Probabilistic Prediction of the State of Discarded Underwater Marine Munitions

• Activity Number: 423 - Recent Advancements in the Analysis of Extremes
• Type: Contributed
• Date/Time: Tuesday, July 31, 2018 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics and the Environment
• Author(s): Jonathan Gillmore Ligo* and Sarah Rennie and Alan Brandt
• Companies: Johns Hopkins Applied Physics Laboratory and Johns Hopkins Applied Physics Laboratory and Johns Hopkins Applied Physics Laboratory
• Keywords: Remediation; Risk Assessment; Regulations; Extreme values; Expert System; Statistics and Policy

Abstract

After years of military activity there are many coastal sites contaminated with discarded military munitions (MM) in need of remediation. Over time, MM can to bury/unbury in the seafloor as well as migrate from their original locations, usually due to extreme events such as strong storms.  Since site surveys are often prohibitively expensive, the Underwater Munitions Expert System (UnMES) has been developed for predicting the distribution and burial of MM at sites of interest. UnMES is a dynamic Bayesian Network (DBN) that summarizes current knowledge on MM migration and burial by integrating environmental, laboratory, and field observations with physics-based modeling of behavior resulting from a sequence of storms. The DBN framework is especially suitable for site remediation planning as its probabilistic structure improves interpretability of prediction uncertainty. Statistical characterization of extreme events needed for implementation of UnMES in the coastal environment includes modeling the upper tail distributions of environmental forcings such as storm-generated wave and currents. UnMES predictions have been compared to field measurements of MM migration and burial.