K.7.0 NOMINAL CASE ANALYSIS FOR EX SITU INTERMEDIATE SEPARATION ALTERNATIVE
This section describes the results of the nominal case risk analysis for the Ex Situ Intermediate Separation alternative. The primary changes in the nominal, compared to the bounding case, were that the distribution coefficient (Kd) for Np-237, a major contributor to the groundwater risk, was assumed to be 1.0 rather than zero, which slows its movement through interaction with the soils. In addition, the residual inventory of isotopes C-14, Tc-99, and I-129 were reduced to 10 percent of the bounding residual inventory. These changes are described in more detail in Section K.6.5. The inventory of contaminants assumed to be released during retrieval for the nominal case analysis is presented in Tables K.7.0.1 and K.7.0.2. Differences from the bounding case are shown in bold.
Atmospheric releases during remediation may be chronic or acute. A nominal case was analyzed only for the acute (accident) case, because risks for the chronic bounding case were already very low.
K.7.1 EX SITU INTERMEDIATE SEPARATIONS ALTERNATIVE EXPOSURE ASSESSMENT
Uncertainties in the Ex Situ Intermediate Separations alternative human health exposure assessment can be divided into two parts. The first part discusses the uncertainties associated with the exposure parameters used in the post remediation land use scenarios. The second part discusses the uncertainties associated with the accidental release scenarios. In both cases, a Monte Carlo approach was used to evaluate both the uncertainty in the exposure assessment and to establish the parameters which contribute the most to the uncertainty in the exposure assessment (i.e., sensitivity analysis).
In the Monte Carlo analysis, PDFs were used to represent the range of values of a given parameter. The Crystal Ball computer software was then used to simulate a large number of realizations of the set of random variables in each model. This computation was repeated a large number of times to produce complete PDFs of the output function. Statistical summaries of the results were then plotted to interpret the data.
K.7.2 POST-REMEDIATION LAND-USE SCENARIOS
The results of the uncertainty analysis of exposure parameters for the post-remediation land-use scenarios in the Ex Situ Intermediate Separations alternative are summarized in Table K.7.2.1. The mean or nominal value of the Monte Carlo result was computed and compared to the bounding or fixed-point estimate of the same function. The nominal (Monte Carlo) results were generally approximately one order of magnitude less than the bounding estimates. This result supports the statement that the bounding value used in the exposure assessment in the EIS is an upper bound estimate.
The ILCR for the nominal and bounding estimates are provided in Table K.7.2.2. These risk estimates reflected the changes in the source and transport assumptions only. They do not incorporate the Monte Carlo estimates of variation in exposure parameters. The changes in source and transport assumptions for the nominal case decreased the ILCR by one to two orders of magnitude for most of the exposure scenario - future time combinations considered (Table K.7.2.2). These risk probability differences decreased the total accumulative cancer risk incidence by approximately an order of magnitude over the 10,000-year period of analysis. Although using less conservative assumptions for the bounding case altered the distribution of the risk through time for some alternatives, especially No Action (see Section K.6.5); for Ex Situ Intermediate Separation, the two cases followed the same general pattern, with the nominal case covering a smaller area, consistent with its lower accumulative risk (Figure K.7.2.1). The spatial pattern of risk onsite was generally similar in the two cases, as illustrated for the residential farmer scenario in Figures K.7.2.2, K.7.2.3, and K.7.2.4. At 2,500 years, the nominal case risk would be confined to the 200 East Area (Figure K.7.2.2), but at 5,000 and 10,000 years, the spatial distributions of the bounding and nominal cases were similar. The bounding case risk at 5,000 years had a higher risk area between 200 West and 200 East Areas, and occupied more area to the north and west (Figure K.7.2.3), but at 10,000 years, the areas for the two cases were essentially identical (Figure K.7.2.4).
K.7.3 ACCIDENTAL RELEASE SCENARIOS
A Monte Carlo uncertainty and sensitivity analysis was also conducted on the parameters used to compute LCFs or the probability of contracting cancer as a result of an accidental air releases associated with the remedial actions. The sensitivity analysis indicated that the parameters which contribute the most to the uncertainty in the LCF for the accidental release scenarios were, as measured by rank correlation: the ULD, the atmospheric dispersion coefficient (Chi/Q), the release volume, the inhalation rate, and the LCF conversion factor. Table K.7.2.1 compares the mean value of the LCF distributions for the four accidental release scenarios with the bounding estimate derived using upper-bound exposure factors. The results indicated that the LCFs predicted using the upper-bound values were in all cases one to two orders of magnitude greater than the mean of the Monte Carlo result. These results demonstrated that the predicted LCF estimates were upper bound estimates of cancer probability and/or fatality rates. The true probability of contracting cancer or fatalities resulting from cancer could actually be much less than the predicted value.
K.7.4 EFFECT OF ASSUMPTIONS ABOUT GROUNDWATER FLOW AND DIRECTION
As described in Section K.4.2.1, Future Groundwater Flow Direction, the groundwater modeling and risk assessment in the EIS assumed groundwater levels equivalent to those recorded in December 1979. The appropriateness of these data as a basis for impact analysis was tested quantitatively by running the model and calculating consequent risks for the Ex Situ Intermediate Separations Alternative using the same assumptions as in the bounding case analysis except that there was no water inflow from site waste water disposal to the vadose zone. This scenario was called the no mound case.
The effects of changing the assumptions for the Ex Situ Intermediate Separations alternative are illustrated for the residential farmer scenario in Figures K.7.4.1 through K.7.4.3. These changes eliminated the future risk north and west of the 200 Areas, consistent with the similar changes in groundwater flow, described in Section K.4.2.1 and Volume Four, Appendix F. However, the total risk increased in the no mound case (Table K.7.4.1). This effect was most pronounced at 5,000 years, when the risk for the bounding case was spread over a long southeast to northwest diagonal across the site, with a higher risk area between 200 West and East Areas (Figure K.7.4.2). The no mound case risk at 5,000 years was confined to a band running east from the Central Plateau, and most of the risk above 1.0E-06 was within the higher isopleth, above 1.0E-04 (Figure K.7.4.2, Table K.7.4.1). At 10,000 years, the areas of risk were comparable in the two cases, although the no mound risk occupied a smaller area with a higher risk level than the bounding case (Figure K.7.4.3, Table K.7.4.1). The total estimated cancer incidence over the 10,000 year period of interest was approximately four times higher in the no mound case than in the case used in the EIS (Table K.7.4.1).
The predicted flow field for this scenario tended to have a more pronounced west to east flow direction with similar gradient magnitude, compared to the December 1979 flow field on which the impact assessments were based. This resulted in a smaller groundwater contaminant plume. Figures K.7.4.1 through K.7.4.3 illustrate the risk that would be associated with the bounding case and no mound case.
Table K.7.4.1 Areas of Risk Contours and Total Concern Incidence, Bounding and No Mound Cases
Generally, a smaller contaminant plume would translate to higher contaminant concentrations and greater risk; however, the small plume would translate into few people being exposed. This effect would be partially offset by the longer vadose zone travel time and contaminant dispersion within the vadose zone.
|
NEWSLETTER
|
| Join the GlobalSecurity.org mailing list |
|
|
|
