Unexploded ordnance (UXO) presents a challenge to active military installations seeking to manage and clean their testing and training ranges as well as to sites designated for base realignment and closure and to formerly used defense sites attempting UXO remediation. In the United States alone, more than 900 sites with varying terrain, foliation, and topography are potentially contaminated with UXO. Using current technologies, the cost of identifying and disposing of UXO in the U.S. is estimated to be as high as $500 billion. New technologies capable of UXO detection that have high detection rates and low false alarm rates are required to reduce the cost of site cleanup drastically.

The aim of this project is to perform basic research on sensor development, sensor utility, and signature possibilities in the uncharted 25 – 300 kHz mid-frequency electromagnetic induction (MF-EMI) electromagnetic frequency band for induction sensing of buried UXO. The goal is to provide more options for discriminating ordnance from non-ordnance, thereby reducing false alarm rates during field surveying. This will be accomplished by innovative instrumentation development in the MF-EMI band, coupled with new modeling development.

Technical Approach

Technical objectives include the following: (1) performing lab measurements of soil electrical properties, including seasonal effects, in order to quantify expected subsurface signal loss rates; (2) extending and verifying a suite of computer programs able to achieve rigorous threedimensional solution of the physics of response by non-idealized UXO and non-UXO targets in the 25 - 300 KHz frequency range; (3) producing high fidelity simulations, involving time, space, and frequency domains, of the response by a wide range of specific UXO morphologies and dispositions as well as by common non-UXO targets; (4) obtaining measured induction responses for an array of UXO and non-UXO targets, using technology to be developed and exploiting existing databases where possible; and (5) using all of the above to identify distinctive UXO signature behaviors, and their discernibility relative to the environment, for combination with those being obtained in frequency ranges both above and below 25 KHz - 300 Khz.


All numerical codes have been completed and rigorously tested against both solutions and other methods. A measurement system has been designed and debugged at the lab level. It has succeeded in producing credible data that makes physical sense, match the well-established lower frequency data, and extend reasonably into the new frequency domain. New measurements have been extended into the new frequency domain with canonical objects and numerous UXO items.


This work builds directly on recent progress in innovative EMI signature identification in the lower frequency EMI range (100 Hz – 25 KHz), thereby amplifying its impact. It is directed toward aiding in (1) the substantial reduction of the false alarm rate in UXO field surveying; (2) the decrease in the cost of remediation of UXO hazard sites; (3) the provision for faster and safer surveying of potential hazard sites; and (4) computational and modeling tool development for a wide range of related electromagnetic applications.