Environmental Cleanup: Too Many High Priority Sites Impede DOD's Program
(Letter Report, 04/21/94, GAO/NSIAD-94-133).

Despite spending a reported $3.76 billion through September 1993 for its
program to clean up hazardous waste at high priority installations, the
Defense Department (DOD) has proceeded slowly during the past decade.
Most of the time and money has been spent on studying the problem, and
few hazardous waste sites have actually been cleaned up.  This report
reviews DOD's (1) progress in its cleanup efforts and estimated costs,
(2) plans for completing the cleanup at high priority installations, and
(3) factors that have affected progress in cleaning up high priority

--------------------------- Indexing Terms -----------------------------

     TITLE:  Environmental Cleanup: Too Many High Priority Sites Impede 
             DOD's Program
      DATE:  04/21/94
   SUBJECT:  Hazardous substances
             Waste disposal
             Defense operations
             Environmental monitoring
             Defense budgets
             Defense cost control
             Interagency relations
             Military facilities
             Environmental research
IDENTIFIER:  EPA National Priorities List
             DOD Installation Restoration Program
             DOD Environmental Restoration Program
             DOD Base Realignment and Closure Account
             EPA National Contingency Plan
             EPA Hazard Ranking System
             Superfund Program
* This file contains an ASCII representation of the text of a GAO        *
* report.  Delineations within the text indicating chapter titles,       *
* headings, and bullets are preserved.  Major divisions and subdivisions *
* of the text, such as Chapters, Sections, and Appendixes, are           *
* identified by double and single lines.  The numbers on the right end   *
* of these lines indicate the position of each of the subsections in the *
* document outline.  These numbers do NOT correspond with the page       *
* numbers of the printed product.                                        *
*                                                                        *
* No attempt has been made to display graphic images, although figure    *
* captions are reproduced. Tables are included, but may not resemble     *
* those in the printed version.                                          *
*                                                                        *
* A printed copy of this report may be obtained from the GAO Document    *
* Distribution Facility by calling (202) 512-6000, by faxing your        *
* request to (301) 258-4066, or by writing to P.O. Box 6015,             *
* Gaithersburg, MD 20884-6015. We are unable to accept electronic orders *
* for printed documents at this time.                                    *

================================================================ COVER

Report to the Chairman, Committee on Governmental Affairs, U.S. 

April 1994



Environmental Cleanup

=============================================================== ABBREV

  CERCLA - Comprehensive Environmental Response, Compensation and
     Liability Act
  DERA - Defense Environmental Restoration Account
  DLA - Defense Logistics Agency
  DOD - Department of Defense
  EPA - Environmental Protection Agency
  IRP - Installation Restoration Program
  NPL - National Priorities List
  TCE - Trichloroethylene

=============================================================== LETTER


April 21, 1994

The Honorable John Glenn
Chairman, Committee on
 Governmental Affairs
United States Senate

Dear Mr.  Chairman: 

The Department of Defense's (DOD) environmental cleanup program has
cost a reported $8.1 billion since the late 1970s; $3.76 billion has
been spent on high priority installations.\1 Concerned about the
effectiveness of DOD's program, you asked us to review DOD's (1)
progress in its cleanup efforts and estimated costs, (2) plans for
completing the cleanup at high priority installations, and (3)
factors that have affected progress in cleaning up high priority

\1 High priority installations include those listed on the National
Priorities List, those proposed for listing, those to be closed, and
those formerly used installations on the National Priorities List
where DOD is the lead agency. 

------------------------------------------------------------ Letter :1

Military installations are similar to small cities in terms of
population, industrial activities, and some types of contaminated
sites.  However, some cover an area larger than a small state.  DOD
has operated industrial facilities on its installations for several
decades that have generated, stored, recycled, or disposed of
hazardous wastes.  Many of these activities have contaminated the
nearby soil and groundwater.  To study and clean up contaminated
sites, DOD established the Installation Restoration Program (IRP) in
1975.  In 1984, the IRP was made part of the Defense Environmental
Restoration Program. 

Hazardous waste contamination can significantly contribute to
mortality and serious illness or pose a hazard to the environment. 
Types of hazardous waste found at most DOD installations and most
private sector sites, include solvents and corrosives; paint
strippers and thinners; and heavy metals, such as lead, cadmium, and
chromium found at most industrial operations.  Other substances such
as nerve agents and unexploded ordnance are found at some military
installations.  Contamination usually results from improper disposal,
leaks, or spills.  The primary contaminants found in a majority of
all DOD and private sector waste sites are petroleum products or
petroleum-related products such as solvents including
Trichloroethylene (TCE). 

The Superfund Amendments and Reauthorization Act of 1986 (10 U.S.C. 
2701) requires DOD to carry out its Defense Environmental Restoration
Program in accordance with the Comprehensive Environmental Response,
Compensation and Liability Act (CERCLA), as amended, commonly
referred to as Superfund (42 U.S.C.  9620).  CERCLA requires federal
entities to comply with the requirements of the law to the same
extent as any nongovernmental entity.  These requirements, set forth
in CERCLA and the National Contingency Plan, require DOD to comply
with specific guidelines regarding the degree of cleanup as well as
any other applicable federal laws and regulations.  DOD must comply
with any legally applicable or relevant and appropriate requirements
in state environmental laws that are more stringent than federal
requirements.  As a result, federal facilities must comply with both
federal and state laws and regulations unless the Environmental
Protection Agency (EPA) grants a site-specific waiver.  The act also
defines the process that federal facilities must follow to clean up
hazardous waste sites and outlines the requirements to be met, state
and local government participation, and the procedures for selection
and approval of remedial actions. 

Cleanup goals and strategies are usually site specific and depend
upon cleanup standards, the exposure potential, population exposed,
and the nature and extent of contamination.  All of these determine
the threat to human health and the environment.  Remedial actions are
taken to prevent or minimize the spread of hazardous substances so
that they do not migrate to cause substantial danger to public health
or welfare or the environment.  DOD gives the highest priority to
installations on EPA's National Priorities List (NPL) and those
scheduled to close. 

The Secretary of Defense delegated cleanup responsibility to the
Army, the Navy, the Air Force, and the Defense Logistics Agency
(DLA).  Cleanup actions are usually accomplished under contract with
private firms, which are monitored by the services.  Most cleanup
actions are funded through the Defense Environmental Restoration
Account (DERA)\2 and the Base Realignment and Closure Account. 

\2 Congress established DERA in 1984 to fund the cleanup of inactive
contaminated sites on DOD installations. 

------------------------------------------------------------ Letter :2

Despite spending a reported $3.76 billion as of September 1993, DOD's
environmental cleanup program for high priority installations has
proceeded slowly over the past 10 years, with relatively few
hazardous waste sites having been cleaned up.  Most of the time and
money has been spent studying the problem.  Of the 7,448 sites on 254
high priority installations,\3 remedial actions have been completed
at 205 sites and are in operation at another 29.  DOD has begun
studying most of the sites and is about to begin or has just begun
remedial design or actions at 3,873 sites.  The cost estimate for
cleaning up high priority installations is $18.2 billion.\4 However,
this cost estimate is based on preliminary information and is likely
to increase. 

EPA's system for identifying high priority sites has led to a large
number of individual sites on installations with that designation. 
In addition, some sites not designated as high priority are more
contaminated than some of the minor high priority sites and pose a
greater risk to human health and the environment than those minor
sites on the list, according to DOD officials.  EPA usually scores
only the four to six worst sites on an installation in determining
whether an installation, which may have hundreds of sites, should be
placed on the NPL.  Many of these sites may have only minor
contamination, but DOD program managers must apply the entire CERCLA
process to all the sites on an NPL installation, including those with
only minor contamination. 

DOD will not be able to efficiently institute cleanup efforts until
it and EPA evaluate the large number of sites currently on the NPL or
the closure list and determine which should be designated as high
priority.  Even a relatively few high priority sites could strain
resources and force difficult choices.  The Deputy Under Secretary of
Defense (Environmental Security) has proposed a new approach to
solving cleanup problems, which includes developing cooperative
rather than adversarial relationships with regulatory agencies,
setting priorities based on risk, and trying to accelerate cleanups. 

We also identified other key factors that have affected DOD's cleanup
of high priority installations in a timely and cost-effective manner. 

  The complex and time-consuming CERCLA study and cleanup process. 

  Prolonged study of hazardous waste sites rather than cleanup. 

  Disagreements with regulatory agencies over the extent of cleanup

  Addressing issues during the CERCLA process, such as liability,
     that generally do not pertain to governmental installations. 

  Scarce resources, including limited technology and expertise. 

\3 Of the 254 installations, 92 are current installations on the NPL,
10 are on formerly used installations, 25 have been proposed for the
NPL, and 127 are scheduled to be closed.  Seventeen installations are
on both the NPL and the closure list, and 5 closure installations are
proposed for the NPL. 

\4 The estimated cost to clean up DOD installations is based on data
provided by the installations, commands, and services. 

------------------------------------------------------------ Letter :3

According to DOD's 1993 Annual Report to Congress, DOD has identified
nearly 20,000 potentially contaminated sites on 1,722 DOD
installations and 8,000 potential sites on 1,632 formerly used
installations in the United States.  Of the 28,000 sites, DOD has
determined that 9,245 sites on current installations and 6,189 sites
on formerly used installations require no further action.  Of the
10,449 active sites on current installations,
5,507 sites are on 244 installations considered as high priority. 
For the 2,815 active sites on formerly owned installations, only 66
sites are on 10 NPL installations.  DOD has initiated or completed
the study phases (preliminary assessment/site inspection and/or the
remedial investigation/feasibility study) at 7,445 of the 19,694
sites, 3,825 sites have reached the remedial design/action phase.  Of
the high priority sites, 205 sites have had remedial action
completed, and another 24 have the remedy in place and operating.  As
of March 1994, none of the 92 NPL installations and 10 formerly used
NPL installations had all sites cleaned up. 

There are 2,521 potential sites on 149 installations scheduled to be
closed.  Seventeen closing installations are listed on the NPL, and
another 5 are proposed for listing.  There are 862 potential sites on
closing NPL installations, with 518 in the remedial design/remedial
action phase.  As of September 1993, 21 of the sites had remedial
actions completed and no further action was required at 85 sites (see
app.  II for discussion).  Table 1 shows the status of DOD's
hazardous waste sites. 

                                     Table 1
                      Status of DOD's Hazardous Waste Sites

                             Reached  With
Type of                        study  IRA\   Reached  RIP\         RA   Response
installation         Total     phase     a   RD/RA\b     c   complete   complete
------------------  ------  --------  ----  --------  ----  ---------  ---------
Active NPL           4,230     4,228   566     2,300    18        104      1,111
Closing NPL            733       733   114       474     0         20         79
Proposed NPL           631       631    64       288     5         21        221
Closing proposed       129       129    21        44     0          1          6
Closing non-NPL      1,659     1,558   124       719     1         59        458
Total                7,382     7,379   889     3,873    24        205      1,867
All sites           19,694    19,631  1,23    12,274    62        571      9,255
\a IRA--Interim remedial action. 

\b RD/RA--Remedial design/remedial action. 

\c RIP--Remedial action in place and functioning as intended. 

Source:  DOD's Annual Report to Congress, March 31, 1994. 

Of the 19,694 potential sites identified on the 1,722 current
installations, DOD has closed out almost half of these sites as
requiring no further action and has 10,449 active sites remaining;
3,773 are on 92 NPL installations, and 533 are on 25 proposed NPL
installations.  Of the 8,004 sites on formerly used installations,
DOD has closed out all but 2,815 active sites.  Of these,
66 are on 10 NPL installations.  Regulatory agencies have not
concurred in all of DOD's "no further action" decisions and may not
concur with DOD's position at times.  For example, EPA did not agree
to the sites closed out on the Pearl Harbor Complex.  In fact, of the
six sites EPA used in scoring the complex for inclusion on the NPL,
two were sites the Navy had determined to require no further action. 

DOD has spent over $3.76 billion to study and begin clean up of the
sites on the high priority installations.  The estimated cost to
complete the studies, investigations, and remedial actions is $18.2
billion.  DOD's estimates of future costs have been based on very
limited information.  As a result, it is difficult to accurately
estimate the remediation costs for a contaminated site before (1) the
nature and extent of contamination is known and (2) DOD selects the
remediation options and cleanup goals that regulatory agencies must
agree to.  However, it may be possible to estimate costs in ranges to
reflect alternative cleanup options.  Also, cleanup costs will depend
on the extent of the cleanup required and the remediation options
selected.  Table 2 shows DOD's estimates to clean up high priority

                           Table 2
           Estimated Cost to Clean up High Priority

                    (Dollars in millions)

Source/activity                                       d cost
--------------------------------------------------  --------
Fiscal Year 1992 Annual Report to Congress
73 NPL installations\a                               $ 8,579
17 NPL installations on Base Closure List              1,819
90 NPL installations (subtotal)                       10,398
Additional or updated DOD data
For 73 existing NPL installations                      3,642
For 17 NPL installations on Base Closure List            229
For four new NPL installations                           608
Increased costs                                        4,479
94 NPL installations (subtotal)                       14,877
Proposed DOD installations
20 installations proposed for the NPL                   820\
5 proposed installations on Base Closure List            230
NPL and proposed installations (subtotal)             15,927
6 formerly used NPL installations                        217
Non-NPL closing installations (all 3 rounds)         2,072\b
Estimated total cost                                $18,216\
\a Includes two formerly used sites. 

\b We generated data using DOD estimating factors for some of the
smaller closing installations. 

\c Data used from DOD's Annual Report to Congress contained some
errors that could have a minor effect on the total projected costs. 

In comparison, DOD has spent a reported $8.1 billion to study,
investigate, and clean up all DOD installations through fiscal year
1993.  DOD's 1991 official cost estimate, which is the most recent
one available, for studying and cleaning up about 20,000 potential
hazardous waste sites on more than 1,722 installations was $24.5
billion.  The estimated cost contained in the Annual Report for 1991
to clean up the NPL installations was $10.4 billion.  DOD officials,
in June 1993, stated the estimate to clean up all sites had risen to
about $30 billion. 

Until the fiscal years 1993 and 1994 appropriations, Congress had
funded DOD's cleanup program usually near the level DOD requested. 
In 1993, Congress reduced DOD's budget request by $313 million and
appropriated $1.2 billion.  In 1994, Congress reduced DOD's budget
request by $347 million and appropriated $2 billion. 

It is too early in the cleanup process to accurately project the
final cost, according to DOD officials.  DOD's cost estimates for
each NPL installation are based on the scope of work called for in
the Federal Facilities Agreements.\5 However, these agreements are
usually signed prior to the completion of the remedial
investigation/feasibility studies.  Until the studies are completed,
usually little is known about the nature and extent of contamination
at known sites and before many sites have been identified.  Even
then, after the nature and extent of the contamination is known,
cleanup costs may vary dramatically, depending upon the cleanup goals
and alternatives selected. 

DOD has 14 high priority installations where the cleanup estimate is
over $200 million each; 3 are over $1 billion.  A number of these
installations have significant contamination problems and could incur
greater cleanup or remediation costs.  For example, Congress directed
the Rocky Mountain Arsenal to be a wildlife refuge after the cleanup
has been completed.  Congress also directed that the Arsenal would be
cleaned up in accordance with CERCLA.  However, Colorado wants the
Arsenal to do additional clean up.  Table 3 depicts the cost to clean
up some large problem installations. 

                           Table 3
           Cost to Clean up Selected Large Problem

                    (Dollars in billions)

                                         DOD    Installation
                                      Annual       estimated
                                      Report       potential
Installation                           costs         costs\a
----------------------------------  --------  --------------

Rocky Mountain Arsenal                  $2.1      $10 to $20
McClellan Air Force Base                 1.7         5 to 10
Aberdeen Proving Grounds                 0.8          2 to 4
Jefferson Proving Grounds                 \b         5 to 10
\a Potential costs are estimates provided by installation officials
and are based usually on the worst-case scenario.  There are no legal
requirements for DOD to implement the worst-case scenarios. 

\b No estimate included in the 1992 Annual Report to Congress. 

In June 1993, DOD officials testified before Congress that too much
of DOD's environmental cleanup program was devoted to studying the
problem rather than cleaning up installations.  Because only 416
sites had been cleaned up at that time, including the 196 sites on
high priority installations, DOD has devised a new approach for
protecting the environment.  This approach is to

  create environmental partnerships with regulatory agencies and
     other stake holders,

  set priorities based on reducing real risk,

  focus efforts on getting cleanups done quicker,

  use existing and emerging technology to solve routine problems,

  increase management attention, and

  improve environmental funding process. 

On July 2, 1993, the President announced a five-part program\6 to
aggressively reinvest in communities and create jobs where
installations are being closed.  One part the President plans to
implement will involve "Fast Track Cleanup using a common sense
approach." The plan calls for a team of DOD, EPA, and state
regulatory agency representatives at each installation, as
appropriate, who will be empowered to run the cleanup program. 
Parcels with no contamination or with contamination below cleanup
levels will be identified quickly and made available for transfer. 
The cleanup team will conduct a bottom-up review of all schedules and
plans to speed up the planning, construction, and installation of
cleanup remedies and will build, publish, and implement an action

\5 Federal Facilities Agreements made between DOD and the regulatory
agencies establish comprehensive installation-specific schedules for
DOD's waste cleanup activities.  Federal Facilities Agreements have
been signed for 93 of the 102 NPL installations and are being
negotiated for the other 9, as of September 30, 1993. 

\6 We will be reporting on this program in a later report. 

------------------------------------------------------------ Letter :4

DOD has 254 high priority installations that contain 7,448 sites.  In
accordance with the National Contingency Plan, EPA has designated 102
of these installations as NPL or Superfund sites, which includes
5,029 individually contaminated sites.  Another 25 installations with
760 sites are proposed NPL sites.  Under current procedures, every
one of the 5,800 individual sites are subject to CERCLA's study and
cleanup requirements.  As of September 30, 1992, 3,154 of the 5,800
sites had reached the remedial design phase.  The majority of the
sites on DOD's NPL installations are not large enough or badly enough
contaminated to be considered Superfund sites, and many would go
undetected in the private sector, according to DOD and EPA officials. 

In the private sector, a site that scores 28.5 under EPA's Hazard
Ranking System\7 is subject to being placed on the NPL.  However, at
each DOD installation, EPA usually scores four to six of what appears
to be the worst contaminated sites based on available data and
combines their scores for the installation's composite score.  If
this composite score is 28.5 or higher, the entire installation is
subject to being placed on the NPL.  The majority of the 7,448
contaminated sites on DOD's high priority installations are not
Superfund type sites.  All but a few of these 7,448 sites were given
high priority status simply because they are located on a military
installation with a small number of badly contaminated sites. 

EPA's process for identifying the highest priority cleanup sites is
not being applied in the best way for DOD installations that contain
significant and insignificant cleanup sites.  EPA resources are too
limited to provide CERCLA oversight for about 20,000 sites on 1,722
DOD installations and about 8,000 potential sites on 1,632 formerly
used installations.  As a result, EPA has curtailed its CERCLA
oversight on non-NPL installations by limiting its oversight to NPL
and Base Realignment and Closure installations.  Even so, EPA
provides regulatory oversight for over 7,400 DOD sites on NPL and
closing installations.  With this number of DOD sites subject to the
CERCLA process, it is nearly impossible for EPA to meet all
procedural requirements at all current installations with existing
resources.  EPA also provides oversight on non-NPL sites in those
cases where cleanup is accomplished through the Resource Conservation
and Recovery Act corrective action procedures. 

DOD officials believe there are some non-NPL installations, which
contain a large number of individual contaminated sites, that are
contaminated enough to qualify for the NPL, but none of these
installations have been listed because EPA lacks the resources to
evaluate additional DOD installations.  EPA officials said they can
add only about 10 federal installations per year to the NPL due to
limited oversight resources.  In January 1994, EPA proposed an
additional nine DOD installations for the NPL.  As a result, DOD
officials believe that some of their worst sites are on non-NPL
installations, and thus, are not considered high priority sites or
given high priority and access to limited resources. 

In testimony before the Senate Committee on Governmental Affairs, the
Deputy Director of the Office of Management and Budget said that her
office and the White House Office on Environmental Policy will work
with the responsible federal agencies to help develop and carry out a
coordinated strategy for federal facilities waste management and
cleanup activities.  To do this, a policy-level interagency group was
created to help develop and implement a coordinated approach for
federal facilities waste management and cleanup programs, including
analysis and resolution of cross-cutting issues.  Office of
Management and Budget and EPA officials stated that one issue to be
addressed is the setting of priorities for determining which
hazardous waste sites are to be cleaned up first. 

By including the minor sites on NPL installations in its oversight
program, EPA is spreading its limited resources quite thin.  The
program managers and others are tied up looking at NPL installations,
including the minor sites, and do not have the time to look at other
sites on other non-NPL installations that are more contaminated than
the minor sites on NPL installations.  As a result, many of DOD's
contaminated sites, including some of the worst contaminated sites,
are being remediated without the regulatory oversight EPA believes is

In order to ensure adequate cleanup funding, DOD has worked with EPA
to set up procedures to include all sites on DOD's NPL installations,
including minor sites, in the NPL cleanup program.  As a result, DOD
unnecessarily expends significant time and resources applying the
CERCLA process to the minor sites.  These valuable resources could be
better used remediating the worst sites that have the greatest
potential for risk to humans.  By requiring DOD to consider the minor
sites on NPL installations, seriously contaminated sites on non-NPL
installations are allowed to worsen while less seriously contaminated
sites on NPL installations receive priority access to DOD and EPA

In cases where a site only has limited contamination and may not
require all CERCLA implementing procedures, CERCLA has provisions
where shortcuts can be made and work completed quicker.  It allows
DOD to reduce the level of effort at some sites.  However, EPA Region
VI officials believe it is essential to keep all sites under the
CERCLA process as long as possible because the site could be
contaminated more than originally thought. 

Officials at 9 of the 20 installations we visited stated they have
had difficulty getting EPA to agree to remove sites from the CERCLA
process.  Installation officials stated that EPA is requiring them to
continue to do additional investigative work at some sites.  Further,
EPA has not agreed to the reduced cleanup program at the sites where
DOD believes a reduced cleanup effort would be sufficient. 

For example, Schofield Barracks officials believe about 80 of nearly
100 sites could be excluded from a portion of the CERCLA process.  At
a number of these sites, quick actions could be taken and the site
eliminated from further CERCLA work.  EPA officials told us that they
had not agreed to these reduced requirements because installation
officials have not provided enough information for EPA to make the
final decisions.  EPA and Schofield have begun working toward
identifying which sites could be closed out with a minimum of action. 

\7 The Hazard Ranking System is a mathematical evaluation methodology
that EPA uses to assess sources of contamination, pathways, and
receptors (i.e., groundwater, surface water, air, and soil) to
determine if a hazardous waste site should be placed on the NPL. 

------------------------------------------------------------ Letter :5

At the commands and installations we visited, officials stated that,
historically, regulatory agencies have taken an adversarial oversight
approach similar to that taken with the private sector.  They stated
cooperation was hindered by EPA offices often being hundreds of miles
away, and the regulators made infrequent trips to see the sites. 
Rapid turnover of EPA's staff often meant the same person seldom
visited a site more than once.  As a result, EPA and DOD were unable
to develop the level of communication and trust needed to build good
working relationships.  Also, EPA officials were unable to gain the
first-hand site knowledge needed to make cleanup decisions. 

Officials at 4 of the 5 EPA regional offices responsible for
oversight of the 20 installations agreed with the need for teamwork
and a less adversarial approach.  Officials from these four regions
said they are attempting to build the better working relationship
needed to instill trust and cooperation and provide oversight.  Some
regions have set up a separate group of regulators specializing in
cleanups of federal facilities, including military installations. 
Both EPA and DOD officials believe a better working relationship
could greatly reduce the amount of time and money required to study
and clean up contaminated sites under the CERCLA process.  However,
federal facilities officials in the fifth region opposed the non-
adversarial approach.  They believed EPA should be tougher. 

One EPA region has implemented a cooperative program at two
installations we visited.  At Fort Wainwright, Alaska, EPA and the
state relocated their installation program managers closer to the
installation.  All three program managers stated that being in close
proximity to each other has enabled them to develop a good working
relationship by meeting regularly.  This allows them to deal with the
small problems before they became big ones, and each representative
has the authority to make decisions for their agency.  They check
with their supervisors on difficult decisions to make sure they make
the right decision.  In addition, they each have technical experts
they can turn to for assistance.  They also have less formal meetings
that allow them to resolve difficult problems and understand each
other's views.  This cooperative effort has worked so well at Fort
Wainwright that the Army uses it in its training program as an
example of how things should and can work. 

Despite its success at Fort Wainwright, other DOD installations have
not benefited from this management approach.  Air Force and EPA
Region X officials said that without frequent meetings and a good
working relationship, the time to study and clean up installations
can be lengthy and cost more.  They cited McChord Air Force Base as a
case where, at each decision point, the agencies used contractors to
prepare the backup data used to formulate each agencies' positions
and to present them to the other participants in reports rather than
the agencies' officials dealing with each other directly. 

According to Air Force officials, the CERCLA review process delayed
the remedial investigation/feasibility study phase of the program. 
McChord officials told us that EPA has issued a lot of guidance
documents for the CERCLA process that requires the installation to
conduct studies, prepare reports, submit draft reports to EPA for
review, and incorporate EPA's comments in the reports and plans. 
They stated this process is a problem because EPA wants the
installation to develop an iron-clad case that will stand up to
public scrutiny.  McChord officials stated EPA has told them that in
order to stand up to public scrutiny, it will require the
documentation to make the case stand up in the courts. 

Each part of the process usually involves several review steps to
issue reports or plans.  For example, a preliminary draft is prepared
and submitted to EPA for review and comment.  EPA develops its
comments using the technical expertise of its contractors.  McChord
then has to get its contractor to review EPA's comments and
incorporate the changes and resubmit it to EPA as a final draft. 
Again, EPA's contractors review and comment on this draft.  McChord's
contractor then incorporates these comments into the final report. 
At any one of these steps, more than one version could be required. 
EPA officials told us that they are trying to implement changes in
how issues are resolved with McChord.  Both EPA and McChord officials
told us that the working relationship between the agencies is

From July 1988 through September 1991, McChord officials analyzed the
time spent preparing and revising documentation submitted to EPA. 
During that time, 25 documents were prepared and often included
multiple versions incorporating EPA's and the state's comments.  Each
version of a plan or report usually costs $20,000 to $30,000.  The
reports and other documents required for the remedial
investigation/feasibility study have cost $3.1 million and have taken
over 37 months; the study is not expected to be completed until the
end of 1994.  McChord officials did not know what the total remedial
investigation/feasibility study costs would be. 

EPA and the state are requiring McChord to make a risk assessment for
the contaminated areas--American Lake Garden Tract and the Washrack
Treatment Area.  Even though industrial activity continues in the
Washrack Treatment Area and is expected to continue, McChord
officials stated that EPA officials told them to assume in this risk
assessment that condos will be built on the site and children will be
playing on it.  Based on instructions from EPA, according to McChord
officials, they are also to assume that all people living in these
condos will get their drinking water from the shallow aquifer. 
However, this aquifer is not used in any place in the area because
the water is contaminated by runoff from agricultural uses and other
small generators such as service stations and dry cleaners.  The
current source of drinking water is the lower aquifers.  Even though
this scenario was required for the risk assessment, it does not mean
that EPA will require McChord to clean up to residential standards. 

McChord provides piped-in drinking water for the American Lake Garden
Tract from the adjoining regional water system.  This has been done
even though the Air Force does not believe it is necessary because
the risk is low. 

Appendix III discusses other factors affecting DOD's cleanup program,
including CERCLA requirements, limited technical knowledge and
expertise, cleanup standards and goals, and restricted funding. 

------------------------------------------------------------ Letter :6

We agree with DOD's proposals to establish a more cooperative
partnership in the environmental community but believe fundamental
changes are necessary to ensure meaningful results.  We recommend
that the Secretary of Defense and the EPA Administrator

  revise the system for designating high priority sites and reduce
     the number of high priority sites currently included as part of
     the high priority program to a more manageable number by
     including only those DOD sites that would qualify as NPL sites
     in the private sector and

  modify EPA's process for applying National Contingency Plan
     requirements to all DOD sites. 

------------------------------------------------------------ Letter :7

To develop the information contained in this report, we reviewed
applicable procedures and records maintained by DOD, EPA, and the
state regulatory agencies.  We interviewed officials from the Defense
agencies, the Army, the Navy, the Air Force, and the Defense
Logistics Agency at the headquarters level in Washington, D.C.; at
major military commands, engineering field divisions, and service
organizations; and selected installations.  Appendix I lists the
organizations and installations we visited. 

We conducted our review from September 1992 to March 1994 in
accordance with generally accepted government auditing standards.  As
requested, we did not request formal agency comments on a draft of
this report.  However, we did discuss its findings, conclusions, and
recommendations with EPA and DOD representatives and have included
their comments where appropriate. 

---------------------------------------------------------- Letter :7.1

Unless you publicly announce its contents earlier, we plan no further
distribution of this report until 30 days after its issue date.  At
that time, we will send copies to congressional committees; the
Secretary of Defense; the Administrator, EPA; and the Director of the
Office of Management and Budget.  We will also make copies available
to others upon request. 

If you or your staff have any questions concerning this report,
please call me on (202) 512-8412.  Major contributors to this report
are listed in appendix IV. 

Sincerely yours,

Donna M.  Heivilin, Director
Defense Management and NASA Issues

=========================================================== Appendix I


Headquarters offices
Army Material Command
U.S.  Army Pacific, Honolulu, Hawaii
Army Environmental Compliance Center, Maryland
Aberdeen Proving Grounds, Maryland
Fort Lewis, Washington
Fort Sill, Oklahoma\1

Fort Wainwright, Alaska
Letterkenny Arsenal, Pennsylvania
Rocky Mountain Arsenal, Colorado
Schofield Barracks, Hawaii


Headquarters offices
Naval Facilities Engineering Command
Chesapeake Engineering Field Division
Pacific Engineering Field Division
Southwest Engineering Field Division
Naval Surface Warfare Center, Dhalgren, Virginia
Pearl Harbor Naval Complex, Hawaii
North Island, San Diego Naval Complex, California\1
Yorktown Naval Weapons Center, Virginia


Headquarters offices
Air Force Center for Environmental Excellence, Texas
Air Force Materials Command, Ohio
Hickam Air Force Base, Hawaii\1
Kelly Air Force Base, Texas\1
Mather Air Force Base, California
McChord Air Force Base, Washington
McClellan Air Force Base, California
Tinker Air Force Base, Oklahoma
Vance Air Force Base, Oklahoma\1


El Toro Air Station, California
Tustin Air Station, California


Headquarters offices
Region III, Philadelphia, Pennsylvania
Region VI, Dallas, Texas
Region VIII, Denver, Colorado
Region IX, San Francisco, California
Region X, Seattle, Washington
Environmental Protection Agency (EPA) Environmental Research
 Laboratory, Oklahoma



\1 Although these are neither National Priorities List (NPL) nor
closure installations, we visited them because of unique
contamination problems including, contamination of groundwater.  Data
from these were not included in our cost and site analyses. 

========================================================== Appendix II


The initial stage of the cleanup program is an installation wide
study to determine if sites are present that pose hazards to public
health or the environment.  Available information is collected on the
source, nature, extent, and magnitude of actual and potential
hazardous substance releases at sites on the installation. 


The next step consists of sampling and analysis to determine the
existence of actual site contamination.  Information gathered is used
to evaluate the site and determine the response action needed. 
Uncontaminated sites do not proceed to later stages of the process. 


Remedial investigation may include a variety of site investigative,
sampling, and analytical activities to determine the nature, extent,
and significance of the contamination.  The focus of the evaluation
is determining the risk to the general population posed by the


Concurrent with the remedial investigations, feasibility studies are
conducted to evaluate remedial action alternatives for the site to
determine which would provide the protection required. 


Detailed design plans for the remedial action alternative chosen are


The implementation of the chosen remedial alternative is implemented. 


Remedial actions can be taken at any time during the cleanup process
to protect public health or to control contaminant releases to the


The remedial action is functioning properly and performing as
designed.  These include such actions as the operation of
pump-and-treat systems that will take decades to complete cleanup. 

========================================================= Appendix III

Other factors affecting DOD's cleanup program include Comprehensive
Environmental Response, Compensation and Liability Act (CERCLA)
requirements, limited technical knowledge and expertise, cleanup
standards and goals, and restricted funding. 


The Superfund remedial process is applicable to 1,100 private sector
NPL sites that are made up of a small number of sources, yet it is
applied to nearly 7,448 sites on 254 high priority Department of
Defense (DOD) installations.  The study and cleanup of each site must
comply with CERCLA requirements, which are time-consuming. 

Most contaminated sites being remediated under DOD's Installation
Restoration Program and eligible for Defense Environmental
Restoration Account funds are subjected to CERCLA cleanup
requirements.  CERCLA requirements were designed to ensure Superfund
sites were cleaned up and to establish who was responsible for
cleanup costs.  Superfund sites usually are considered (1) to be the
worst contaminated sites in America and (2) to pose a significant
risk to human health and the environment.  As a result, according to
DOD officials, CERCLA cleanup procedures are numerous,
time-consuming, costly, complex, exacting, and call for a three-way
decision-making process by DOD, EPA, and state officials. 

Who is liable for the cleanup costs at private sector NPL sites is
nearly always a concern.  For example, the owners of landfills on the
NPL have often gone bankrupt or are no longer in business, and EPA
wants to get the individuals or firms, including federal agencies,
that contributed hazardous waste to the landfills to pay for the
cleanup.  Establishing liability requires extensive testing and
sampling to determine the source of contaminants and the
responsibility for the contamination.  Establishing liability also
involves a great deal of legal and administrative effort that is not
required if there is no question of liability. 

Contaminated DOD sites usually do not involve any party other than
DOD to contribute to cleanup costs.  Liability concerns as to who
should pay for the cleanup only arise at DOD installations when (1)
the contamination migrates beyond installation boundaries and other
sources of contamination commingle with that generated on the
installation, (2) a private firm or contractor that was allowed to
use military facilities generated the contamination, (3) a plume of
contamination has migrated onto the installation, (4) a contractor
operates a government-owned facility and there is a dispute as to
whether the contractor was responsible for the contamination, and/or
(5) property on closing installations is transferred and the new
owners can file claims for cleanup at a later time.  Because
liability is usually not a concern at DOD installations, performing
all of the CERCLA required steps may not be needed. 


Cleaning up contaminated sites is a relatively new area of endeavor,
with the vast majority of all site remediation occurring in the last
5 years.  Knowledge and expertise is especially limited when
confronted with massive amounts of contaminated soil or groundwater. 
Expertise and technology are increasing, but so far it is often not
technically possible or economically feasible to return a large
complex site to pristine condition.  Sometimes, it may be necessary
to protect human health by removing "hot spots" of high contaminant
concentrations and/or limiting exposure to the contaminants while
awaiting the development of better technology. 

Remediation strategies for contaminated soil are designed to remove
the contamination, prevent its migration to groundwater, limit human
exposure, and protect environmental interests such as ecological
considerations.  Timely remedial actions can prevent contaminant
migration into drinking water and save millions in cleanup costs. 
However, such actions depend upon the elapsed time since the spill or
leak occurred, the speed of migration,\1

the depth of the water table, and DOD's ability to respond quickly. 
Cleaning up the soil or preventing contaminant migration also depends
upon the developing technologies.  DOD is working with EPA on a
number of these. 

Natural remediation of highly contaminated drinking water aquifers
could take hundreds of years.  According to EPA officials, cleaning
up aquifers is a relatively new field and efforts to speed up the
process have been expensive and have achieved limited success.  Few
such cleanup efforts have advanced beyond the study phases in the
private sector and on DOD installations.  Technologies used include
pump-and-treat systems and new bio-remediation\2 methods designed to
enhance natural remediation.  Groundwater experts believe it is
necessary to isolate the contamination source and then by using
various methods, including natural remediation and pump-and-treat
systems, confine the spread of the pollution and clean up the

EPA and DOD officials believe pump-and-treat systems can be used to
contain the spread of the contamination plume,\3 although experts
believe it is nearly impossible to bring a highly contaminated
aquifer to drinking water standards.  Human health can also be
protected by treating contaminated water extracted for drinking
purposes.  For example, Schofield Barracks installed an air stripping
treatment system that reduces the amount of Trichloroethylene (TCE)
in drinking water from 40 parts per billion to less than 1 part per
billion.  Tinker Air Force Base plans to install a carbon filtration
system to remove TCE from the water used at the installation where
TCE contamination has been tested as high as 300,000 parts per


The "how clean is clean" question is now being debated across the
nation and in the courts, and the answer depends upon who sets the
cleanup standards and goals and what criteria were used.  The
President has introduced in his CERCLA reauthorization legislation, a
requirement for the development of standard cleanup levels.  Because
national standards do not exist for most contaminants in soil, DOD
must work with EPA and state governments to negotiate and set cleanup
goals for each site.  Under CERCLA, a state may set the cleanup
standard if its cleanup standard is more stringent than EPA's and the
standards meet all other CERCLA tests as applicable or relevant and
appropriate requirements.  State laws may also determine what is
considered a hazardous substance, and therefore, subject to the
cleanup process.  For example, petroleum products are considered
hazardous waste in California and Alaska, but not in Oklahoma. 

Cleanup studies include risk assessments to ascertain the threat
posed by a specific site.  These assessments consider the cleanup
standards, exposure to contaminants, and methods of limiting the
exposure.  In selecting a remedy, DOD must consider the risk and the
technology or other approaches available to accomplish the cleanup or
protect human health and the environment.  Exposure can be avoided by
limiting access to the contaminated area with fences and other
barriers, treating drinking water to remove contamination before it
is consumed, and restricting use of the property.  The exposure risk
is often greater for residential property where people may live 24
hours per day for a lifetime than for nonresidential property such as
a forest preserve. 

EPA's policy for CERCLA implementing regulations call for residential
cleanup standards as the starting point at sites where the eventual
use of the site is residential or unknown.  Residential standards are
usually more stringent than cleanup standards for contaminated sites
in industrial areas.  Most contaminated sites on DOD installations
were created by industrial operations and are located in industrial

Industrial sites on installations that will remain open will continue
to operate after the contaminated site is cleaned up.  Where DOD
installations remain open, DOD can control future land use and limit
access and/or exposure to protect human health and the environment. 
For installations designated to close, it probably will be necessary
to clean up to the reuse plan, which could include residential

Cleanup standards are usually the same for small lightly contaminated
sites and large heavily contaminated sites.  The cleanup standard can
be achieved at the small simple sites by removing and or treating a
relatively small amount of contaminated soil.  However, it is often
not technically possible or economically feasible to remove or treat
the huge volume of contaminated groundwater and soil found at large
complex sites, such as those on McClellan Air Force Base, Rocky
Mountain Arsenal, and Aberdeen Proving Grounds. 

For example, at the direction of Congress, the Army plans to reduce
human exposure to the contaminants on Rocky Mountain Arsenal by its
conversion to a fish and wildlife preserve where people will likely
only visit for a few hours, infrequently.  However, the United States
Court of Appeals for the Tenth Circuit ruled that Colorado can impose
its hazardous waste management laws and regulations.  Colorado
considers the Arsenal to be a storage and disposal site for hazardous
waste.  This is being done despite the fact that there is a CERCLA
cleanup underway.  All parties agree on the plan to make the Arsenal
a fish and wildlife preserve after the cleanup.  EPA officials also
pointed out that in some cases the cleanup required may be more
extensive where some endangered species could be affected. 
Addressing the state standards could result in significantly higher
costs for cleanup of DOD installations. 

On March 2, 1994, the long-standing United States verses Colorado
litigation was continued when the Army sued Colorado over what it
considered too stringent a cleanup standard for the chemical DIMP. 


The cost of cleaning up DOD installations can be affected by the
point in time during the fiscal year that funding is made available
to installation commanders.  Most hazardous waste cleanup projects
are funded with Defense Environmental Restoration Account (DERA)
funds.  A separate fund, the Base Realignment and Closure Account,
has also been established for closure bases.  DERA appropriations
have risen from $150 million in fiscal year 1984 to $1.96 billion in
fiscal year 1994.\4 DOD has requested $2.18 billion for DERA in
fiscal year 1995.  Cleanup funds are allocated to the installations
through the Army and the Air Force major commands.  The Navy
Engineering Field Divisions administer Navy DERA funds.  Under
current funding procedures, DOD installations must obligate these
DERA funds in the fiscal year received. 

At 6 of the 20 installations we visited, officials stated that when
they receive the funds in the latter part of the fiscal year, it
makes it difficult to get project planning, analysis, and the
contracting process accomplished in a short period of time. 
Sometimes they had to choose between (1) timely actions to obligate
the funds and (2) properly completing study requirements and the
necessary contracting procedures.  DOD officials told us that some
installations use a minor amount of discretionary funds to get the
pre-contract work done for selected projects in anticipation of
receiving year-end DERA funds. 

Installation and command officials said funds are often received late
in the fiscal year for a number of reasons, including funds becoming
available when other projects are canceled.  Installation officials
said they need a full year from the time the funds are received to do
all of the prerequisite work to get a contract signed and the funds

Office of the Secretary of Defense and command officials said it is
necessary to hold a certain percentage of funds for unforeseen
problems.  Also, some projects are delayed for unforeseen reasons,
such as contracting problems or changes in regulatory requirements. 

Government contracting procedures are designed to obtain the best
quality for the best price but are often time-consuming.  Taking
shortcuts to save time in order to obligate year-end funds can
increase the price and lead to disagreements with the contractors
over what needs to be done.  This can lead to inadequate contractor
performance and increased costs.  For example, at the Pearl Harbor
Naval Complex, the Pacific Division's\5 contracting office received
two-thirds of its funding in August and September.  This resulted in
the Pacific Division staff working overtime to negotiate and award
contracts for this money at the end of the year.  Because the fee
proposals and contract review period is compressed, the Navy's
ability to evaluate the contractors' proposals is constrained.  In
addition, contractors generally come in with a high cost estimate
because they do not have time to itemize and effectively determine a
scope of work needed to remediate a site.  As a result, the Navy does
not have assurance that it received the best cost for the required

\1 Migration speeds depend upon, among other things, contaminants
liquidity, weight, and volume; soil permeability; and the amount of

\2 Bio-remediation consists of using micro-organisms to neutralize
the contaminants. 

\3 Pathway of chemical constituent flow in the underground water

\4 Congress reduced DOD's fiscal year 1993 budget request by $313
million and appropriated $1.2 billion.  In fiscal year 1994, DOD
requested $2.3 billion, but Congress reduced the DERA appropriation
by $347 million. 

\5 The Pacific Division of the Naval Facilities Engineering Command
does the contracting for Pearl Harbor. 

========================================================== Appendix IV


David R.  Warren, Associate Director
Uldis Adamsons, Assistant Director
Jacob W.  Sprouse, Jr., Adviser


Robert G.  Hammons, Evaluator-in-Charge
Robert Jones, Evaluator