A copula based representation for tailings dam failures

Abstract

In this article, we model the dependence between dam factor and D max, where dam factor is an indicator of risk of a tailings dam failure, which involves the height H of the tailings dam, the volume of material housed by the tailings dam VT and the volume dispensed by the tailings dam, VF, when the dam breaks. And, Dmax is the maximum distance traveled by the material released by the tailings dam, after the collapse. With the dependence found via copula models and Bayesian estimation, given a range of dam factor, we estimate the probability of the released material to exceed a certain threshold. Since the dam factor involves the released volume VF (unknown before the dam break), we present a naive way to estimate it using VT and H. In this way, it is possible to estimate the dam factor of a tailings dam and with such a value to identify the probability of the tailings dam to show a Dmax that exceeds a certain threshold.

Keywords

• copula models
• bayesian estimation
• conditional probability