Seeing Farther: Using the LANL Area G PA/CA Models to Investigate Radiological Risk 10,000 Years After Site Closure - 21277

To be meaningful, performance objectives for radioactive disposal sites must be linked to periods of compliance. Radiological risk, however, does not disappear after a specified compliance period has elapsed. In this work, the Los Alamos National Laboratory (LANL) environmental performance assessment (PA) for Material Disposal Area G (Area G) is extended to 10,000 years to examine the evolution of radiological risk beyond the 1000-year period required by regulation. Per Department of Energy (DOE) order 435.1, a PA model is required to examine radiological risk to members of the public and site intruders up to 1,000 years after site closure. In this analysis the modeling period is extended to 10,000 years after closure. Additionally, our modeling incorporates new data on bedrock depth below the Area G mesa-top disposal units, providing an update to the bedrock layer used in previous LANL Area G PA assessments. To support the extension of the PA modeling to 10,000 years, and associated composite analysis (CA) modeling, erosion modeling of the site is also extended using 10,000-year simulations. Our first objective is to assess how the site evolves farther into the future, focusing on long-term land-surface evolution, transport pathways, and other impacts to radiological dose. Dose to receptors on the mesa, in the canyons on either side, and downstream are evaluated through 10,000 years. A second objective is to evaluate if radiological risk increases considerably after the 1,000- year timeframe stipulated in DOE regulations, and, if so, examine different time frames that may be more informative. Drawbacks of the longer model time include increased parameter uncertainty. Uncertainty is addressed using scenario modeling with high-, moderate-, and low-erosion cases. We also provide probabilistic dose results (as 10/25/50/75/90% quantiles through time), in addition to mean annual dose metrics associated with prior Area G PA/CA assessments. Our long-term analysis indicates that simulated mean annual radiological doses continue to meet the 1000-year performance objectives until at least 10,000 years following site closure, with the exception of the Agricultural Intruder, who remains below the performance objective until 2,000 years. Probabilistic dose results show that the 90. percentile remains below 1000-year metrics for members of the public and intruder scenarios other than the Agricultural Intruder for 10,000 years. Typically, only the mean of the results is considered. However, here we show how probabilistic analysis improves understanding of the underlying distribution of results, and we analyze the impact of highly skewed distributions on the calculated mean. We conclude that extending the Area G PA/CA model to investigate radiological risk 10,000 years after closure is not critical from a regulatory standpoint, as current modeling indicates the site will meet 1,000-year performance objectives through much of the 10,000-year extended period. However, increasing simulation times to 10,000 years is useful for contextualizing site environmental processes and their impact on radiological risks. (authors)