On the performance of computational methods for the assessment of risk from ground-water contamination

The effect of parameter uncertainty and overly conservative measures on risk assessment has been addressed in numerous researches. Most of the work conducted to date is based on the use of the classic Monte Carlo simulation method (MCS) as a probabilistic modeling tool. Although the MCS is robust and asymptotically convergent, it lacks computational efficiency when the simulated probability is small. Furthermore, the sensitivity information can only be obtained with additional computational effort. First- and second-order reliability methods (FORM and SORM) have been developed in the structural analysis field and have been recently applied to ground-water contaminant transport and remediation problems. In this work, we extend the application of the reliability methods to the probabilistic assessment of cancer risk due to ground-water contamination. Results of the reliability methods compared well with a published case study of PCE contamination of a ground-water supply in California. The target risk level is extended over a larger range, and the sensitivity of the probability of failure to the relevant random variables is analyzed. The application of the methods to another case study, cancer risk due to the ingestion of benzene contaminated water, further illustrates a systematic way of directly accounting for the intrinsic uncertainty of the transport and fate model parameters involved in the risk assessment procedure. The probability of exceeding the target risk level in this case was found to be most sensitive to the uncertainty in the parameters describing the ground-water transport process.