Index > 3 Characterisation of radioactively contaminated sites >

3.3.6 Establishing derived concentration guideline levels DCGLs

Contents
3.3.6.1 Introduction
3.3.6.2 Direct application of DCGLs
3.3.6.3 DCGLs and key-nuclides to assess the radioactivity of difficult to measure nuclides
3.3.6.4 Use of DCGLs for sites with multiple radionuclides
3.3.6.5 Integrated surface and soil contamination DCGLs

3.3.6.1 Introduction

The information gained from laboratory analyses is essential in establishing and applying the DCGLs for a site. DCGLs provide the goal for essentially all aspects of designing, implementing, and evaluating the final status survey. The DCGLs discussed in this manual are limited to soil contamination (and structure surfaces); the user should consult the responsible regulatory agency(ies) if it is necessary to establish DCGLs for other environmental media (e.g., ground water, and other water pathways). This section contains information regarding the selection and application of DCGLs.

The development of DCGLs is often an iterative process, where the DCGLs selected or developed early in the Radiation Survey and Site Investigation (RSSI) Process are modified as additional site-specific information is obtained from subsequent surveys. One example of the iterative nature of DCGLs is the development of final clean-up levels, and soil screening levels1 (SSLs) are selected or developed at a point early in the process, usually corresponding to the scoping survey in EURSSEM [EPA-1996], [EPA-1996a]. A soil screening level can be further developed, based on site-specific information, to become a preliminary remediation goal, usually at a point corresponding to the characterization survey. If the preliminary remediation goal is found to be acceptable during the characterization survey, it is documented as the final clean-up level in the record of decision (ROD) for the site. The record of decision is typically in place prior to any remedial action, because the remedy is also documented in the record of decision.

3.3.6.2 Direct application of DCGLs

In the simplest case, the DCGLs may be applied directly to survey data to demonstrate compliance. This involves assessing the surface activity levels and volumetric concentrations of radio-nuclides and comparing measured values to the appropriate DCGL.

Example 3.3: Direct application of a DCGL

Consider a site that used only one radionuclide, such as 90Sr throughout its operational lifetime. The default DCGL for 90Sr in soil and on building surfaces may be obtained from the responsible agency. Survey measurements and samples are then compared to the volume activity and the surface concentration DCGLs for 90Sr directly to demonstrate compliance. While seemingly straightforward, this approach is not always possible (e.g., when more than one radionuclide is present).

3.3.6.3 DCGLs and key-nuclides to assess the radioactivity of difficult to measure nuclides

For sites with multiple contaminants, it may be possible to measure just one of the contaminants and still demonstrate compliance for all of the contaminants present through the use of key-nuclide measurements. Both time and resources can be saved if the analysis of one radionuclide is simpler than the analysis of the other.

Example 3.4: Application of a DCGL in the case of a key-nuclide

Using measured 137Cs (= key-nuclide) concentration as a surrogate for 90Sr (= difficult to measure nuclide) reduces the analytical costs because wet chemistry separations do not have to be performed for 90Sr on every sample.

In using one key-radionuclide to measure the presence of others, a sufficient number of measurements, spatially separated throughout the survey unit, should be made to establish a “consistent” ratio. The number of measurements needed to determine the ratio is selected using the data quality objectives (DQO) process and based on the chemical, physical, and radiological characteristics of the nuclides and the site. If consistent radio-nuclide ratios cannot be determined during the historical site assessment (HSA) based on existing information, EURSSEM recommends that one of the objectives of scoping or characterization be a determination of the ratios rather than attempting to determine ratios based on the final status survey. If the ratios are determined using final status survey data, EURSSEM recommends that at least 10% of the measurements (both direct measurements and samples) include analyses for all radio-nuclides of concern.

In the use of key-nuclides, it is often difficult to establish a “consistent” ratio between two or more radio-nuclides. Rather than follow prescriptive guidance on acceptable levels of variability for the “key-nuclide – difficult to measure nuclide” ratio, a more reasonable approach may be to review the data collected to establish the ratio and to use the DQO process to select an appropriate ratio from that data.

Example 3.5: Illustration of the application of key-nuclide measurements

Ten soil samples within the survey unit were collected and analyzed for 137Cs (= key-nuclide) and 90Sr (= difficult to measure nuclide) to establish a ratio.
The ratios of 90Sr to 137Cs were as follows: 6.6, 5.7, 4.2, 7.9, 3.0, 3.8, 4.1, 4.6, 2.4, and 3.3. An assessment of this example data set results in an average 90Sr to 137Cs ratio of 4.6, with a standard deviation of 1.7. There are various approaches that may be used to develop a ratio from this data – but each must consider the variability and level of uncertainty in the data. One may consider the variability in the ratio by selecting the 95% upper bound of the ratio (to yield a conservative value of 90Sr from the measured 137Cs), which is 8.0 in this case. Similarly, one may select the most conservative value from the data set (7.9).

The DQO process should be used to assess the use of key-nuclides. The benefit of using this approach is the reduced cost of not having to perform costly wet chemistry analyses on each sample. This benefit should be considered relative to the difficulty in establishing the surrogate ratio, as well as the potential consequence of unnecessary investigations that result from the error in using a “conservative” “key-nuclide – difficult to measure nuclide” ratio. Selecting a conservative “key-nuclide – difficult to measure nuclide” ratio ensures that potential exposures from individual radio-nuclides are not underestimated. The nuclide method can only be used with confidence when dealing with the same media in the same surroundings – for example, soil samples with similar physical and geological characteristics. The EURSSEM user will need to consult with the responsible regulatory agency for concurrence on the approach used to determine the surrogate ratio.

Once an appropriate “key-nuclide – difficult to measure nuclide” ratio is determined, one needs to consider how compliance will be demonstrated using key-nuclide measurements. That is, the user must modify the DCGL of the measured radionuclide to account for the inferred radionuclide. Continuing with the above example, the modified DCGL for 137Cs must be reduced according to the following equation:

DCGLCs,mod = DCGLCs × (DCGLSr) / ((CSr/CCs) × DCGLCs + DCGLSr) ……………………… (3-1)

where CSr/CCs is the surrogate ratio of 90Sr to 137Cs.

Assuming that the DCGLSr is 15 Bq/kg, the DCGLCs is 10 Bq/kg, and the surrogate ratio is 8 (as derived previously), the modified DCGL for 137Cs (DCGLCs,mod) can be calculated using Equation 3-1:

DCGLCs,mod = 10 × ( 15 / (8 × 10 + 15)) = 1.6 Bq/kg

This modified DCGL is then used for survey design purposes.

The potential for shifts or variations in the radionuclide ratios means that the key-nuclide method should be used with caution. Physical or chemical differences between the radio-nuclides may produce different migration rates, causing the radio-nuclides to separate and changing the radio-nuclide ratios. Remediation activities have a reasonable potential to alter the “key-nuclide – difficult to measure nuclide” ratio established prior to remediation. EURSSEM recommends that when the ratio is established prior to remediation, additional post-remediation samples should be collected to ensure that the data used to establish the ratio are still appropriate and representative of the existing site condition. If these additional post-remediation samples are not consistent with the pre-remediation data, surrogate ratios should be re-established.

Compliance with surface activity DCGLs for radio-nuclides of a decay series (e.g., thorium and uranium) that emit both alpha and beta radiation may be demonstrated by assessing alpha, beta, or both radiations. However, relying on the use of alpha surface contamination measurements often proves problematic due to the highly variable level of alpha attenuation by rough, porous, and dusty surfaces. Beta measurements typically provide a more accurate assessment of thorium and uranium contamination on most building surfaces because surface conditions cause significantly less attenuation of beta particles than alpha particles. Beta measurements, therefore, may provide a more accurate determination of surface activity than alpha measurements.

The relationship of beta and alpha emissions from decay chains or various enrichments of uranium should be considered when determining the surface activity for comparison with the DCGLW values. When the initial member of a decay chain has a long half-life, the radioactivity associated with the subsequent members of the series will increase at a rate determined by the individual half-lives until all members of the decay chain are present at activity levels equal to the activity of the parent. This condition is known as secular equilibrium.
equilibrium.

Example 3.6: Calculation of the surface activity

Consider that the average surface activity DCGLW for natural thorium is 1,000 Bq/m2 (600 dpm/100 cm2), and all of the progeny are in secular equilibrium – that is, for each disintegration of 232Th there are six alpha and four beta particles emitted in the thorium decay series. Note that in this example, the surface activity DCGLW of 1,000 Bq/m2 is assumed to apply to the total activity from all members of the decay chain. In this situation, the corresponding alpha activity DCGLW should be adjusted to 600 Bq/m2 (360 dpm/100 cm2), and the corresponding beta activity DCGLW to 400 Bq/m2 (240 dpm/100 cm2), in order to be equivalent to 1,000 Bq/m2 of natural thorium surface activity. For a surface activity DCGLW of 1,000 Bq/m2, the beta activity DCGLW is calculated as follows:

1,000 Bq of chain)/m2 × 4 β /(dis of 232Th)
-———————————————————————— = 400 β Bq/m2 ……………………..(3-2)
(10 Bq of chain) / (1 Bq of 232Th)

To demonstrate compliance with the beta activity DCGLW for this example, beta measurements (in cpm) must be converted to activity using a weighted beta efficiency that accounts for the energy and yield of each beta particle. For decay chains that have not achieved secular equilibrium, the relative activities between the different members of the decay chain can be determined as previously discussed for “key-nuclide – difficult to measure nuclide” ratios.

Another example for the use of key-nuclides involves the measurement of exposure rates, rather than surface or volume activity concentrations, for radio-nuclides that deliver the majority of their dose through the direct radiation pathway. That is, instead of demonstrating compliance with soil or surface contamination DCGLs derived from the direct radiation pathway, compliance is demonstrated by direct measurement of exposure rates. To implement this key-nuclide method, historical site assessment (HSA) documentation should provide reasonable assurance that no radioactive materials are buried at the site and that radioactive materials have not seeped into the soil or groundwater. This key-nuclide approach may still be possible for sites that contain radio-nuclides that do not deliver the majority of their dose through the direct radiation pathway. This requires that a consistent relative ratio for the radio-nuclides that do deliver the majority of their dose through the direct radiation pathway can be established. The appropriate exposure rate limit in this case accounts for the radio-nuclide(s) that do not deliver the majority of their dose to the direct radiation pathway. This is accomplished by determining the fraction of the total activity represented by radio-nuclide(s) that do deliver the majority of their dose through the direct radiation pathway, and weighting the exposure rate limit by this fraction.

Note: That the considerations for establishing consistent relative ratios discussed above apply to this key-nuclide approach as well. The responsible regulatory agency should be consulted prior to implementing this “key-nuclide – difficult to measure nuclide” approach.

3.3.6.4 Use of DCGLs for sites with multiple radionuclides

Typically, each radionuclide DCGL corresponds to the release criterion (e.g., regulatory limit in terms of dose or risk). However, in the presence of multiple radio-nuclides, the total of the DCGLs for all radio-nuclides would exceed the release criterion. In this case, the individual DCGLs need to be adjusted to account for the presence of multiple radio-nuclides contributing to the total dose. One method for adjusting the DCGLs is to modify the assumptions made during exposure pathway modelling to account for multiple radio-nuclides. The key-nuclide measurements discussed in the previous section describe another method for adjusting the DCGL to account for multiple radio-nuclides. Other methods include the use of the unity rule and development of a gross activity DCGL for surface activity to adjust the individual radionuclide DCGLs.

The unity rule, represented in the expression below, is satisfied when radio-nuclide mixtures yield a combined fractional concentration limit that is less than or equal to one:

C1/DCGL1 + C2/DCGL1 + … + Cn/DCGLn < 1_ ……………………………………………….. (3-3)

where:
C = concentration
DCGL = guideline value for each individual radionuclide (1, 2, …, n)

For sites that have a number of significant radio-nuclides, a higher sensitivity will be needed in the measurement methods as the values of C become smaller. Also, this is likely to affect statistical testing considerations – specifically by increasing the numbers of data points necessary for statistical tests.

3.3.6.5 Integrated surface and soil contamination DCGLs

Surface contamination DCGLs apply to the total of fixed plus removable surface activity. For cases where the surface contamination is due entirely to one radionuclide, the DCGL for that radionuclide is used for comparison to measurement data (Section 3.3.6.2).
For situations where multiple radionuclides with their own DCGLs are present, a gross activity DCGL can be developed. This approach enables field measurement of gross activity, rather than determination of individual radio-nuclide activity, for comparison to the DCGL. The gross activity DCGL for surfaces with multiple radio-nuclides is calculated as follows:

  • Determine the relative fraction (f) of the total activity contributed by the radionuclide.
  • Obtain the DCGL for each radionuclide present.
  • Substitute the values of f and DCGL in the following equation:

Gross Activity DCGL = 1 / (f1/DCGL1 + f~2~/DCGL2 + … + fn/DCGLn)………………………….(3-4)

Example 3.7: Calculation of an integrated surface and soil contamination DCGL

Assume that 40% of the total surface activity was contributed by a radionuclide with a DCGL of 8,300 Bq/m2 (5,000 dpm/100 cm2); 40% by a radionuclide with a DCGL of 1,700 Bq/m2 (1,000 dpm/100 cm2); and 20% by a radionuclide with a DCGL of 830 Bq/m2 (500 dpm/100 cm2). Using Equation 3-4:

Gross Activity DCGL = 1 / (0.40/8,300 + 0.40/1,700 + 0.20/830)
= 1,900 Bq/m2

Note: That Equation 3-4 may not work for sites exhibiting surface contamination from multiple radio-nuclides having unknown or highly variable concentrations of radionuclides throughout the site. In these situations, the best approach may be to select the most conservative surface contamination DCGL from the mixture of radionuclides present. If the mixture contains radionuclides that cannot be measured using field survey equipment, laboratory analyses of surface materials may be necessary.

Because gross surface activity measurements are not nuclide-specific, they should be evaluated by the two-sample non-parametric tests described in Section 3.10 to determine if residual contamination meets the release criterion. Therefore, gross surface activity measurements should be performed for both the survey units being evaluated and for background reference areas. The background reference areas for surface activity typically involve building surfaces and construction materials that are considered free of residual radioactivity (see Section 3.3.5). The total surface activity due to residual contamination should not exceed the gross activity DCGL calculated above.

For soil contamination, it is likely that specific radio-nuclides, rather than gross activity, will be measured for demonstrating compliance. For radio-nuclides that are present in natural background, the two-sample non-parametric test described in Section 3.10.3 should be used to determine if residual soil contamination exceeds the release criterion. The soil contamination due to residual activity should not exceed the DCGL. To account for multiple background radio-nuclides, the DCGL should be adjusted in a manner similar to the gross activity DCGL described above. For a known mixture of these radio-nuclides, each having a fixed relative fraction of the total activity, the site-specific DCGLs for each radio-nuclide may be calculated by first determining the gross activity DCGL and then multiplying that gross DCGL by the respective fractional contribution of each radio-nuclide.

Example 3.8: Calculation of the DCGL for a known mixture of radio-nuclides

If 238U, 226Ra, and 232Th have DCGLs of 190 Bq/kg (5.0 pCi/g), 93 Bq/kg (2.5 pCi/g), and 37 Bq/kg (1.0 pCi/g) and activity ratios of 40%, 40%, and 20%, respectively, Equation 3-4 can be used to calculate the gross activity DCGL.

Gross Activity DCGL = 1 / (0.40/190 + 0.40/93 + 0.20/37)
= 85 Bq/m2

The adjusted DCGLs for each of the contributory radio-nuclides, when present in the given activity ratios, are then 34 Bq/kg (0.40 × 85) for 238U, 34 Bq/kg (0.40 × 85) for 226Ra, and 17 Bq/kg (0.20 × 85) for 232Th. Determining gross activity DCGLs to demonstrate compliance enables an evaluation of site conditions based on analysis for only one of the contributory contaminants (surrogate approach), provided the relative ratios of the contaminants do not change.

For situations where the background radio-nuclides occurring in background have unknown or variable relative concentrations throughout the site, it may be necessary to perform the two-sample non-parametric tests separately for each radio-nuclide present. The unity rule should be used to determine that the sum of each radio-nuclide concentration divided by its DCGL is less than or equal to one.

Therefore, at each measurement location calculates the quantity:

C1/DCGL1 + C2/DCGL2 + … + Cn/DCGLn ……………………………………………….. (3-5)

Where:
C is the radio-nuclide concentration.

The values of C are the data to be used in the statistical tests to determine if the average over the survey unit exceeds one.

The same approach applies for radio-nuclides that are not present in background, with the exception that the one-sample nonparametric statistical test described in Section 8.3 is used in place of the two-sample non-parametric test (see Section 3.5.1). Again, for multiple radio-nuclides either the surrogate approach or the unity rule should be used to demonstrate compliance, if relative ratios are expected to change.

1 Soil Screening Levels are currently available for chemical contaminants and are not designed for use at sites with radioactive contamination.

In Spain, USNRC soil secreening level have been used: 1) Federal Register/ Vol 63 No.222. NRC. Supplemental Information on the Implementation of the Final Rule on Radiological Criteria for License Termination 2) Federal Register/ Vol 64 No.234. NRC. Supplemental Information on the Implementation of the Final Rule on Radiological Criteria for License Termination.
– by Rafael Garcia-Bermejo Fernandez about 6 years ago
The data in Example 3.5 are not skewed (mean=4.56 and median=4.15) . If they were skewed is better to use the 95% or 99% UCL of the median according the recommendations in GILBERT, R. O.; 1987. Statistical Methods for Environmental Pollution Monitoring. Van Nostrand Reinhold, New York.
– by Rafael Garcia-Bermejo Fernandez about 6 years ago