Deer Meat Contaminated With Cs-137 at Brookhaven National Laboratory
Prepared by the Bureau of Environmental Radiation Protection
Table of Contents
Because of soil contamination at Brookhaven National Laboratory (BNL), some of the deer killed on the site roadways were found to have elevated levels of Cs-137. BNL does not allow hunting on its site. However, as the site property is not fenced, this finding raised concern over the potential radiation doses to those who eat venison from deer harvested on or near the BNL site. As a result, NY State Department of Environmental Conservation (DEC) initiated a program with the objective of harvesting 20 deer from an area adjacent to BNL, where deer moving off-site can be legally hunted during season. As part of this program, meat samples were collected from seven deer between August 18 and October 12, 1998. In addition, BNL obtained samples from two more deer killed in the vicinity of the laboratory in December 1998. All vicinity deer samples show elevated concentrations of Cs-137, when compared to deer harvested from areas on Long Island which are too far to be affected by contamination on the BNL site.
Because of the small number of deer samples collected from the area that surrounds BNL, all available results of meat samples from deer killed on-site were combined with results for vicinity deer samples. The combined data set includes 32 deer killed between 1992 and 1999. This data set was used to calculate dose projections from one year of consumption of contaminated deer meat by the average hunter and under reasonable maximum exposure (RME) assumptions. The projected doses were estimated to be 5 mrem for the average hunter and 9 mrem for RME conditions. Both projected doses are below the DEC 10 mrem guidance value for one year of exposure.
It is recommended that the public, particularly hunters, be advised of the presence of elevated Cs-137 concentrations in deer close to the BNL site. However, since the projected doses are below the existing guideline, it is concluded that no additional formal restrictions need to be placed on hunting in the areas where it is not currently prohibited, nor on consumption of deer harvested from these areas.
The 1996 Site Environmental Report which was issued by the Brookhaven National Laboratory (BNL) in 1998, provides data on Cs-137 concentration measured in deer meat samples for deer taken on-site and off-site [BNL 98a]. As a result, STAR (Standing for Truth About Radiation Foundation) petitioned the New York State Department of Environmental Conservation (DEC) to implement a formal hunting ban on BNL property and issue a consumption advisory for deer from the area around BNL. The petition also requested that, in the alternative, DEC limit the number of permits granted for the area around the laboratory, and, if hunting is to continue, DEC sample the local deer population for radionuclides.
DEC forwarded the petition to the Bureau of Environmental Radiation Protection (BERP). New York State Department of Health (DOH) and asked for an opinion on the health risk posed by the observed levels of Cs-137 to individuals consuming meat of deer taken from the BNL area. The following analysis was developed in response to DEC'S request.
The risk to an individual from the ingestion of meat containing Cs-137 depends upon the radiation dose to that individual that results from the Cs-137 intake. The radiation dose to an individual from the ingestion of a given amount of Cs-137 depends upon a number of factors. These factors include the total body and skeletal muscle masses, the amount of Cs-137 ingested, its uptake, distribution and elimination rate from the body. While these factors vary from one individual to another, methods which can be used for radiation protection purposes have been developed for estimating the dose from unit intake of a specific radionuclide . The methods developed by the International Commission on Radiological Protection (ICRP) have been used in developing the radiation protection criteria used by the United States and NY State agencies. These methods are used in this analysis.
Existing criteria for not allowing the sale of meat products contaminated with radioactive materials only address situations where contamination results from a major nuclear accident, such as that which occurred in some European countries from the 1986 Chernobyl reactor accident. (The limit for Cs-137, derived from the Protective Action Guide, adopted by USFDA in 1998 for such emergencies is 32 pCi/g (1,200 Bq/kg) [FDA 98].) There are no meat-specific criteria for cases similar to that in BNL. On the other hand, criteria have been set which address doses to the public from all exposure pathways to residual contamination, including that of consuming meat of animals that graze on contaminated lands. These latter criteria are more applicable to the BNL situation.
At the national level, the US Nuclear Regulatory Commission (NRC) adopted regulations which set a limit of 25 mrem/year for the unrestricted release of sites contaminated with radioactive materials [NRC 97]. The US Environmental Protection Agency (EPA) published draft regulations proposing a limit of 15 mrem/year from all pathways, with the ground water contamination limited to values not to exceed the Safe Drinking Water Act Maximum Contaminant levels (MCLs) [EPA 94]. New York State has not adopted similar regulations yet. However, DEC has issued a Technical Administrative Guidance Memorandum (TAGM) in which a limit of 10 mrem/year is used [DEC 93]. The 10 mrem value, which is the most protective of these values, will be used to evaluate the need for placing hunting restrictions or issuing advisories relating to the consumption of deer meat harvested from the BNL vicinity.
The dose to a person eating deer meat containing Cs-137 can be estimated using the expression:
H = W x C x DCF
Radioactive materials which are taken up by the different organs and tissues deliver radiation dose to the body as they decay over time. The committed effective dose equivalent H, measures the total dose equivalent that accumulates over a period of 50 years following intake of the radionuclides.
Almost all of the cesium ingested in the form of soluble compounds is absorbed into the body from the gastrointestinal tract [ICRP 89]. Measurements of Cs-137 uptake in 10 volunteers following the consumption of venison contaminated as a result of the Chernobyl accident gave a mean absorption of 78% of the ingested cesium (Henrichs, as reported in ICRP #56) [ICRP 89]. However, because of the paucity of data on the uptake of cesium incorporated in food, the ICRP dose projections assume that 100% of the ingested cesium is absorbed.
Cesium entering the body distributes in soft body tissues. Skeletal muscles have been reported to have higher concentrations than other tissues, but the differences are not large. Cesium is eliminated from the body mostly through urine, following a retention function of the form:
R(t) = a e^(-0.693 t /T1) + (1 – a) e^(-0.693 t /T2)
Where R(t) is the fraction of the originally absorbed Cs-137 that remains in the body at time t after ingestion.
t is the time following ingestion measured in days,
a is the fraction of ingested cesium which is eliminated with a half life equal to T1 days,
1-a is the fraction of ingested cesium which is eliminated with a half life equal to T2 days.
Observations have shown a to range from about 0.05 to 0.15, T1 is about 1 or 2 days, and T2 ranges from 50 to 200 days. In order to calculate projected doses from Cs for radiation protection purposes, the ICRP adopted the values a = 0.1, T1 = 2 days and T2 = 110 days in the retention equation. Since cesium is eliminated from the body with a half life of the order of 100 days, 99% of the radiation dose is received by the body within the first two years after the ingestion of the material.
The annual intake of Cs-137 is determined by the amount of venison consumed and the concentration of Cs-137 in the meat. The amount of venison an individual eats per year varies widely from one individual to another in the community. While the majority of the population eats no or little deer meat, venison may form a large portion of the meat in some hunters' diet. Thus, the hunters and members of their families are expected to receive the highest dose equivalent in the population. The average annual Cs-137 intake by a member of this group will be used to estimate the radiation dose. The data reported in Exposure Factors Handbook issued by EPA in 1996 show that 0.96% of the population in the Northeastern US consumes game meat and that the mean intake is 1.13 +/- 0.22 gram per day per kilogram of body weight (g/kg/day) [EPA 96]. For a 70 kg individual, this gives an annual intake of 28.9 +/- 5.5 kg/year (63.5 +/- 12.1 lbs/year). A value of 29 kg/year is used in this analysis. This corresponds to about two 10 ounce meals per week.
About 60% of radioactive cesium taken up by deer, as a result of grazing on contaminated lands, is retained and is distributed throughout the deer body. The highest concentrations are found in muscle, which is the normally edible portion of the deer. Cesium concentration in the liver, which is also often consumed, is about 30% of the concentration in muscle [Coughtrey 83].
Cesium is eliminated from deer with a half life of about 2-3 weeks. The elimination half life varies slowly with the mass of the deer ranging from 16 days for a 100 lb deer to 20 days for a 200 lb animal [Coughtrey 83]. As a result, the amount of cesium in a deer ingesting a fixed daily amount reaches an equilibrium value in about 100 days from the start of intake. At equilibrium, the amount of cesium entering the body is equal to the amount that is eliminated. The total activity in the animal at equilibrium is about 25 times the daily intake of cesium. Also, if a deer stops grazing in contaminated areas, the amount of cesium in the deer decreases to 10% of its peak value after about two months.
Tables 1, 2 and 3 summarize all available data on K-40 and Cs-137 concentrations in deer meat for samples obtained from deer killed on BNL property, areas adjacent to BNL and areas away from BNL, respectively. Results of analysis of deer meat samples taken from deer killed on-site roads were included in BNL's Site Environmental Reports for 1992, 1993 and 1996. The 1992 Report gives the Cs-137 concentrations in flesh and liver samples of one deer killed on-site [BNL 93]. These concentrations are given as 6.15 pCi/g and 1.12 pCi/g in the flesh and liver respectively. The 1993 BNL Site Environmental Report indicates that there were additional deer meat samples collected in 1992, but analyzed in 1993 [BNL 94]. It gives a range of 0.28 - 6.15 pCi/g for the on-site samples, and 0.18 - 0.25 pCi/g for samples collected from the Ridge/Calverton areas. No specific information is given in the 1993 report on the numbers of samples nor on individual sample concentrations. The 1996 report gives data on Cs-137 concentration in 5 flesh and 5 liver deer samples collected from the site and on 3 flesh and 6 liver samples from deer killed off-site [BNL 98a]. The off-site samples are collected from areas many miles away from BNL and are not likely to include deer that have grazed on the BNL site. (Deer foraging range is expected to be within an area of about I mile radius, unless unusual circumstances drive the deer to migrate to greater distances [Lowery 98].) The concentrations in the 1996 on-site meat samples range from 1.01 to 11.74 pCi/g in meat and from 0.65 to 3.36 pCi/g in liver. The average concentrations in the 1996 meat and liver samples taken on-site are 6.03 +/- 3.85 pCi/g and 2.04 +/- 1.02 pCi/g, respectively. In addition to the data published in the Site Environmental Reports of 1992. 1993 and 1996, BNL provided data on samples taken from 11 deer in 1997 and 1998, 10 on-site and one off-site [BNL 98b]. The 1997 on-site samples include flesh and liver samples from 6 deer. The Cs-137 concentrations in these samples range from 0.19 to 6.04 pCi/g in flesh and from 0.03 to 3.73 pCi/g in the liver. The average concentrations for the on-site samples in 1997 are 1.87 +/- 2.20 pCi/g for flesh and 0.84 +/- 1.43 pCi/g for liver. The one off-site sample analyzed in 1997 was obtained from an area about 1/2 mile SW of Shultz Road and Wading River Manor Road. The concentration in this sample is measured at 4.71 +/- 0.80 pCi/g in flesh.
BNL analyzed meat and liver samples taken from four on-site deer during January through April 1998. These included four meat and three liver samples. The Cs-137 concentration was below the detection limit in two of the meat samples. The other two meat samples had measured concentrations of 0.24 0.05 pCi/g and 0.24 0.15 pCi/g, respectively. All three liver samples showed Cs concentrations that are below the detection limit.
The on-site deer are likely to have grazed on contaminated lands, and therefore are expected to have elevated Cs-137 concentrations. Deer harvested from areas close to BNL could have spent some time grazing within the contaminated areas on-site. Therefore, while hunting is not allowed on the BNL site, it is anticipated that some of the deer hunted in areas near BNL may have elevated Cs-137 concentrations. In order to obtain a sample more representative of potentially contaminated deer legally accessible to hunters, additional samples were collected by DEC in the fall of 1998 from a zone which falls within about one mile from the BNL boundary. Meat samples from some of these deer were sent to both BNL and DOH laboratories for analysis. While a target of 20 deer meat samples was originally set, by the end of December 1998, samples from only seven deer harvested from this zone were provided to DOH and BNL laboratories for analysis. Samples from four out of the seven deer were analyzed by both DOH and BNL, two only by DOH and one only by BNL. Results obtained by DOH and BNL for the samples from the four deer analyzed by both laboratories are compared in Figure 2. In some cases, multiple samples from the same deer were analyzed by the same laboratory. In these cases, the concentrations shown in Table 4 are the calculated weighted average concentration for each deer. The Cs-137 concentrations measured in these samples ranged from 1.13 to 5.53 pCi/g. The average concentration for the seven deer is 2.18 +/- 1.49 pCi/g. The results of all deer meat samples are given in Tables 1, 2 and 3, where Cs-137 concentrations are listed for on-site, BNL vicinity and away-from-site deer samples, respectively.
Additional meat analysis results were provided by BNL which included analysis of samples collected between 1997 and 1999 [BNL 99]. The new data included one 1997 and two 1998 site vicinity deer, four on-site deer (two in 1998 and two in 1999) and two away from the site deer. The results of these samples are included in Tables 1, 2 and 3. The measurement results for vicinity and on-site deer are combined in Table 4 and shown in Figure 1. Deer numbers 1-11 are vicinity deer, while deer numbers 12-31 are on-site. Where more than one meat sample was analyzed from the same deer, the weighted arithmetic average was used in obtaining the concentrations listed in Table 4.
Radiation dose projections are given below using two approaches the dose to the average individual of the hunters' group, and a more conservative approach in which the dose is estimated for the reasonable maximum exposure (RME) conditions.
A. Average Dose Estimate
The committed effective dose equivalent H from one year's intake of contaminated venison is calculated using the equation given in Section III. As discussed above, the average amount of deer meat consumed in one year by hunters or members of their families is assumed to be 29 kg/y (64 Ibs/yr). The average concentration of Cs-137 in the BNL vicinity deer meat samples is 3.13 +/- 2.15 pCi/g (Table 4). The value of the dose conversion factor for the ingestion of Cs is given in EPA's Federal Guidance Report No. 11 as 5E-5 mrem/pCi [EPA 88]. These values are used to estimate the effective dose equivalent H(Av) to the average individual from one year consumption of contaminated deer meat.
H(Av) = 3.13 x 29,000 x 5 x 1E-5 mrem = 5 mrem
While the calculated dose is about 1/2 of the dose limit criterion used for this evaluation (Section II), the number of deer samples used to estimate the dose is too small to give sufficient assurance that the dose to the average hunter will not exceed the 10 mrem criterion (the sample size needed to perform statistical tests on the data is discussed in Section A below).
A larger set of data is obtained if the samples taken on-site are combined with vicinity deer samples to give a data set that is assumed to be representative of the pool of deer available to hunters. Combining all available meat analysis results gives a set of 32 deer samples in which the Cs-137 concentration varies between levels below the detection limit (assumed to be equal to 0.1 pCi/g for calculation purposes) and 11.74 pCi/g (Table 4).
The overall average concentration for this set is 3.38 +/- 3.10 pCi/g. Using this average value gives the projected committed effective dose equivalent to the average individual from one year of deer meat consumption to be
H(Av) = 3.38 x 29,000 x 5E-5 = 5 mrem
B. Dose Estimate Based on Reasonable Maximum Exposure Considerations
Since the radiation dose is proportional to the Cs-137 concentration in the meat, the dose calculated above using the average measured concentration represents the best estimate of the dose to the average individual. However, because the purpose of this analysis is to determine if actions need to be taken to reduce the likelihood of individuals receiving dose in excess of the limiting criterion, more conservative estimates are also considered.
USEPA's report " Risk Assessment Guidance for Superfund " takes a conservative approach by requiring that reasonable maximum exposure (RME) estimates be developed [EPA 89]. The EPA procedures call for estimating the reasonable maximum exposure by using the 95% upper confidence limit on the arithmetic average for variables used in the calculation. For the set of 32 samples, this corresponds to a Cs-137 concentration of 4.28 pCi/g. (For a set of n = 32 data points, the 95% upper confidence limit of the arithmetic mean is equal to the arithmetic mean + 1.65 x standard error (se), where se = std deviation / SQRT(n).) Using this value, together with the average value for deer meat consumption, gives an estimate for the RME dose of
H(RME) = 4.28 x 29,000 x 5E-5 = 6 mrem
A more conservative estimate may also be calculated using the 95% upper confidence limits of both the average concentration of Cs-137 and the deer meat consumption values. This gives the dose estimate as
H = 4.28 x 39,900 x 5E-5 = 9 mrem
The health effect of concern from exposure to low levels of radiation dose is the potential increase of cancer risk. The risk from a radiation induced cancer depends on the radiation dose received. The risk factor has been estimated from the review of epidemiologic studies of human populations that received large radiation doses, including the Japanese bomb survivors and others who received radiation therapy for the treatment of cancer of benign diseases. Risk factors are estimated for low doses by extrapolation from high dose information on the basis of linear, no-threshold assumptions. A lifetime mortality risk factor for a population exposed to radiation equal to 5E-7 per mrem was derived by the ICRP (ICRP Report #60) [ICRP 91]. USEPA uses a mortality risk factor of 5.09E-7 per mrem, and a lifetime cancer incidence risk factor of 7.6E-7 per mrem [EPA 94]. Since the risk is assumed to be proportional to the dose, the dose to the average individual in a population gives the best estimate of the risk to the population. Using the average concentration of Cs-137 and average annual consumption of meat, together with a risk factor of 5E-7 per millirem, one obtains a lifetime cancer mortality risk of about 2E-6 from eating contaminated venison for one year. The corresponding cancer incidence risk is about 4E-6.
The dose estimates given above show that both the calculated average and RME doses are below the 10 mrem criterion. However, because of uncertainty, statistical tests are needed to determine if the measurement results support a conclusion that the deer population meets the criterion for allowing hunting outside the BNL boundary with no additional restrictions and to estimate the probability that such a decision is likely to be made in error.
The set of the 32 sample results does not follow a symmetric distribution or a log-normal distribution (Figure 3), and is analyzed using nonparametric statistical methods. The analysis seeks to answer the two points raised in the previous paragraph. The methods used in this analysis are those described in US NRC's report NUREG-1505, Rev. I [NRC 98].
In the following, we test the hypothesis that the deer population in question has an average Cs-137 concentration greater than a derived concentration guideline level (DCGL) (the null hypothesis). The DCGL is the level which leads to a dose of 10 mrem from one year consumption of deer meat by the average individual in the hunters group. This hypothesis can be rejected (i.e. making the conclusion that no additional restrictions are required) only if the measurement data provide sufficient evidence that the average concentration is less than the DCGL.
There are two situations which will lead to a decision not to reject the null hypothesis:
(1) the null hypothesis is true; and
A decision error is made if it is concluded that the criterion for not imposing additional restrictions is met when the true condition is the opposite. In this case, the null hypothesis is rejected when it is actually true. This type of decision error is referred to as Type I decision error. A Type II decision error occurs when the null hypothesis is not rejected when it is actually false.
While Cs-137 is present in the environment as a result of nuclear weapons testing, the concentrations in deer meat away from BNL, as reported by BNL (Table 3), are much smaller than those observed in deer taken from, or near, BNL. If it is assumed that all Cs-137 contamination found in BNL deer is due to the contamination on the site, there will be no need for a matched sample to determine background values, and the one set of measured concentrations can be used to perform the statistical tests. The Sign statistical test, which applies in this case, is used in this evaluation.
The following quantities are used in the statistical test:
The value of DCGL for Cs-137 in deer meat that corresponds to an effective dose equivalent of 10 mrem to the average individual is equal to 6.9 pCi/g. A value of alpha = 0.01 will be used in this evaluation. In this case, there will be a 99% probability that a decision to place additional restrictions will be indicated by the data when such restrictions are actually warranted.
A. Test for Adequacy of Sample Size
Vicinity deer data set
Vicinity and on-site deer data set
The N values are obtained from Table A.2a, NUREG-1505, using the calculated values for delta/sigma and assuming alpha = 0.01 and beta = 0.10. Therefore, the vicinity deer meat sample set which has 11 measurements, is not sufficient to perform the Sign test. However, the combined data set of vicinity and on-site samples, which has 32 deer is adequate to perform this statistical test.
B. Testing the Null Hypothesis
The null hypothesis is tested by the Sign test. To apply this test, the number of measurements S+ for which the concentration is less that the value of DCGL is calculated for the data sample being tested. The value of S+ is then compared to a critical value k which depends upon both the sample size. n. and the probability of the Type I Error, a. The null hypothesis is rejected if S+ is greater than the critical value, k. The values for k are given in Table A.3 of NUREG-1505. For n = 32 and = 0.01, k= 23.
For the deer samples in the combined data set, 28 of 32 have concentrations that are less than 6.9 pCi/g. Therefore, S+ is equal to 28. Since S+ is greater than 23 for this data set, the null hypothesis is rejected. This indicates that the existing data do not support the need for placing additional restrictions on hunting near BNL.
Cs-137 occurs in the environment as a result of fallout from the atmospheric nuclear weapon tests that were carried out in the 1950s and 1960s. Measurable levels of Cs that are attributed to fallout can still be observed in deer meat. BNL's 1993 and 1996 Site Environmental Reports report on Cs-137 concentrations measured in a number of deer meat samples taken from areas in Suffolk County away from BNL. The reported concentrations in these samples range from values below the detection level up to 0.90 +/- 0.16 pCi/g.
The Cs-137 concentrations in the majority of samples taken from deer killed on-site or from areas close to BNL exceed the values attributed to Cs-137 from fallout. This indicates uptake of Cs by deer grazing on contaminated areas on-site. While hunting is not allowed on the BNL property, the site is not fenced. Hunting is allowed in areas surrounding the site that are within the normal range for on-site deer. Samples from the vicinity of BNL analyzed in 1997 and 1998, confirm the presence of elevated Cs-137 concentrations in deer meat, and identify an exposure pathway to hunters and their families from contamination at BNL. The public health impact of this contamination was evaluated using results of Cs-137 concentration measurements in 32 deer samples taken from areas on-site as well as areas close to the BNL site. The projected average dose associated with the consumption of contaminated deer meat is estimated to be about 5 mrem for one year of intake. This value is below the levels for unrestricted release of contaminated sites which have been set by regulatory authorities in the US and in NY State. Therefore, actions by NYS to prohibit or restrict hunting in the areas where it is currently not prohibited would not be indicated by the existing data. Also, the issuance of consumption advisories in this case would imply a need for restrictions which are not indicated by the existing criteria.
Even though the data show that there is no need for formally restricting hunting in the area surrounding BNL, nor for the issuance of deer meat consumption advisories, the public should be informed of the fact that deer harvested from the BNL vicinity tend to have elevated Cs-137 concentrations. Hunters who do not wish to expose themselves or their families to the incremental radiation dose that may occur, can choose to hunt other areas away from BNL.
The analysis and discussion presented above can be used to address the different possible actions suggested by STAR. These are discussed below:
Last Modified: February 10, 2009