Hazard area and recurrence rate time series for determining the probability of volcanic disruption of the proposed high-level radioactive waste repository at Yucca Mountain, Nevada, USA

摘要

The post-12-Ma volcanism at Yucca Mountain (YM), Nevada, a potential site for an underground geologic repository of high-level radioactive waste in the USA, is assumed to follow a Poisson process and is characterized by a sequence of empirical recurrence rate time series. The last ten time series are used as a prediction set to check the predictive ability of the candidate model produced by a training sample using autoregressive integrated moving average modeling techniques. The model is used to forecast future recurrence rates that, in turn, are used to develop a continuous mean function of the volcanic process, which is not only required to evaluate the probability of site disruption by volcanic activity but accommodates a long period of compliance. At the model validation stage, our candidate model forecasts a mean number of 6.196 eruptions for the prediction set which accounts for seven volcanic events of the 33 post-12-Ma eruptions at the YM site. For a full-scaled forecasting, our fitted model predicts a waning volcanism producing only 3.296 new eruptions in the next million years. We then present the site disruption probability as the chance that a new eruption will occur in the "hazard area" based on a model developed for licensing commercial space launch and reentry operations in the space transportation industry. The results of the site disruption probability and sensitivity analysis are summarized with a numerical table generated from a simple equation sufficient for practical use. We also produce three-dimensional plots to visualize the nonlinearity of the intensity function associated with the underlying model of a nonhomogeneous Poisson process and emphasize that the interpretation of site disruption probability should always be accompanied by a compliance period.