Application of the Weibull distribution to describe the vertical distribution of cesium-137 on a slope under permanent pasture in Luxembourg

The application of solutions of the Advection-Dispersion-Equation (ADE) for soil profiles is sometimes questionable. An alternative method, based on the Weibull distribution, has been developed, which can approximate the vertical distribution of radiocesium (137Cs) and allows extrapolation to account for the complete inventory. The structure of the equation allows simple parameters describing the soil depth profile to be derived. Reliable estimates of the total 137Cs inventory can help to explain the lateral distribution. This equation was used to analyse the 137Cs-activities of Chernobyl fall-out measured on a slope under permanent pasture in Luxembourg.     

A 10-year study of the 137Cs distribution in soil and a comparison of Cs soil inventory with precipitation-determined deposition

During a 10-year period, 1988-1998, surface soil samples have been collected at Blentarp in southern Sweden and analysed for 137Cs from the Chernobyl accident and from the nuclear weapons tests. The distance between the sampling plots on the different sampling occasions has been no more than 3 m. The results show that the depth distribution of 137Cs is very similar for each of the sampling occasions, indicating that the caesium migration at this site is very small. The total activity measured in the soil cores is in agreement with the calculated activity of 137Cs deposited at the site after nuclear weapons tests and the Chernobyl accident, based on air activity concentration and the amount of precipitation. The calculated deposition of 137Cs originating from the bomb tests amounts to 1.41 kBq m-2 for the period 1962-1986, which is in agreement with the activity of nuclear weapons fallout measured in the soil samples (1.60 kBq m-2 as a mean value of the first four years of sampling). The calculated activity of 137Cs of Chernobyl origin was 0.79 kBq m-2, which agrees well with the value of 0.79 kBq m-2 measured in the soil samples in 1988.     

Global weapons' fallout 137Cs in soils and transfer to vegetation in south-central Chile

The contamination and depth distribution of 137Cs in soil due to the fallout from atmospheric weapons' tests were measured at 29 sites in the 9th and 10th administrative regions in Chile located in the 40 degrees latitude in the southern hemisphere. The depth distribution in most of the sites follows no systematic pattern in the upper few centimetres, but below this depth an exponential decline could be deduced. The calculated relaxation depth appears to be a good indicator for estimating the long-term 137Cs distribution in these soil profiles. It ranges from 4.4 +/- 1.9 cm in Palehumults to 8.4 +/- 4.4 and 9.7 +/- 5.1 cm in Hapludands and Psamments, respectively. For these soil types the value for the relaxation depth tends to increase with decreasing clay content and increasing volume of coarse pores. 137Cs activity densities at the selected sites ranged from 450 to 5410 Bq m(-2) and correlate significantly (r = 0.791) with the mean annual rainfall rate of the sampling sites. 137Cs concentration ratios of prairie plants/soil were found to be in the range 0.008-2.3 and could be related to relaxation depths in undisturbed soils.     

Cs contamination of the Techa river flood plain near the village of Muslumovo

The results of a radiometric survey of the Techa river flood plain near the village of Muslumovo in the Chelyabinsk region of Russia are presented. The observed territory extended 16.6 km along the riverbed, with a total area of 2.5 km². The collimated scintillation detector technique was applied to in situ field measurements of ¹³⁷Cs deposition on the soil. Maps of ¹³⁷Cs deposition and soil penetration depth were developed on the basis of approximately 5000 measurements. The total ¹³⁷Cs deposition within the surveyed territory has been estimated at 6.6 TBq. The means of the total ¹³⁷Cs soil depositions at half-kilometer sites on the flood plain and its distribution along the river have also been calculated. A maximum ¹³⁷Cs contamination above 7.5 MBq/m² is associated with a bank height up to 1 m above the usual water level. The data identify zones of intensive radionuclide sedimentation and transit zones.     

Changes in 137Cs concentrations in soil and vegetation on the floodplain of the Savannah River over a 30 year period

(137)Cs released during 1954-1974 from nuclear production reactors on the Savannah River Site, a US Department of Energy nuclear materials production site in South Carolina, contaminated a portion of the Savannah River floodplain known as Creek Plantation. (137)Cs activity concentrations have been measured in Creek Plantation since 1974 making it possible to calculate effective half-lives for (137)Cs in soil and vegetation and assess the spatial distribution of contaminants on the floodplain. Activity concentrations in soil and vegetation were higher near the center of the floodplain than near the edges as a result of frequent inundation coupled with the presence of low areas that trapped contaminated sediments. (137)Cs activity was highest near the soil surface, but depth related differences diminished with time as a likely result of downward diffusion or leaching. Activity concentrations in vegetation were significantly related to concentrations in soil. The plant to soil concentration ratio (dry weight) averaged 0.49 and exhibited a slight but significant tendency to decrease with time. The effective half-lives for (137)Cs in shallow (0-7.6 cm) soil and in vegetation were 14.9 (95% CI=12.5-17.3) years and 11.6 (95% CI=9.1-14.1) years, respectively, and rates of (137)Cs removal from shallow soil and vegetation did not differ significantly among sampling locations. Potential health risks on the Creek Plantation floodplain have declined more rapidly than expected on the basis of radioactive decay alone because of the relatively short effective half-life of (137)Cs.     

High accuracy in situ radiometric mapping

In situ and airborne gamma ray spectrometry have been shown to provide rapid and spatially representative estimates of environmental radioactivity across a range of landscapes. However, one of the principal limitations of this technique has been the influence of changes in the vertical distribution of the source (e.g. 137Cs) on the observed photon fluence resulting in a significant reduction in the accuracy of the in situ activity measurement. A flexible approach for single gamma photon emitting radionuclides is presented, which relies on the quantification of forward scattering (or valley region between the full energy peak and Compton edge) within the gamma ray spectrum to compensate for changes in the 137Cs vertical activity distribution. This novel in situ method lends itself to the mapping of activity concentrations in environments that exhibit systematic changes in the vertical activity distribution. The robustness of this approach has been demonstrated in a salt marsh environment on the Solway coast, SW Scotland, with both a 7.6 cm x 7.6 cm NaI(Tl) detector and a 35% n-type HPGe detector. Application to ploughed field environments has also been demonstrated using HPGe detector, including its application to the estimation of field moist bulk density and soil erosion measurement. Ongoing research work is also outlined.     

Application of in-situ measurement to determine Cs in the Swiss Alps

Establishment of ¹³⁷Cs inventories is often used to gain information on soil stability. The latter is crucial in mountain systems, where ecosystem stability is tightly connected to soil stability. In-situ measurements of ¹³⁷Cs in steep alpine environments are scarce. Most studies have been carried out in arable lands and with Germanium (Ge) detectors. Sodium Iodide (NaI) detector system is an inexpensive and easy to handle field instrument, but its validity on steep alpine environments has not been tested yet. In this study, a comparison of laboratory measurements with GeLi detector and in-situ measurements with NaI detector of ¹³⁷Cs gamma soil radiation has been done in an alpine catchment with high ¹³⁷Cs concentration (Urseren Valley, Switzerland). The aim of this study was to calibrate the in-situ NaI detector system for application on steep alpine slopes. Replicate samples from an altitudinal transect through the Urseren Valley, measured in the laboratory with a GeLi detector, showed a large variability in ¹³⁷Cs activities at a meter scale. This small-scale heterogeneity determined with the GeLi detector is smoothed out by uncollimated in-situ measurements with the NaI detector, which provides integrated estimates of ¹³⁷Cs within the field of view (3.1 m²) of each measurement. There was no dependency of ¹³⁷Cs on pH, clay content and carbon content, but a close relationship was determined between measured ¹³⁷Cs activities and soil moisture. Thus, in-situ data must be corrected for soil moisture. Close correlation (R² = 0.86, p < 0.0001) was found for ¹³⁷Cs activities (in Bq kg⁻¹) estimated with in-situ (NaI detector) and laboratory (GeLi detector) methods. We thus concluded that the NaI detector system is a suitable tool for in-situ measurements in alpine environments. This paper describes the calibration of the NaI detector system for field application under elevated ¹³⁷Cs activities originating from Chernobyl fallout.     

Uncertainty analysis of the external gamma-dose rate due to the variability of the vertical distribution of 137Cs in the soil

The external gamma-dose rate at 1 m height above a flat area due to the presence of fallout radiocesium in the soil is frequently calculated from the observed depth profile of the 137Cs activity as well as the soil mass per unit area. At a given site, these depth profiles may, however, vary considerably, thus introducing an uncertainty to the external gamma-dose calculated in this way. To assess this source of uncertainty for a typical grassland site, the activity of Chernobyl-derived 137Cs and the wet bulk density in the three upper soil layers at 100 plots in a 100 m x 100 m pasture were determined. Analysis of these data shows that the frequency distribution of the dose rates calculated from the corresponding depth profiles of all plots is similar to a log-normal distribution (mean 25 nGy h-1, median 22 nGy h-1, standard deviation 11 nGy h-1; range 1.6-56 nGy h-1). The various sources which contribute to the uncertainty of the dose rate are quantified. The semi-variogram indicates that any spatial dependence of the dose rates occurs on this pasture only over distances that are smaller than the shortest sampling interval (here about 10 m). It is estimated which errors have to be expected for the median dose rate when the depth profiles of 137Cs and of the wet bulk density are determined only for a small number of plots. It is preferable to calculate the mean dose rate as a mean from the n individual dose rates rather than from an averaged 137Cs depth profile of the n plots.     

Comparative estimations of 137Cs distribution in a boreal forest in northern Sweden using a traditional sampling approach and a portable NaI detector

Field-portable detectors have been frequently used in routine monitoring and hazard assessment studies. However, there have been few thorough attempts to evaluate their potential as an alternative to the traditional procedure of collecting samples and analysing them in the laboratory. Thus, in this study the two approaches were compared in terms of their utility for monitoring (137)Cs activity in the Ny?nget catchment in northern Sweden. The objectives were: (i) to determine the (137)Cs activity in soils associated with three types of vegetation, (ii) to map the geographical distribution of (137)Cs using the portable NaI detector connected to a GPS system (GDM-40), (iii) to identify (137)Cs anomalies in the catchment, and (iv) to compare the measurements obtained with the NaI detector and traditional sampling followed by laboratory analysis. Our results demonstrate that the GDM-40 has very good potential for making (137)Cs inventories and for detecting (137)Cs anomalies within large areas. The GDM-40 measurements identified differences between different hydrological areas that were not determined with the soil sampling method. The GDM-40 method is much faster than a traditional soil sampling method. However, soil sampling cannot be totally excluded because it is needed to calibrate the GDM-40. The agreement between the (137)Cs activity values obtained by the two approaches was 20% which is good in the field where so many factors vary.     

Vertical distribution and inventories of 137Cs in the Syrian soils of the eastern Mediterranean region

Vertical distribution and inventories of (137)Cs have been determined using radiocesium distributions in presumably undistributed soil profiles, collected from 36 sites distributed all over Syria (eastern Mediterranean region). Vertical distributions of (137)Cs in the collected profiles were found to be strongly correlated with soil type and five groups were identified. Based on these profiles, total (137)Cs inventory (bomb test and Chernobyl) varied between 320 Bq m(-2) and 9,647 Bq m(-2). Geographical mapping of (137)Cs inventories showed that the highest values were found in the coastal, middle and north-east regions of Syria indicating that Chernobyl atmospheric contribution to the total (137)Cs deposition in the region is predominant. In contrast, the lowest values were found in the south-east region (Syrian Badia), where a relatively uniform distribution was observed, which may only be attributed to the past global nuclear bomb test. The measured inventories were also compared with a mathematical model for estimating bomb derived (137)Cs reference inventories.     

Analysis of radiocaesium in the Lebanese soil one decade after the Chernobyl accident

Fallout from the Chernobyl reactor accident due to the transport of a radioactive cloud over Lebanon in the beginning of May 1986 was studied 12 years after the accident for determining the level of (137)Cs concentration in soil. Gamma spectroscopy measurements were performed by using coaxial high sensitivity HPGe detectors. More than 90 soil samples were collected from points uniformly distributed throughout the land of Lebanon in order to evaluate their radioactivity. The data obtained showed a relatively high (137)Cs activity per surface area contamination, up to 6545Bqm(-2) in the top soil layer 0-3cm. The average activity of (137)Cs in the top soil layer 0-3cm in depth was 59.7Bqkg(-1) dry soil ranging from 15 to 119Bqkg(-1) dry soil. The horizontal variability was found to be about 45% between the sampling sites. The depth distribution of total (137)Cs activity in soil showed an exponential decrease. Estimation of the annual effective dose due to external radiation from (137)Cs contaminated soil for selected sites gave values ranging from 19.3 to 91.6 micro Svy(-1).     

Numerical simulation of 137Cs and (239,240)Pu concentrations by an ocean general circulation model

We simulated the spatial distributions and the temporal variations of 137Cs and (239,240)Pu concentrations in the ocean by using the ocean general circulation model which was developed by National Center of Atmospheric Research. These nuclides are introduced into seawaters from global fallout due to atmospheric nuclear weapons tests. The distribution of radioactive deposition on the world ocean is estimated from global precipitation data and observed values of annual deposition of radionuclides at the Meteorological Research Institute in Japan and several observed points in New Zealand. Radionuclides from global fallout have been transported by advection, diffusion and scavenging, and this concentration reduces by radioactive decay in the ocean. We verified the results of the model calculations by comparing simulated values of 137Cs and (239,240)Pu in seawater with the observed values included in the Historical Artificial Radionuclides in the HAM database, which has been constructed by the Meteorological Research Institute. The vertical distributions of the calculated 137Cs concentrations were in good agreement and are in good agreement with the observed profiles in the 1960s up to 250 m, in the 1970s up to 500 m, in the 1980s up to 750 m and in the 1990s up to 750 m. However, the calculated 137Cs concentrations were underestimated compared with the observed 137Cs at the deeper layer. This may suggest other transport processes of 137Cs to deep waters. The horizontal distributions of 137Cs concentrations in surface water could be simulated. A numerical tracer release experiment was performed to explain the horizontal distribution pattern. A maximum (239,240)Pu concentration layer occurs at an intermediate depth for both observed and calculated values, which is formed by particle scavenging. The horizontal distributions of the calculated (239,240)Pu concentrations in surface water could be simulated by considering the scavenging effect.     

A simplified 137Cs transport model for estimating erosion rates in undisturbed soil

(137)Cs is an artificial radionuclide with a half-life of 30.12 years which released into the environment as a result of atmospheric testing of thermo-nuclear weapons primarily during the period of 1950s-1970s with the maximum rate of (137)Cs fallout from atmosphere in 1963. (137)Cs fallout is strongly and rapidly adsorbed by fine particles in the surface horizons of the soil, when it falls down on the ground mostly with precipitation. Its subsequent redistribution is associated with movements of the soil or sediment particles. The (137)Cs nuclide tracing technique has been used for assessment of soil losses for both undisturbed and cultivated soils. For undisturbed soils, a simple profile-shape model was developed in 1990 to describe the (137)Cs depth distribution in profile, where the maximum (137)Cs occurs in the surface horizon and it exponentially decreases with depth. The model implied that the total (137)Cs fallout amount deposited on the earth surface in 1963 and the (137)Cs profile shape has not changed with time. The model has been widely used for assessment of soil losses on undisturbed land. However, temporal variations of (137)Cs depth distribution in undisturbed soils after its deposition on the ground due to downward transport processes are not considered in the previous simple profile-shape model. Thus, the soil losses are overestimated by the model. On the base of the erosion assessment model developed by Walling, D.E., He, Q. [1999. Improved models for estimating soil erosion rates from cesium-137 measurements. Journal of Environmental Quality 28, 611-622], we discuss the (137)Cs transport process in the eroded soil profile and make some simplification to the model, develop a method to estimate the soil erosion rate more expediently. To compare the soil erosion rates calculated by the simple profile-shape model and the simple transport model, the soil losses related to different (137)Cs loss proportions of the reference inventory at the Kaixian site of the Three Gorge Region, China are estimated by the two models. The over-estimation of the soil loss by using the previous simple profile-shape model obviously increases with the time period from the sampling year to the year of 1963 and (137)Cs loss proportion of the reference inventory. As to 20-80% of (137)Cs loss proportions of the reference inventory at the Kaixian site in 2004, the annual soil loss depths estimated by the new simplified transport process model are only 57.90-56.24% of the values estimated by the previous model.     

The extremely high 137Cs inventory in the Sulu Sea: a possible mechanism

Large-volume seawater samples were collected in the Sulu and South China Seas and their (137)Cs activities were determined by gamma-ray spectrometry using a low background type high-purity Ge detector. Vertical distributions of (137)Cs activity showed an exponential decrease in the South China Sea, whereas a subsurface maximum at 200m depth and monotonic decrease below 300m were observed in the Sulu Sea. A significant difference in intermediate water (137)Cs activities in the 500-2000m depth was observed between the Sulu and South China Seas, i.e., the (137)Cs activities in the Sulu Sea were remarkably higher than those in the South China Sea. The difference in the (137)Cs inventory below 500m was approximately 1200Bqm(-2) between the Sulu and South China Seas. The (137)Cs total inventory of 3200Bqm(-2) in the Sulu Sea was 5.7 times higher than that expected from global fallout. A possible mechanism controlling this extremely high (137)Cs total inventory may be inflows of the (137)Cs rich water masses through the Luzon Strait, lateral transport across the Mindoro Strait into the Sulu Sea, and then subduction into the deep layer in the basin.     

Formation of radioactivity enriched soils in mountain areas

A field study was carried out in the Mercantour Mountains at 2200 m altitude to investigate the processes of soil enrichment in atmospheric Chernobyl (137)Cs. Soils with high (137)Cs activities have been collected in the pasture areas with frequently measured (137)Cs activity values of the order of 7000 Bq m(-2). At some single spots (about 6% of the studied area), activity in soils reached 300000 Bq m(-2), which represents 44% of the (137)Cs of the total area. Data further showed that spatial distribution of Cs depends widely on its origin: Chernobyl Cs is mainly concentrated in "enriched" soils, whereas older Cs and (241)Am fallout from nuclear weapons tests (NWTs) and natural atmospheric (210)Pb in soils is less heterogeneously distributed.In order to elucidate the processes which have led to the enrichment in Chernobyl (137)Cs in the Alps in May of 1986, we have studied the repartition of atmospheric (7)Be isotope (half-life=53.3 d) in the pasture compartments (soil, litter, grass, and snow). Snow (7)Be data give evidence that fallout enrichment is related to snow accumulation (snow drift). The transfer of beryllium occurs rapidly to the grass and litter, where the strongest pollutant accumulations were measured. However, (7)Be transport to the soil required more than 8 months.     

Long-term studies on transfer of 137Cs from soil to vegetation and to grazing lambs in a mountain area in northern Sweden

Studies were made during 1990-1997 on the transfer of 137Cs from soil to vegetation (herbage) and to grazing lambs on a mountain farm with an uncultivated grazing area of about 10 km2. The farm is situated in an area in Northern Sweden which was contaminated by the Chernobyl fallout in 1986. The mean concentration of 137Cs in the soil to a depth of 10 cm for eight sampling sites observed in the 8-year period was 14.51 kBq/m2, while in the cut herbage the average concentration was 859 Bq/kg d.w. and in lamb meat 682 Bq/kg w.w. A slow vertical migration of 137Cs in the 0-10 cm soil layer was indicated. Although the 137Cs concentration in herbage gradually decreased, the concentration in lamb meat varied from year to year. Soil ingestion by the lambs as a pathway for activity transfer was shown to be negligible, while ingestion of fungi with high concentrations of 137Cs was demonstrated to occur, as large numbers of fungi spores were counted in samples of the lambs' faeces. Fungi ingestion might therefore partly explain the varying mean yearly 137Cs concentrations in lamb muscle. The mean transfer parameters were as follows: for "soil to herbage" 61.3 Bq/kg d.w. herbage per kBq/m2 soil, for "herbage to lamb meat" 0.81 Bq/kg w.w. meat per Bq/kg d.w. herbage, and for "soil to lamb meat" 47.1 Bq/kg w.w. meat per kBq/m2 soil. A trend of decreasing values of the transfer parameter for "soil to herbage" indicated that 137Cs was becoming less available for root-uptake with time. The effective ecological half-life of 137Cs in soil, herbage and lamb meat was calculated to be 19, 7 and 16 years, respectively. It can be concluded that natural areas are vulnerable to 137Cs contamination, resulting in high concentrations in plants, fungi and lamb meat for a long time.     

Radioactivity concentrations of 40K, 134Cs, 137Cs, 90Sr, 241Am, 238Pu and 239+240Pu radionuclides in Jordanian soil samples

In November 2000, surface and core soil samples were collected from different regions of Jordan. The samples were analyzed by direct gamma spectrometry and combined radiochemical separation procedure to quantify (40)K, (134)Cs, (137)Cs, (90)Sr, (241)Am, (238)Pu and (239+240)Pu radioactivity. Concentrations (Bq.kg(-1) dry weight) have been observed to vary in the range 1.5-2.6 for (134)Cs, 2.8-11.4 for (90)Sr, and 0.13-0.48 for (241)Am, 0.016-0.062 for (238)Pu, 0.28-1.01 for (239+240)Pu and 155-543 for (40)K. The typical concentration of (137)Cs found in topsoils (0-2 cm) ranged in 7.5-576 Bq.kg(-1), dry weight. These values were greater than those observed in samples taken at greater depths (up to 32 cm). Activity ratios of (134)Cs/(137)Cs, (90)Sr/(137)Cs, (239+240)Pu/(137)Cs, (238)Pu/(137)Cs, (241)Am/(137)Cs, (239+240)Pu/(238)Pu and (241)Am /(238)Pu have mean values of 0.0049 (R=1), 0.29 (R=0.76), 0.41 (R=0.90), 0.39 (R=0.85), 0.41 (R=0.88), 7.72 (R=0.97) and 16.66 (R=0.98), respectively. The underlying concentrations were correlated and relatively higher than those reported in neighboring countries. One moss sample, as a biomonitor indicator, was measured and evaluated along with the soil samples. Its data showed higher concentrations of all measured radionuclides due to accumulations over years. The depth distribution of the fission product (137)Cs and the total deposition (Bq.m(-2)) were also studied in selected samples. Estimations of the annual effective dose equivalent due to (137)Cs-soil contamination showed values up to more than 200 microSv.     

Vertical migration studies of 137Cs from nuclear weapons fallout and the Chernobyl accident

The vertical migration of (137)Cs originating from nuclear weapons fallout (NWF) and the Chernobyl accident has been studied at 33 sampling sites in western Sweden. An attempt to describe the present depth distribution with a solution to the convection-diffusion equation (CDE) with a pulse-like fallout event as the initial condition was made. A sum of two CDEs describing the NWF and Chernobyl debris was fitted to the actual depth profiles measured by soil sampling. The fitted depth profiles were used to correct in situ measurements for the actual depth distribution, showing good agreement with the accumulated activities in soil samples. As expected, the vertical migration was very slow and most caesium was still present in the upper soil layers. The ranges of the apparent convection velocity, v, and apparent diffusion coefficient, D, were between 0 and 0.35 cm/year and 0.06 and 2.63 cm(2)/year, respectively.     

Calibration of a portable HPGe detector using MCNP code for the determination of 137Cs in soils

In situ gamma spectrometry provides a fast method to determine (137)Cs inventories in soils. To improve the accuracy of the estimates, one can use not only the information on the photopeak count rates but also on the peak to forward-scatter ratios. Before applying this procedure to field measurements, a calibration including several experimental simulations must be carried out in the laboratory. In this paper it is shown that Monte Carlo methods are a valuable tool to minimize the number of experimental measurements needed for the calibration.     

The effective and environmental half-life of 137Cs at Coral Islands at the former US nuclear test site

The United States (US) conducted nuclear weapons testing from 1946 to 1958 at Bikini and Enewetak Atolls in the northern Marshall Islands. Based on previous detailed dose assessments for Bikini, Enewetak, Rongelap, and Utirik Atolls over a period of 28 years, cesium-137 (137Cs) at Bikini Atoll contributes about 85-89% of the total estimated dose through the terrestrial food chain as a result of uptake of 137Cs by food crops. The estimated integral 30, 50, and 70-year doses were based on the radiological decay of 137Cs (30-year half-life) and other radionuclides. However, there is a continuing inventory of 137Cs and 90Sr in the fresh water portion of the groundwater at all contaminated atolls even though the turnover rate of the fresh groundwater is about 5 years. This is evidence that a portion of the soluble fraction of 137Cs and 90Sr inventory in the soil is lost by transport to groundwater when rainfall is heavy enough to cause recharge of the lens, resulting in loss of 137Cs from the soil column and root zone of the plants. This loss is in addition to that caused by radioactive decay. The effective rate of loss was determined by two methods: (1) indirectly, from time-dependent studies of the 137Cs concentration in leaves of Pisonia grandis, Guettarda specosia, Tournefortia argentea (also called Messerschmidia), Scaevola taccada, and fruit from Pandanus and coconut trees (Cocos nucifera L.), and (2) more directly, by evaluating the 137Cs/90Sr ratios at Bikini Atoll. The mean (and its lower and upper 95% confidence limits) for effective half-life and for environmental-loss half-life (ELH) based on all the trees studied on Rongelap, Bikini, and Enewetak Atolls are 8.5 years (8.0 years, 9.8 years), and 12 years (11 years, 15 years), respectively. The ELH based on the 137Cs/90Sr ratios in soil in 1987 relative to the 137Cs/90Sr ratios at the time of deposition in 1954 is less than 17 years. The magnitude of the decrease below 17 years depends on the ELH for 90Sr that is currently unknown, but some loss of 90Sr does occur along with 137Cs. If the 15-year upper 95% confidence limit on ELH (corresponding to an effective half-life of 9.8 years) is incorporated into dose calculations projected over periods of 30, 50, or 70 years, then corresponding integral doses are 58, 46 and 41%, respectively, of those previously calculated based solely on radiological decay of 137Cs.