The investigators have developed a computer-based expert system that synthesizes understanding of the characteristics of burial, mobility, and re-exposure behavior of underwater unexploded ordnance (UXO) under coastal wave-dominated conditions: the Underwater Munitions Expert System (UnMES). This project will provide new capabilities for UnMES, including additional burial mechanisms, coastal environments, and longer time scales. The final product will be a practical software tool for guidance of underwater munitions site assessment and remediation efforts. The expert system is designed to perform under uncertainty, when knowledge of relevant conditions is limited, and to act as a repository for knowledge gained during the research efforts of multiple SERDP underwater Munitions Response projects. UnMES will effect the transition from research to application.
UnMES is a probabilistic Bayesian network for predicting the location of munitions and their degree of burial at underwater sites. This effort includes: (1) incorporation of recent SERDP research into burial by fluidization in high-energy conditions that mobilize sand sediments and inject pressure fluctuations into the seabed; (2) extension of the geographic coverage of UnMES to include modeling of beach face processes as well as munitions behavior on cohesive sediments found in estuarine environments; (3) augmentation of the UnMES architecture to predict patterns that evolve over time scales of months to years driven by infrequent extreme events; (4) a focus on statistical design that will provide improved insight into validation of probabilistic predictions; and (5) collaboration with real-world remediation site managers to design relevant visualization and guidance output products. This work will be in direct collaboration with other SERDP studies that focus on UXO response in a variety of underwater environments.
The probabilistic construct of UnMES feeds naturally into risk-assessment models used by site managers for remedial investigation decisions. Guidance regarding the timing, location, and operational choices for both Wide Area Assessments and subsequent clean-up efforts will be more efficiently planned and executed. Prediction of conditions that affect detection and classification performance by geophysical, acoustic, and optical sensors will guide optimal selection of sensor technologies for use within the characterized sub-regions. As UnMES will ultimately be used as an applied decision tool, continued development should incorporate input from an engaged user community. Therefore, a particular focus of this effort will be to develop visualization and risk assessment products in collaboration with managers of selected remediation sites of interest in order to ensure that UnMES directly addresses their needs.
The UnMES construct also provides a natural framework for expansion to incorporate future modules for the prediction of munitions’ condition (e.g. corrosion, biofouling, etc.), as well as linking with sensor detection performance models currently under development in the SERDP MR Program area.