Electromagnetic induction (EMI) sensors have shown promise for detection and characterization of metallic items in the marine environment. However, several technical and operational challenges still need to be addressed. These include reduced signal levels due to stand-off distance from the seafloor, the masking of target signals by conductive seawater, and reduced positional accuracy and non-uniform data coverage due to navigational difficulties underwater. This project aims to quantify and address these difficulties so that reliable and cost-effective detection and classification of munitions in the underwater environment becomes a reality. Specific objectives include:

  1. Develop a multilayer conducting model that can be used to better understand the challenges involved with marine sensing;
  2. Validate, evaluate, and test the use of terrestrial processing and interpretation strategies within the marine environment, including the use of library-based UXO classification techniques;
  3. Develop interpretation methods that are tolerant of the increased positional uncertainties encountered in marine sensing;
  4. Develop methods to improve target detection in the marine environment; and
  5. Verify the multilayer conductivity model so that it can be used as a reliable tool to evaluate the effect of different environment conditions on EMI sensors.

Technical Approach

This research will build upon the extensive body of knowledge and techniques that have already been developed for detection and characterization of buried metallic items in the terrestrial realm. The project will focus on EMI systems that are deployed to collect data while in motion, although the methods would also be applicable to static cued systems. A significant proportion of the project will involve the development of numerical models and algorithms, with a small field and laboratory component.

For the first objective of developing a multilayer conductivity model,  an integral equation technique implemented in a previous Strategic Environmental Research and Development Program (SERDP) project (MR-2412) will be extended. In order to test the applicability of terrestrial processing and interpretation strategies the team will obtain access to data collected by existing marine EMI sensors. Previous projects have found that efforts to apply new algorithms to “realworld” datasets typically provide useful insights into the development of improved processing methods. This project will supplement any field datasets with synthetically generated datasets where the team can rapidly assess the impact of a wide range of environmental variables or sources of noise.

To address the third objective, a robust inversion approach will be developed that explicitly accounts for sensor positioning uncertainties. It will also explore the use of an alternative technique that decouples the source estimation process from the position estimation process. To improve target detectability, the project will investigate the applicability of emerging synthetic source techniques that combine signals generated by different transmitters to enhance target signal levels. For the final objective of model verification, a variety of field and laboratory experiments will be conducted partly using equipment utilized in a precursor SERDP project. For the laboratory studies, a scaled physical model of the marine EMI sensing problem will be developed.


EMI is the primary technique used for detection and characterization of buried metallic items in the terrestrial realm. A number of underwater EMI sensors are either nearing the end of their development cycles or are already being utilized for production surveys. This project will attempt to improve our understanding of the marine EMI sensor process and the processing and interpretation strategies used on these new marine EMI sensors. The ultimate objective is to provide stakeholders with efficient tools to rapidly and effectively clear areas contaminated with underwater ordnance. 

  • UXO Detection and Classification ,

  • Electromagnetic Induction (EMI) Sensors ,

  • Environmental Models