Evaluation of Discrimination Technologies and Classification Results
Greg Van | Parsons
Objectives of the Demonstration
This report describes in detail the procedures, methods, and resources Parsons used to complete demonstration projects at the Former Camp Beale, CA; Former Pole Mountain Target and Maneuver Area (PMTMA), WY; Fort Sill, OK; Massachusetts Military Reservation (MMR), MA; former Camp George West, CO; and former Waikoloa Maneuver Area (WMA), HI, for the Environmental Security Technology Certification Program (ESTCP) Munitions Response (MR) projects MR-201104 (Evaluation and Discrimination Technologies and Classification Results) and ESTCP MR-201157 (Demonstration of MetalMapper Static Data Acquisition and Data Analysis).
To varying degrees, these projects were conducted with the following two primary objectives:
- Test and validate detection and discrimination capabilities of currently available and emerging advanced electromagnetic induction (EMI) sensors developed specifically for discrimination on real sites under operational conditions.
- Investigate in cooperation with regulators and program managers how classification technologies can be implemented in munitions and explosives of concern (MEC) cleanup operations.
Two advanced EMI sensors—the MetalMapper and the man-portable Time-domain Electromagnetic Multi-sensor Towed Array Detection System (TEMTADS) 2x2 (referred to here as TEMTADS), were tested as part of these projects. Both sensors employ multiple transmitters to induce electromagnetic fields in different orientations or locations relative to a given source object, and multiple receivers oriented in three directions (X, Y, and Z) to accurately measure the electromagnetic response generated by the source regardless of its orientation relative to the instrument. Both instruments also measure response over a significantly longer period than more traditional electromagnetic sensors. The MetalMapper uses three orthogonally-oriented transmitters to induce electromagnetic fields in the X, Y, and Z directions and seven triaxial receivers to measure the response generated by a source object following a pulse from each transmitter. The TEMTADS uses four horizontally-oriented (Z direction) transmitters arranged in a 2x2 array, and four triaxial receivers—each situated in the middle of the transmitters—to accomplish the same task. Using the standard settings applied for each instrument over the course of these projects, the MetalMapper typically measured response to approximately eight microseconds following each transmitter pulse, while the TEMTADS measured response to approximately 25 milliseconds versus the 1.25 milliseconds typically measured by the more standard EM61-MK2 (EM61) metal detector.
Data collected by the two instruments were processed using the UX-Analyze add-on to Geosoft’s Oasis montaj processing environment to determine various parameters for each target. The primary parameters of concern were the intrinsic polarizability (response over time) of each axis of the source object. Because the polarizabilities essentially do not change between different versions of the same object, the polarizabilities measured for an unknown source object at a field site can be compared to polarizabilities measured for standard objects such as ordnance. UX Analyze was also used to determine the degree of match between field data points and site specific libraries containing known polarizabilities for various versions of the ordnance suspected to be present at each site. Following library matching, each target was classified as either a target of interest (TOI) or non-TOI using a classification scheme developed for each site. A ranked dig list was submitted to ESTCP for each site, with the targets most likely to be TOI at the top and the targets most likely to be considered clutter at the bottom.
The standard military MR work conducted (i.e., seeding, dynamic data collection, intrusive work) on the projects proceeded without issue. There was little of note with these activities with the exception of minor changes to the intrusive procedures as the projects progressed, to streamline the process and to ensure that results were tracked accurately. The projects were also largely very successful with regard to the advanced EMI sensor data collection and classification efforts. Dig lists successfully identified 98%–100% of the TOI at each of the sites, with the exceptions of MMR, where the TOI detection rate was as low as 83% during the second phase of intrusive investigation, and the former WMA, where it was concluded that all detection survey targets would need to be excavated to remove all hazardous material. The goal of the project at MMR was more to reduce the amount of large TOI at the site than to identify all TOI present, so the misclassification of some TOI was not a significant concern. Classification using the MetalMapper at the former WMA was determined to be ineffective, primarily due to the excessive geologic response from the iron rich volcanic environment. Reductions in clutter digs ranged from about 61% at Fort Sill to about 91% for the second project at Camp Beale, with the exception of the former WMA MetalMapper classification study, where it was determined that no reduction was possible.
The single largest implementation issue on any of the projects was the failure of one of the MetalMapper transmitters during the second Camp Beale project. Diagnosis and repair of this issue required about 130 hours of field time for trouble-shooting and standby time at the beginning of the project. A transmitter board also needed to be replaced during the Fort Sill project, although diagnosis and repair of the problem did not take nearly as long as the additional time needed at Camp Beale. Additionally, the serial-to-universal serial bus (USB) converter that was originally shipped with the MetalMapper caused repeated software crashes during both the first Camp Beale project and the Fort Sill project. Each crash necessitated a restart of the collection software and re- addition of the points to be collected, which severely impacted the number of points that could be collected each day. Improved versions of this equipment were identified, and the problem has been solved. The largest implementation issue with the TEMTADS was a re-shot rate nearing 25%. The TEMTADS survey area at Camp Beale was tree-covered and much rockier than the MetalMapper survey area, and it is suspected that many of the TEMTADS targets were selected based on geophysical noise in the low-amplitude EM61 data.
Points of Contact
Mr. Greg Van
SERDP and ESTCP