Atoll Soil Types in Relation to the Distribution of Fallout Radionuclides (Rongelap Atoll Study Resulting from Bravo Event, Castle Operation Fallout)

Page 2

Biology and Medicfne ATOL.LSOIL TYPES IN RELATION TO THE Laboratory of Radiation Biology University of Washington Operated by the University of Washington under Contract No. AT(45-1)1385 with the Unibed States Atomic Energy Commission.

Page 3

The rediskributiionof radionuclides in atoll soils following falloutfrom a nuclear device is described. The soils are calc~eous, containing no inorganic COI1OMS, and their exchange capacityis directly related to organic content. Comparison of 9amrna-rayspectra of depth increments from young and old soils ~Mws that cs137 and S1312’move most readily in old soil, while theprincipal gax;;:~-emittingradionuclide moving in young soil differencesin :LEtical movement between soil types is obscured @ the highly va~~::~lesurface distribution of the radionuclides. l’hereis a vertical gradient in the distribution of radionuclides 137 90evenwithin the surface inch. Litter redeposits Cs and Sr 65at the soil surface and bird droppings have added Zn and Co’”. ~ young SOilS the highest levels of radioactivity are associated withsoil algae f~~d as a suface crust in undisturbed areas andin coral fragments in eroded areas. Horizontal movement is localizedand probaly is of little overall importance. Buried organichorizons contain more Cs137 than ~d roots are generally more radioactive soilexcept at the soil surface. Pumice adsorbradionuclides but pumice is found adjacent soil layers, than the surrounding particles in the soil CalInixingby animals in old soils and by erosion in young soils

Page 4

1S important in the redistribution of radionuclides near the soil surface, C;Z37 and Sr9a are the principal radionuclides entering 8 wle within tl~esoil-plant system. Any loss from this system appearsto be small, but a definite conclusion can not be drawn

Page 5

TABLE OF CONTENTS Depth Gradient of Gamma Spectra Litter and Guano Depth Gradient of Sr90 Zhmied Organic Layers

Page 6

LIST OF TABLES 1. Nutrient levels and Al horizon of the soil reaction in the surface five most extensive soil series. samples from two sets of plots. m. Composition of representative soil types, from Rongelap Atoll (after Kenady 1962). w. Relative Sr 90 ccmtent of depth increments collected in 1959 from two Rongelap Atoll soil types. v. ~r90 and cclcium in the surface 1-1/2 inches of Rongelap soil, Rongelap gravelly sand, collected

Page 7

Beach Ridge Sand, Rongel&p Island. Gogan se~~~~, Kabelle Island. Lomuilal Sand, Kabelle Island. Gamma-ray spectra of equal amounts of old soil from ment from old soil compared with the gamma spectra Gamma-ray spectra of depth increments of old soil taken from the surface inch. Gamma-ray spectra af litter and guano collected in an area of old soil. 9* Ganuna-rayspectra of equal amounts of young soil from different depths. 10. Gamma-ray spectrum of young soil from a depth increment of 9 to 10 inches compared with the gamma 11. Radioautqraph obtained by exposing no-screen X-ray film approximately one cm below the soil 12. Radioautographs and photographs of soil core 130 Jh area of young soil subject to erosion. 14. Photograph and radioautqraph of old soil showing correspondence of radioactive portions to organic

Page 8

ATOLL SOIL TYPES IN RELATION TO THE DISTRIBUTION OF FALLOUT RADIONUCLIDES The redistribution of radionuclides in atoll soils following contaminationwith radioactive fallout is the subject of this Marshall Islands, in the central opportunity for such studies Rongelap Atollf northern Ocean, presents a unique since it was substantially contaiiinatedwith radioactive fal~out t Only once. The fallout resulted from a thermonuclear device detonated at Bikini Atoll eighty miles to the west on March 1, 1954l Although there was some additional contamination from nuckar tests in 1956 and 1958 the total contribution of radionuclides from the fallout of these subsequent test series amounted to a fraction of one per cent of the amount from the 1954 fallout. Gamma radiation dose rates at Rongelap at detonation plus one day ranged from 3.5 r/’hrat the southern islets of the atoll to 35 r~r at the northern islets (Dunning 1957). These rates declined at approximately the rate predicted for mixed fission products by Miller and heb (1958). Rongelap Atoll has a lagoon area of 388 square miles and an average depth of 168 feet (Nugent 1946, p. 748). The emergent land area is about three square tniles,,consisting of :

Page 9

acre to the largest, Rongelap, in size from a fraction of an which is four miles long and one-halfmile across at its widest point. There is one small iskt on the western reef and the other islets extend along the northern,eastern and southern reefs. The islets on the northern reef are not as well developed as those to the east and south. There are two seasons--a dry season from December to March and a wet season from April to November. Annual rainfall is less than fifty inches, and there is no well-developed fresh-water lens. Some importanb features of Rongelap Atoll? including aerial photographs are given by Wlens (1962). Class~fication and mapping of the soil t~es at Rongelap primarily calcium carbonate, originating from corals, foramhifera, coralline algae and mollusk shells. There is a very small amount of pumice drift in the soils. Since these soils contain no inorganic colloids, exchange capacity and organic content are linearly related. In some areas, particularly along the seaward sides of the islets, buried Al horizons are deep as eighty inches (Fig. 1). These highly organic presumably result from storm debris covering previously established soil and vegetation. The pH, determined in the field from a 1x1 soil-water ratio with a Beckman Btadel N-2 pH meter, is generally between

Page 10

Micromonolith of Pit 35 Sot! pit 35, Rongelap Island. Dark layers ore buried horizons. Fig. 1. Beach Ridge Sand, Rongelap Island. This soil typeoccurs on the island beach ridges. The pit shown,...<above is seven feet deep and contai s six buried -.

Page 11

7 and 8 in the surface horizon but occasional values from 4.6 to 9.5 were found. The pH increases with depth and decreasing The amounts of exchangeable cations in the different soil types are given in Table I . The calcium content is so high that more calctum is brought into solution with repeated extractions. Consequently, strontium units have little or no signific~ce relative to atoll soils and, when given, are based m total calcium rather than on exchangeable calcium. METHODS OF COLLECTION AND The vertical distribution of radioactivity in soil and litterwas studied by analyzing samples taken mostly by l-inch soil increments, in a few cases by 1/4- and l/8-inch increments and by radioautographs of sections of soil cores the method of Held et al. (1965). The increment were made during both the wet and dry seasons 1961 aid 1963. The cores were collected only there i,s considerable horizontal variation in the levels of radioactivity (Table II) each set of increments was collected to insure the sampling of a single vertical column. The large amount of horizontal variability also made it profitable to compare the relative amounts of radionuclides at different depths from many profiles than to make precise

Page 14

6pectraof equal amounts of soil from different depths were _ared directly to arrive at (1) a qualitative evaluation Of the vertical distribution of the radlonuclides and (2) a Semiquantitativeestimate of relative amounts of radionuclides at different deptlns. Gamma-ray spectra were made with a system sodium iodide crystal a~~la 256-channel analyzer. In addition, for analyses of the pure betaRESULTS AND DISCUSSION Vertical distribute.nnin relation to soil t~e is made by comparingyoung and old soils. Examples are Kabelle Sand, a young soil, (Fig. 2) and Gogan Gravelly Sandy Loam (Fig. 3) and !tmluilalSand (Fig. 4), old soils. Their characteristics are given in Tables I and XII. The young soil has little organic material except that in tm algal surface crust about 1 cm thick. Hermit crabs, Coenobita perlatus, at the base of a few shrubs~ the borsowing ghost crab, Ocymode ceratopthalmia, and ants are the maf.nanimals present. Litter accumulation is found only at the bases of the scattered shrubs and is a minor part of this soil ~stem.....,.,.>

Page 15

Black areas are crusts formed by algae, l I General view of pit 6 ores. Fig. 2. X.aks Ue Sand. A very young soil conmosed. lagoon Bands deposited on the lagoon-side

Page 16

Stond of Pisonia grand is, Kab.e!le ls!cnd, pit 38 area. Fig. 3. Gogan series, Kabelle Xsland. This is a well- 1 developed, productive soil, usually associated

Page 17

strata are buried horizons Messerschmidia argentia at soil pit 12. Fig. 4. Lomuilal Sand, Kabelle Island.

Page 19

M contrast, the old soil has well-developed A. and Al horizonsand supports dense vegetation, which produces a heavy litterfall during the dry season and which contributes to the redistribution of radionucli.des in the system. Sea birds nest h the vegetation and land crabs, predomiEarthwoms are seld~ found and terrestrial isopods, although foundin old soils on some few soil insect% which are beeties as well as ants. mainly tenebrionid and ;;zabid Depth Gradient of Gamma Spectra The gamma sFectra, with background subtracted, of the o to l-inch, 1 to 2-inch, and 9 to 10-inch depth increments from old soil collected in March 1959 are given in Fig. 5. The spectra of the increments between z inches and 9 inches have been Omitted from the figure for clarity but shed a gradual change from the condition at 1 to 2 inches to that at 9 to 10 inches. The radismuclides corresponding to the photopeaks are indicated in the figure. 60 65 and h 155 show significant counts in the surface increments but are not detectable in samples frcm greater depths. The amounts

Page 20

Fig. 5. Camrm-ray spectra Oi eq~al amounts of old SOLI from Mfferent depths.

Page 21

photo@~& region of the O to l-inch increment, the relatively broadpeak is a combination of the 0.60-mev photqeak of Sb 125 hcrement the peak becomes sharper and is oriented toward the 0.66-mevphotopeak of Cs137-Ba137m. At the 9 to ment there is almost complete orientation toward ~a137m 125peak, with little Sb remaining. In Fig. 6 a comparison k given of the spectrum of the 9 to 10-inch increment (Fig. 5) Fig. 7 S?IOWSthe gamma spectra of increments taken from an undisturbedarea in 1958. The first spectrum is from the O to depth increments to a depth of 1 inch. The highest levels of , the rare earths, and CO , which move m.cmeslowly than ~J37 or Sr904 are in the 1/2 to 5/8-inch increments, whereas the CS137, which moves most rapidly in this soil type, is in Litter and Guano The gamma-ray spectrum of litter, consisting of leaves, twitisand floral parts splattered with tern droppings, collected 125much higher than the Sb penk, me 1017 peak photopeak to be skewed to the left, indicating &he gresence of the 1.12-mev

Page 22

Fig. 6. Gina-ray spectrum of a 9 to 10-inch depth increment from old soil compared with the

Page 23

Fig. 7. Ganuna-rayspectra of depth increments gf old soil t~ from the surface inch. The origin ef each spectm~ been arbitrarily separated from the preceding one hi factor of ten to facilitate Inspection of the fiyuo.

Page 25

Fig. 8. Gamma-ray cpectra of litter and guano collected in an area of old soil.

Page 26

~n65peak. The presence 035 Zn65 is corroborated by the 0.5-mev Fig. 8 also shows the gamma-ray spectrum of noddy ternpxko - guanocollected in this area. The 1.12-mev peak predominates 60 65overthe 1.17 peak of Co and the 0.51 peak of Zn is evident. me foliar contribution to the litter contains only Cs 137 from In undisturbed areas Cs137 90 and Sr are being deposited witiithe litter and are thus replacing at the surface some of theCs137and Sr90 lost by leaching. There is not sufficient data from the field work to determine whether there eventually’ 9 willbe a loss of these radionuclides from the soil-plant systems Or a steady state (excluding physical decay of the radionuclides). mmolith lysimeters and controlled and uniform addition of the radionuclideswduld define this point. Fig. 9 gives the spectra of the O to l-i.nclh,1 to 2-inch, $ and EU were detected only in the surface layers, and with increasing depth the 0.60 to 0.66-mev photopeak region of trum of the 9 to 1(1-~nch increment is compared with that of an ~125 spike in Fig. 10, showing that the photopeaks of the soil and spike gamma spectra are identical.

Page 27

Fig. 9. Gamma-ray spectra of equal amounts of young soil from different depths.

Page 28

Fig . 10. (%~ma-ray spectznm of young sofl from a depth increment of 9 to 10 inches compared

Page 29

The radionuclide content of leachates collected in the field ngrees with these results (Cole et al. 1961). Leachates from 125yOung soil contained only !% and Sr90~ while leachates from betterdeveloped soils contained mainly Sr90 137, ~lth The relative Sr90 content of depth increments of the two soil types is given in Table XV. There is a rapid decrease in tent between soils probably are not significant. The extreme values from the results cate samples taken from of Sr’” analyses of subsamples of replismall areas differ by a factor of more than ten (Table S1). It is likely that the variability is due largely to the spotty nature of the distribution of the fallout radimuclides, which is evident in the radioaut~raphs discussed below and from X-ray films that were exposed at the soil surface and just bel~ the surface (Fig. 11), and to small differences in the characteristics of the SOi,l within a single Sampling by l/8-inch increments in 1959 of an undisturbed old sOil, (Pig. 4 and Table I ) on Rongelap Island indicates a gradient of Sr90 levels in the top inch of soil (Table V) . me levels in the second inch are about one tenth those in the

Page 31

Fig. 11. Radioautograph obtained by exposing no-screen X-ray film approximately one &m below the soil surface in situ..—

Page 32

Table V. @o and Ca in the surface 1-1/2 inches of Rongelap soil, Rongelap Gravelly Sand, collected in September 1959. Percent of total in

Page 33

firstinch and remain at approxhately the same level from 7/8 inch to -1-~/2 inches in depth. Radioaut~raphs of cores taken in 1963 show marked differences in the distribution of radioactivity. Fig . 12 ShOWS photo. graphs and the corresponding radioautqraphs of sections of pores of young soils from an undisturbed area (Fig. 2) and from an area t Subject to erosion (Fig. 13) and an old soil (Fig. 3). The radioactivity corresponds closely to the dark area in the photograph Of the core from the undisturbed young soil. This dark area is composed almost entirely of a mixture of soil algae, forming a crust which has retained most of the fallout radionuclides. In an eroded area of young soil the radioactivity is associated with large coral fragments which are infiltrated with algae. m~s area is subject to erosion by both wind and water, which accounts for the coral fragments containing algae ~d radionuclides occurring beluw the soil surface. Radioactivity is not associated primarily with smaller particles as

Page 36

in other soils, presumably because smaller particles have been erodedaway since fallout occurred. A radioaut~raph of a core from the old soil (Fig. 12) shows that the radioactivity extends to a greater depth and Is more diffusely distributed than In the young so~ls. The surface littercontains little radioactivity compared with the soil There is evidence of localized horizontal movement of radionuclides in the soil, …