(The) impact of below detection limit samples in (NPP) decommissioning residual risk assessment

Abstract

There are some technical problems to finalizing the regulations and standards for nuclear power plant (NPP) decommissioning including site soil analyses for license termination. It is difficult to precisely include the detection limit observations from site soil analysis, because typical soil collection data is reported above, at, and below the detection limit. In addition, multiple detection limits are required in order to adequately estimate the soil radionuclide concentrations. Below the detection limit observations are referred to as “left-censored data”. Left-censored data are known to be less than some detection limits, as defined by researchers analyzing the environmental data. Conventional approaches to managing the left-censored data either ignore or simply replace the detection limits with zero, a fraction of the detection limit, or the detection limit itself. However, these approaches are statistically biased and limited in their usefulness. The Kaplan-Meier method, robust regression on order statistics method, and maximum likelihood estimation method are proposed to estimate the left-censored data more precisely. The proposed methods are applied to a Monazite powder manufacturing plant and the Colorado School of Mines Research Institute. Summary statistics such as the sample mean, sample standard deviation, sample percentiles are calculated from the data set and are used in risk and volume estimation calculations. Previous environmental analyses, such as the exposure assessment, recognize the uncertainty issue that is addressed in this research using the Latin hypercube sampling approach. Probabilistic distributions of the input parameters are developed for the uncertainty analyses. Furthermore, the framework methodology for the decision-making process in the proposed method is presented. The framework defines the critical steps, the amount of radioactivity, types of distributions, censoring percentage, sample size, and number of detection limits

it also allows the user to select the appropriate approach based on their site-specific data and analyses. The goal of this research is to develop a more precise risk assessment, an estimation of the volume of soil removal, and the examination of cost savings.