Lessons Learned from the Cleanup of Formerly Used Defense and Military Munitions Sites
GAO-09-779T: Published: Jun 10, 2009. Publicly Released: Jun 10, 2009.
Under the Defense Environmental Restoration Program (DERP), the Department of Defense (DOD) has charged the Army Corps of Engineers (the Corps) with cleaning up 4,700 formerly used defense sites (FUDS) and active sites that were under its jurisdiction when they were initially contaminated. The 661-acre Spring Valley site in Washington, D.C is one such site. Like many other FUDS, the U.S. Army used the Spring Valley site during World War I for research and testing of chemical agents, equipment, and munitions. Most of the site is now privately owned and includes private residences, a hospital, and several commercial properties. The primary threats at the site are buried munitions, elevated arsenic in site soils, and laboratory waste; perchlorate was also found onsite. This testimony discusses GAO's past work relating to remediation efforts at FUDS and military munitions sites to provide context for issues at Spring Valley. Specifically, it addresses: (1) the impact that shortcomings in information and guidance can have on decision-making; (2) the impact that incomplete data can have on cost estimates and schedules; (3) how funding for a particular site may be influenced by overall program goals; and (4) how better coordination can increase public confidence in cleanups and facilitate effective decision-making. GAO has made several prior recommendations that address these issues, with which, in most cases, the agency concurred.
GAO's past work has found significant shortcomings in the Corps' use of available information and guidance for making decisions relating to cleanup of FUDS. For example, in 2002, GAO found that the Army determined that there was no evidence of large-scale burials of hazards remaining at Spring Valley before it had received all technical input. This experience is not unique. In a 2002 national study, GAO reported that the Corps did not have a sound basis for determining that about 1,468 of 3,840 FUDS properties--38 percent--did not need further study or cleanup action. GAO attributed these shortcomings to limitations in the Corps guidance that did not specify what documents or level of detail the agency should obtain to identify potential hazards at FUDS or how to assess the presence of potential hazards. GAO's past work has also shown that incomplete data on site conditions and emerging contaminants can interfere with the development of accurate cost and schedule estimates. At Spring Valley, the Corps' estimates of cleanup costs increased by about six fold, from about $21 million to about $124 million from fiscal year 1997 through fiscal year 2001. As assumptions about site conditions changed and new hazards were discovered, the estimates continued to rise and currently stand at about $174 million. Again, these problems are not unique. In 2004, GAO evaluated DOD's cleanup of sites with military munitions and found several similar weaknesses in preliminary cost estimates for numerous sites across the country. GAO's past work has shown that funding available for specific sites may be influenced by overall program goals and other priorities. Spring Valley has received priority funding due to its proximity to a major metropolitan area and high visibility; however, GAO's past work shows that this is usually not the case with most FUDS sites. Over the past 10 years DOD has invested nearly $42 billion in its environmental programs, but it typically requests and receives a relatively smaller amount of funding for environmental restoration activities at FUDS sites compared to funding available for active sites. GAO's past work has found that better coordination and communication with regulators and property owners can increase public confidence and facilitate effective decision-making for contaminated sites. With regard to Spring Valley, GAO reported in 2002 that the Corps, the Environmental Protection Agency (EPA) and the District of Columbia had made progress because they had adopted a partnership approach to cleanup decisions. However, this kind of cooperation and coordination does not always occur nationwide. For example, in 2003, GAO reported that the Corps only informed states of upcoming work and requested input from them about half of the time. Similarly, GAO found that the Corps did not always communicate with property owners about the decisions it makes regarding contamination at FUDS sites and more often than not did not inform property owners about how to contact the Corps in the event that further hazardous substances were identified at the site.