JEFFERSON PROVING GROUNDS ADVANCED TECHNOLOGY DEMONSTRATION

UXO technology deficiencies came to the forefront of our nation’s newspapers with the public’s realization that the base realignment and closure (BRAC) process would not result in the immediate turnover of formerly used, Department of Defense (DoD) properties. A legacy of bombs, missiles, and rockets decades old, and even cannonballs from the past century restricts unlimited public use or access to these lands. In addition, active DoD installations considering alternative land uses must face unknown hazards, as record keeping of past ordnance usage was nonexistent or incomplete. There is an enormous demand to characterize properties just so the extent of the UXO hazard can be defined. In addition, there is a demand for lands to be returned to the public domain through UXO remediation efforts. UXO cleanup efforts are estimated to cost in the tens of billion of dollars.  The U.S. Congress established the UXO-ATD program to focus technology on reducing the unfunded liability and the time needed to characterize and remediate property. Congress recognized the need for more cost-effective and safer technologies. The USAEC manages the ATD program at JPG under the Congressional mandate to demonstrate advanced and innovative UXO technologies. A meaningful framework for understanding UXO technology performance was established by publishing public criteria and metrics. The ATD program would not only benefit restoration managers, who need to know more than just how to spell "ordnance" correctly, but also technology developers who would have quantifiable goals to seek against published performance.  NAEVA Geophysics employed a towed-array system of three ganged EM-61 units in data collection at JPG III.  Phases I and II In the first two phases, conducted in 1994 and 1995 respectively, ordnance was emplaced that was representative of different UXO conditions. Two sites, 16 and 32 hectares, were established for ground-based and airborne technology demonstrations. There were 29 demonstrations in Phase I and 17 demonstrations in Phase II. These demonstrations showed that airborne platforms and ground penetrating radar (GPR) sensors did not perform well under the test conditions at JPG. Demonstrators who used a combination of sensors (electromagnetic induction and magnetometry) had the best performance. The better performers in Phase II detected over 80 percent of the ordnance, but they also reported three to twenty times more targets (false alarms) than actual ordnance. The inability to distinguish ordnance from the prevalent farming debris at the site was noted, because this would likely be a major cost factor in remediating UXO properties. Excavation demonstrations of remotely operated systems were also demonstrated at the two Phases. Excavators could unearth ordnance at only a fractional rate (<5%) of how fast demonstrators could detect it.  Phase III In Phase III, the ordnance layout was changed form the earlier Phases to represent geographically-defined UXO scenarios. An Aerial Gunnery Range (1), Artillery and Mortar Range (2), Grenade and Submunition Range (3), and Interrogation and Burial Area (4) were established on the 16 hectare site. Demonstrators were allowed to select the scenarios that best represented their system’s capabilities for detection, localization and or characterization of the UXO. Remote excavation technologies were also solicited.    Sensor Type Demonstrator (Scenario #) PD False Alarm (FA) Rate (#/Hectare) FA Ratio (#/Ordnance Detected) Electromagnetic CHEMRAD (1, 2) 0.5 12.90 1.91 Induction (EM) GRI (EM) (1, 2, 3) 0.87 123.89 8.46 Geopotential (1, 2, 3) 0.06 9.04 8.54 Gradiometer Foerster (1) 0.60 36.46 4.85 Magnetometer Battelle (2) 0.12 1.71 1.00 (Mag) GRI (Mag) (1, 2, 3) 0.70 223.68 18.82 Rockwell (1, 2) 0.34 25.93 5.70 EM & Grad Geophex (1,2) 0.77 32.44 3.11 EM & Mag ADI (3; Mag only in 1, 2) 0.78 109.48 8.30 GRI (Combined) (1, 2, 3) 0.93 240.53 15.23 Geo-Centers (1, 2, 3) 0.93 81.80 5.18 Geometrics (2) 0.90 38.44 3.00 NAEVA (1, 2) 0.94 24.84 1.96 SCA_ADI (3; Mag only in 1, 2) 0.63 46.80 4.36 SCA_Geo-Centers (1, 2, 3) 0.76 43.55 3.36 SCA_Geometrics (2) 0.96 41.86 3.06 Ground Penetrating ENSCO (1, 2) 0.70 48.66 5.14 Radar & EM & Grad Averages 0.68 67.18 6.00   Map showing NAEVA's target locations.  Detection Performance Three of the demonstrators who has participated in at least two search scenarios, Geo-Centers, NAEVA, and GRI (combined), scored above 90 percent. All three used the combination of sensor technology that was most successful in Phase II, magnetometer and electromagnetic induction. NAEVA is further distinguished because not only did they achieve the highest PD, but their false alarm rate, 24.8 false alarms per hectare, is relatively low among all demonstrators. Geo-Centers’ false alarm rate is over three times that of NAEVA’s and GRI (combined)’s is ten times NAEVA’s. In examining the false alarm ration, a measure of the ordnance likelihood in target excavations, over one-third of NAEVA’s target excavations would have resulted in ordnance. This rate is twice that of Geo-Centers’ 16 percent rate and more than 5 times GRI’s 6 percent rate.    NAEVA Geophysics' performance at JPG III outranked that of all demonstrators from all three phases.