Cleaning Up Unexploded Ordnance Downloaded via 148.251.232.83 on September 4, 2018 at 14:14:51 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles. J A C Q U E L I N E A. M A C D O N A L D Technological limitations, cost uncertainties, regulatory disputes, environmental issues, and public safety concerns plague the transfer of military bases to civilian control. 372 A ENVIRONMENTAL SCIENCE & TECHNOLOGY / SEPTEMBER 1, 2001 2001 American Chemical Society
PHOTOS COURTESY OF U.S. ARMY CORPS OF ENGINEERS Photographsfrom leftto right:unexploded hand grenades,60-mm mortar,and practice rocketsand detonation fuses. CORBIS The presence of unexploded ordnance (UXO) makes military base closures difficult and creates complex safety and environmental problems. There are long-term concerns about possible soil and groundwater contamination from munitions residues, such as explosives and lead, and the dangers from disturbing or mishandling munitions that were fired but failed to detonate. The U.S. Department of Defense (DOD) faces high liability risks because of these problems, and the presence of UXO dramatically decreases the value of the lands. Cleaning up UXO contamination is clearly desirable, but successful remediation at closed and transferred bases is complicated by several important issues: The number of UXO-contaminated sites in the United States is unknown; there are technological limitations to detecting and clearing ordnance; and cleanup costs, which are very high, are poorly characterized. Moreover, there is an ongoing dispute over how best to regulate these sites. EPA, other federal agencies, the states, and DOD have not agreed on which office should have regulatory authority over the cleanup process, how that process should proceed, and what the ultimate cleanup goals should be. Although cost-effective remediation methods for UXO are yet to be developed, Congress has mandated that DOD turn over more property to the public or sell it. Because more military bases are likely to close, Congress and the Bush administration should address the problem of UXO as they consider policies for guiding these closures. Dimensions of the problem The presence of UXO is inevitable on any land used for military training or weapons development and testing. Dud rates how often a munition fails to explode are as high as 10% (1). In time, large amounts of UXO can accumulate, even if most of the surface UXO is periodically cleared away. The types of UXO vary widely, depending on land use. UXO can include small arms ammunition, bombs weighing up to one ton, artillery rounds, mortars, aircraft cannon, tank-fired projectiles, submunitions (which are designed to scatter over a large area), rockets, guided missiles, grenades, torpedoes, mines, chemical munitions, bulk explosives, and pyrotechnics. Each munition type poses different challenges for detection and clearance. Historically, the military has not tracked where ordnance landed after it was fired. Some munitions are followed now, but a requirement to monitor all munitions after they are fired would severely restrict combat training. Estimates of the number of UXO sites, based on historical surveys, vary. EPA has data (compiled in 1999) indicating that more than 7500 sites already transferred or slated for transfer out of military control could contain UXO. The Defense Science Board (DSB) Task Force on UXO estimates 1500 sites, including ranges on active installations, but it acknowledges this number is very uncertain because of the absence of detailed surveys (1). Most recently, DOD estimates (in a draft report intended for Congress but not finalized) that more than 4000 ranges, including those at active and closed installations, will require UXO cleanups (2). Sites range from small parcels of land to vast tracts covering thousands of acres. For example, UXO is being cleared from nearly the entire Hawaiian island of Kahoolawe, which was used as a live fire target area; in all, 28,800 acres of this island will be cleared. The DSB projects that even if only 5% of acreage suspected of containing UXO requires remediation, costs could exceed $15 billion (1). DOD estimates the total cost for closed, transferred, and transferring ranges at $11.5 $84.2 billion and the total cost for all ranges as $106.9 $391.5 billion in fiscal year 2000 dollars (2). An April 2001 U.S. General Accounting Office (GAO) audit concluded that DOD underestimated costs but that developing an accurate estimate currently is not possible (3). The uncertainty is partly due to a lack of agreement on the goals for UXO cleanup, such as the depth to which explosives should be cleared. SEPTEMBER 1, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 373 A
The number of casualties caused by UXO in the United States has been small compared to rates abroad. A limited EPA survey published last year of 61 DOD facilities containing 203 inactive, closed, transferred, or transferring ranges found reports of 24 incidents involving UXO, including 5 explosions that resulted in 3 deaths and 2 injuries (3). In a review of Explosives Safety Board information and news accounts, EPA documented 65 civilian deaths and 131 dismemberments since World War II. However, these numbers are also only estimates. The DSB Task Force on UXO cites two injuries and two deaths due to UXO at DOD ranges in the continental United States in 1996, but the task force did not research the total number of casualties (1). The task force predicts that casualties will increase as civilians gain access to closed bases (1). Internationally, 2000 civilians are injured or killed every month by land mines (1). Other UXO risks primarily from residual soil and groundwater contamination are poorly understood. Little information is available, for example, on the toxicity of the explosive RDX, a possible carcinogen that is highly mobile in the subsurface and has been found in the groundwater beneath the Massachusetts Military Reservation on Cape Cod. For almost all UXO sites, no data exist on the extent to which munitions components have or are likely to migrate into groundwater. Transfers increasing Military downsizing began with the end of the Cold War. In 1988 and 1990, Congress enacted Base Realignment and Closure (BRAC) legislation that required four rounds of base closures (in 1988, 1991, 1993, and 1995). As a result, the military has closed about 20% of major domestic installations and many smaller ones (4). This February, Sen. John McCain (R-AZ) introduced legislation that requires additional rounds of base closures. Under BRAC, DOD is required to clean up the closed properties and then either sell or transfer them to the public. Congress intended for base closures to save DOD the operation and maintenance costs associated with running so many facilities and for land sales to generate revenue. Although many bases have been closed, environmental contamination, often due to UXO, has delayed land transfers and sales. As of September 30, 1997, only about 14% of BRAC lands had been transferred, according to the GAO (4). RAND Corporation s discussions with DOD staff indicate that little additional land has been transferred since 1997. Although DOD spent nearly $7 billion cleaning up UXO risks primarily from residual soil and groundwater contamination are poorly understood. former installations, many of the sites are not suitable for reuse, and the military has had difficulty finding buyers, or even transferees willing to take the land for free. In the early years of the BRAC program, DOD predicted that it would receive $4.7 billion from the sale of BRAC land. Later revenue projections have decreased that figure to $122 million, in part because of the difficulty of transferring contaminated land (5). UXO has also been found on many parcels of land not associated with the BRAC program but used by the military and sold or returned to federal agencies or the private sector before 1986. Known as Formerly Utilized Defense Sites, or FUDS, most of this acreage is owned by the U.S. Department of the Interior, but some is also privately owned. UXO also has been found on FUDS that are now privately owned. PHOTO COURTESY OF USACE Technical limitations The traditional method for clearing UXO is called mag and flag. This involves surveying the range with a hand-held metal detector and placing a flag at the center of anomalous signals. Flagged areas are later excavated to search for UXO. Mag and flag has numerous problems: Detection of all the UXO present is not guaranteed, the false alarm rate is extremely high, anomalies like hubcaps and metal scrap account for 75% or more of the cost of UXO clearance, and the process is slow (6). Previous studies have shown that excavation by hand or backhoe can take as much as an hour per target, depending on the depth, and with the large rate of false alarms, much time is spent in fruitless digging (6). Extensive excavation can denude a site of its vegetation, damaging the environment; excavating UXO is risky. During the past decade, research on better UXO detection and clearance methods has increased. The most promising research is focused on improving detection capabilities to reduce the false alarm rate and minimize excavation. Two basic types of metal detectors are used for finding UXO: magnetometers, which measure distortions in the Earth s magnetic field caused by buried metal objects; and electromagnetic induction (EMI) devices, which induce an alternating electric current underground, then measure the return signal (the flux in the magnetic field created by the electric current) from metal objects. Traditionally, these devices were set to find any anomaly that caused a distortion or created a return signal above some threshold de- 374 A ENVIRONMENTAL SCIENCE & TECHNOLOGY / SEPTEMBER 1, 2001
tection level. Recent research has focused on carefully analyzing the return signals to look for patterns consistent with UXO. These newer methods rely on signal-processing algorithms to correlate the strength, frequency, or decay rate of the return signal with known patterns for different types of UXO. Two types of EMI devices are being investigated for this purpose. Frequency-domain EMI measures the frequency of the return signal from the buried object, whereas time domain EMI measures the decay rate of the return signal. New detection methods combine magnetometers with frequency-domain, time-domain, or both types of EMI detectors and then process the data using advanced algorithms to distinguish UXO from scrap. The new sensors and algorithms are promising. For example, at Fort A. P. Hill in Virginia, EMI frequency-domain sensors combined with traditional metal detectors and advanced signal-processing algorithms were shown to maintain a detection probability of 95% (i.e., found 95% of the UXO) and reduce the false alarm rate by seven-fold compared to using just conventional detection methods (7). The experiments, however, were conducted under ideal conditions at a test site that had first been entirely cleared and then reseeded with a limited amount of wellcharacterized UXO and clutter. The most comprehensive effort to develop and test commercial detection instruments was a series of field evaluations at Jefferson Proving Ground, IN, conducted in four phases between 1994 and 1999. Sites measuring 40 and 80 acres were seeded with unmarked, inert ordnance and clutter at depths and positions typical of different types of military munitions. First, instruments were calibrated in a small area seeded with marked ordnance. Then, multiple targets placed close together that had proved impossible to resolve in the first phase were replaced with single targets. For the third phase, targets, other than bombs, that were buried deeper than 2 m for the first two phases were excavated and either buried at shallower depths or removed entirely. During the first three phases of tests at Jefferson Proving Ground, none of the technologies successfully discriminated UXO from clutter. The final phase of testing involved modifying the experimental ranges once again, to focus on improving discrimination. The sites were cleared and then reseeded with samples of ordnance and non-ordnance. All of the targets were marked with flags. Vendors were provided with information about the seeded targets so they could calibrate their instruments. The challenge was to identify whether the marked targets were UXO or non-uxo. Even under these ideal conditions, only half of the systems could discriminate UXO from non- UXO at a rate that was significantly greater than the chance probability of 50% (see Figure 1). Research under controlled conditions, such as at Fort A. P. Hill and Jefferson Proving Ground, demonstrates the problems with available instruments. Numerous technical challenges to UXO detection must be overcome. UXO can be buried at depths >3 m, depending on UXO type and soil and weather conditions. Locating targets at depth can be difficult to impossible, depending on the target size and other factors. The detection capability of EMI devices decreases at a rate inversely proportional to the distance from the target raised to the sixth power (7). Deep targets have a smaller signal-to-noise ratio than shallow targets. Noise from cosmic and atmospheric sources, machinery, and radio transmitters interferes with target detection and leads to a high false alarm rate. The Jefferson Proving Ground tests also assessed ground-penetrating radar, synthetic aperture radar, and infrared sensors. None of these systems performed nearly as well as the EMI and magnetometry devices. Thus, it appears that the greatest potential for near-term improvements is in developing signal-processing algorithms that can provide more accurate information from EMI devices and magnetometers. The DSB recommends long-term research on novel sensor techniques such as seismic and acoustic methods, neutron activation systems, and synthetic dog s noses that can detect explosives. FIGURE 1 Detection test reliability Results shown are from Phase IV of the Jefferson Proving Ground Unexploded Ordnance (UXO) detection technology tests. Half of the vendor s technologies tested in this phase could distinguish UXO from non-uxo to some degree, although not to the Army s target goal of 95% correct identification of UXO and 75% correct identification of non-uxo. The diagonal line is the line of no discrimination capability. Filled circles are various vendor technology results. Any technology with results on or below this line has no ability to distinguish UXO from non-uxo. Unexploded ordnance correctly discriminated (%) 100 90 80 70 60 50 40 30 20 10 0 0 Source: Reference (5). 10 20 30 40 50 60 70 80 90 100 Non-unexploded ordnance correctly discriminated (%) Regulatory disputes The issue of how to regulate UXO was first raised by Congress in the 1992 Federal Facilities Compliance Act, which requires formerly exempt federal agencies to comply with the Resource Conservation and Recovery Act (RCRA), the statute that provides for cradle-to-grave management of solid and hazardous wastes. EPA determined that munitions used for nor- SEPTEMBER 1, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 375 A
mal military purposes training, R&D, and testing at active military-controlled sites need not be regulated under RCRA, because DOD already has extensive systems in place to guard against any public risks. However, EPA deferred the issue of whether munitions on closed military range lands should be regulated under RCRA. In the agency s view, sufficient regulations were not in place to govern munitions at these sites. EPA gave DOD the option of writing a Range Rule to specify how UXO at these sites would be managed, thereby avoiding RCRA regulation. In 1997, DOD published a draft Range Rule in the Federal Register (8), but EPA objected that the rule provided inadequate public health protection. The Departments of the Interior and Agriculture, along with a coalition of state attorneys general, argued that DOD lacked authority to even issue the rule. Negotiations among career staff at EPA and DOD failed to resolve concerns. In November 2000, Sherri Goodman, Undersecretary of Defense for Environmental Security, decided to withdraw the Range Rule from further review and end the negotiations (9). According to agency staff familiar with the negotiations, the primary sticking point was which agency should have final say in the steps to be taken for UXO clearance. DOD proposed that EPA s role should be limited to that of concurring with DOD s decision. However, EPA insisted throughout the negotiations that it should have primacy and that UXO clearance should occur using a process identical to that for cleanup of hazardous sites under Superfund. For now, the draft Range Rule remains shelved. Without such a rule, however, there is no certainty over what cleanup process and standards ultimately will be required for UXO sites, making further progress on this issue extremely difficult. Future base closures As Congress and the Bush administration consider policies for managing military land, the UXO issue should not be overlooked. Decisions to transfer more military land back to the public should consider whether continued attempts to reuse UXO-contaminated land for nonmilitary purposes make sense, or whether this land should remain under military oversight in perpetuity or at least until better detection and clearance technologies are available. If the latter, a strong commitment to fund research on and development of new detection and clearance methods, along with a commitment to fund UXO remediation over the long run, will be needed if military ranges are to be considered for new uses in the future. Because Goals for UXO clearance must be specified. some UXO is likely to remain even after the best current technologies are employed, difficult decisions must be made about what types of land uses can be allowed. Where UXO remains, some agency of the government must have long-term responsibility for enforcing land use restrictions to prevent public harm. Clear guidance from Congress and the Bush administration is needed to resolve UXO problems at military ranges already transferred or slated for transfer. Congress and the administration should clarify who has regulatory authority, and end goals for UXO clearance must be specified. The federal agencies have been unable to resolve these difficult issues. To make progress, new legislation from Congress or guidance from the highest levels of the Bush administration may be needed. PHOTOS COURTESY OF USACE References (1) Defense Science Board Task Force on Unexploded Ordnance Clearance, Active Range UXO Clearance, and Explosive Ordnance Disposal Programs. Report of the Defense Science Board Task Force on Unexploded Ordnance (UXO) Clearance, Active Range UXO Clearance, and Explosive Ordnance Disposal (EOD) Programs; Office of the Under Secretary of Defense for Acquisition and Technology, U.S. Government Printing Office: Washington, DC, 1998. (2) Department of Defense. Unexploded Ordnance Response Technology and Cost: A Report to Congress; Department of Defense, U.S. Government Printing Office: Washington, DC, 2001. (3) General Accounting Office (GAO). Environmental Liabilities: DOD Training Range Cleanup Cost Estimates Are Likely Understated; GAO-01-479; GAO, U.S. Government Printing Office: Washington, DC, 2001. (4) General Accounting Office (GAO). Military Bases: Status of Prior Base Realignment and Closure Rounds; GAO/NSIAD-99-36; GAO, U.S. Government Printing Office: Washington, DC, 1998. (5) Warren, D. R. Military Base Closures: Observations on Legislative Proposal for No-Cost Transfer of Surplus Property; GAO/T-NSIAD-99-215; GAO, U.S. Government Printing Office: Washington, DC, 1999. (6) Robitaille, G.; Adams, J.; O Donnell, C.; Burr, P. Jefferson Proving Ground Technology Demonstration Program Summary; SFIM-AEC-ET-TR-99030; Army Environmental Center: Aberdeen Proving Ground, MD, 1999. (7) Heberlein, D. C. Progress in Metal-Detection Techniques for Detecting and Identifying Landmines and Unexploded Ordnance; Institute for Defense Analyses: Alexandria, VA, 2000. (8) Department of Defense. Closed, Transferred, and Transferring Ranges Containing Military Munitions: Proposed Rule. Fed. Regist. 1997, 62, 50,795 50,843. (9) Center for Public Environmental Oversight. Defense Department Withdraws Proposed Range Rule. Military Environ. 2001, 8 (January), 1 2. Jacqueline A. MacDonald is an engineer with RAND Corp., Pittsburgh, PA, a nonprofit institute that aims to improve decision making through research and analysis. 376 A ENVIRONMENTAL SCIENCE & TECHNOLOGY / SEPTEMBER 1, 2001