraped away and a cutting tool(a flat piece of steel) was inserted underneath the cover,freeing the In addition to the surface - soil sampling program, vegetation and files taken from the same location as the plant) were collected wherever suitable plant species were located on Bikini and Ene-u Islands (see Fig. 3 , Appendices A and B). Leaves,

Page 11

fruit, roots, litter, and stems of edible (Pandanus, breadfruit, coconut, papaya, banana, and squash) and nonedible indicator (SeaevoZa and Messerschmidia) plants were collected when available. We attempted to take at least one soil profile, and preferably as many as three, through the root zone of each sampled plant. In addition to the soil profiles taken through the root zone of sampled vegetation, other soil profiles were collected on a random basis on both Bikini and Eneu Islands. The geographical locations of these profiles are also shown in Fig. 3. All profile samples were taken from pits dug with a backhoe. the pit was dug, the sidewalls were carefully cut back a few centimeters to ensure a clean, undisturbed profile. For each profile, 100-cm samples were collected from the sidewall at increments of 5 cm in the upper part and of 10 to 20 cm through the lower part of the profile. Total depth for profile samples varied from Vegetation and litter samples were carefully selected and classified by age. For example, collected leaves were classified as young, mature, and senescent. Two ages of litter were readily identifiable and were collected accordingly. However, fruit samples representing different stages of growth were unavailable. A l l samples were handled separately, placed in plastic bags, and sent to LLL for processing and analysis, All soil and vegetation samples were trosocopy and by wet chemistry for the following radionuclides (see Ref. activities appear to approximate lognormal distributions and thus, we calculated geometric means of 6oCo, concentrations in soil for Eneu Island as a whole and for the four areas of interest on Bikini Island. For each profile collected, we plotted (on semilog paper) the concentrations of the selected nuclides as a function of depth. The profile data were compared in an attempt to characterize the different areas o f the islands Because they are the major contributors to the dose from ingestion of selected for more detailed analysis. for these nuclides from measured

Page 14

Paci f ic Ocean Paci f ic Ocean Fig.3. Sampling sites of the Bikini (a) and Eneu (b) soil profile and

Page 15

concentrations in plant samples and from the average measured concentration in the associated 0- to 25-cm soil profile. We define the concentration To reduce the variability in average soil concentrations (used to calculate the concentration factor), we used ,=. 0- to 25-cm soil profile that encompasses a large fraction of the effective absorptive root zone rather than the deeper 0- to 55-cm profile that encompasses the entire root zone. Concentration factors calculated on the basis of the average soil concentration in the upper 2 5 cm of the profile are somewhat greater but do not differ substantially from those based on the deeper profiles (Tables 1 and 2). All concentration factors reported here are therefore those derived from average 0- to 25-cm soil The average 0- to 25-cm soil concentration is calculated as the weighted geometric mean for the separate concentrations, measured at various increments throughout the profile. Concentration values less than the minimum detection limit are set equal to the detection limit, following the U.S. Environmental Protection Agency technique. Whenever the number of samples is large enough, soil concentrations are plotted against concentrations in plants from the same sampling site; the results are analyzed with linear regression methods. These linear regression results the 0.1-level of a standard F test. For each combination of nuclide, plant organ, and species considered, statistically significant regression equations are compared to the median of the calculated concentration factors and a single representative concentration factor is assigned. Predictions of radionuclide levels in foodstuffs can be made from concentration factors if measured soil concentrations are available; however, concentration , ratios are also needed if the only available data are from mature leaf samples. The concentration ratio is defined as the ratio of the concentration in fruit to the concerttration in leaves of the same species; or, as the concentration in leaves of one species to the concentration in leaves of another species. We calculted preliminary ratios for all available species from the 1975 Bikini survey. However, because o f the small number of samples involved, a statistical analysis of these results was not possible.

Page 17

Table 2. soil - mature leaf concentration factors calculated for 0- t o ~ 5 - c ~ and deeper soil profiles. a Sample locations are shown i n Fig. 2.

Page 18

Results and Discussion SURFACE SOIL SURVEY Although more samples are available from the 1975 Bikini survey than from is little consistency in the peograph137cs, 2 3 9 y 240Pu, and 241Am on Bikini The maps and overlays in Appendix C present the activities of these radionuclides in picocuries per gram of dry soil over the sites from which the samples were collected. A list of concentrations of all detectable nuclides for each sampling site is included in pocket on inside back cover). A dry-soil density of 1.5 g/cm3 may be used to convert the integrated profile uata into activity per u z i ~area. However, some caution must be exercised in such calculations because a significant fraction of the total activity may be located below the sampling depth. Table 4 presents the means of the surface-soil concentrations of the dominant nuclides for Eneu Island and for the four areas of interest on Bikini Island. The values for Eneu are consistently ten times lower than concentrations for any part of Bikini Island. As expected, on Bikini Island, Area 1 shows the lowest soil concentration, since it is an exposed beach area that has been cleared for housing. Data from Bikini' and Enewetak12 have revealed that soil activity is directly related to the amount of vegetation present in the area surrounding the sampling site. One possible reason for this is that a heavy vegetative cover can protect the underlying soil, minimizing the effects of weathering processes (e.g., wind and rain erosion) that transport surface activity through the soil column to the water lens. Follow-up field work at Enewetak Atoll has also shown that, in heavily vegetated areas, litter increases the soil Although soil concentrations of radionuclides in Area 2 appear to be higher f o r 9oSr and 239 ,24OPu than in any other area on Bikini, statistical analysis of the wSr concentrations for each of the four areas on Bikini, using the MannWhitney nonparametric test, shows no significant difference between the concentrations i n the various areas. However, a more extensive analysis is needed to better define the …