The remediation of shallow, offshore water sites impacted by discarded military munitions and unexploded ordnances (UXO) is of increasing importance. Various sensing techniques, including sonar, laser, optical, electromagnetic induction (EMI), and magnetometry, have been developed to address the detection and characterization of metallic items in complex marine environments. Among these, marine EMI sensing has emerged as a promising technique for underwater munitions detection and classification, due to its ability to map, detect, and classify potential UXO items based on their polarizabilities.

Marine EMI sensing, however, faces distinct challenges in dynamic underwater environments, primarily due to conductive seawater and limited accessibility that necessitate large standoff distances from the seafloor. These conditions result in diminished target response signals, thereby limiting detection performance and affecting subsequent inversion and classification processes. To address this, high-powered transmitters can be employed to enhance target responses, but an alternative approach involves leveraging measurements from multiple transmitters and receivers, as exemplified by the Ultra Transient Electromagnetic Induction (UltraTEM) system. This approach utilizes a synthetic aperture (SA) method to constructively superpose measured responses, thereby increasing the detectability of buried targets.

This paper explores the applicability of the SA method in the context of UXO detection through time-domain EMI sensing. The SA principle and method are formulated and implemented for marine EMI dynamic acquisition. Practical application examples using UltraTEM marine high-altitude data from the Sequim Bay test site demonstrate the detection enhancement achieved with the SA method.

This project shows that the TEM synthetic aperture method effectively boosts target detectability by optimally stacking measured responses from multiple transmitter-receiver configurations. Optimal weights for each measurement are determined through energy-based optimization, ensuring maximum utilization of data. The composite implementation of the SA method involves constructing both SA receivers and SA transmitters. This method, applied with a moving window, acts as a local amplified detector, significantly enhancing target responses along profiles at large standoffs.

However, the SA method assumes signal coherence from nearby objects, leading to potential false amplifications of correlated noise. These false responses may exceed detection thresholds but are likely to be rejected during inversion and classification analysis due to their non-physical nature. Mitigating correlated noise in measurements could further enhance SA method performance.