Nuclear Cleanup: DOE Should Reevaluate Waste Disposal Options	 
Before Building New Facilities (25-MAY-01, GAO-01-441). 	 
								 
Unless the Department of Energy (DOE) revisits its disposal needs
and its current option for disposing of wastes off-site, it could
miss opportunities to reduce cleanup costs at the Fernald, Oak	 
Ridge, and the Idaho National Engineering and Environmental	 
Laboratory (INEEL) sites and at other sites, such as Paducah,	 
that might propose the development of new on-site facilities.	 
Building in a decision checkpoint before major investment	 
decisions are finalized could identify instances when the use of 
off-site disposal would be less expensive, or when the cost	 
difference no longer outweighs the long-term risks associated	 
with on-site disposal. Such validation of the cost comparison is 
especially important in instances where DOE is aware that the	 
scope or timeframe of the cleanup effort has changed		 
dramatically. Remaining open to new proposals for off-site	 
disposal would also inject an element of competition into this	 
process. Thus, even if the validation did nothing more than	 
confirm the original decision to dispose of the wastes on-site,  
it has the potential to ensure that costs are kept to a minimum. 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-01-441 					        
    ACCNO:   A01059						        
  TITLE:     Nuclear Cleanup: DOE Should Reevaluate Waste Disposal    
             Options Before Building New Facilities                           
     DATE:   05/25/2001 
  SUBJECT:   Environmental policies				 
	     Nuclear facilities 				 
	     Radioactive waste disposal 			 
	     Site selection					 
	     Cost control					 
	     Cost effectiveness analysis			 
	     Decision making					 
	     DOE Fernald Environmental Management		 
	     Project (OH)					 
								 
	     DOE Comprehensive Environmental			 
	     Response, Compensation, and Liability		 
	     Act Program					 
								 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Testimony.                                               **
**                                                              **
** 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.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-01-441
     
Report to the Secretary of Energy

United States General Accounting Office

GAO

May 2001 NUCLEAR CLEANUP DOE Should Reevaluate Waste Disposal Options Before
Building New Facilities

GAO- 01- 441

Page i GAO- 01- 441 On- site Waste Disposal Letter 1

Results in Brief 2 Background 2 Sites Made Decisions Using Preliminary
Information About

Disposal Needs 5 DOE Has Not Used Updated Information to Reassess Disposal

Decisions Before Making Major Investments in On- site Facilities 10
Conclusion 20 Recommendation 20 Agency Comments 20 Scope and Methodology 21

Appendix I DOE Sites? Analyses of the Primary Tradeoffs Between On- Site and
Off- Site Disposal Alternatives 23 Fernald 23 Oak Ridge 24 INEEL 26

Appendix II Scope and Methodology 28

Appendix III Comments From the Department of Energy 30

Appendix IV GAO Contacts and Staff Acknowledgements 31

Related GAO Products 32

Tables

Table 1: Comparison of On- site and Off- site Disposal Cost Estimates Used
To Support the Record of Decision at Each Site 5 Table 2: Evaluation Methods
Used to Assess CERCLA Balancing

Criteria 8 Table 3: Key Tradeoffs in Each Site?s Analysis of CERCLA

Balancing Criteria 9 Table 4: Projected Waste Volumes for Disposal Used for
the ROD

and Current Estimate 11 Contents

Page ii GAO- 01- 441 On- site Waste Disposal

Table 5: The Effect of Current Prices for Low- level Waste Disposal on DOE?s
Off- site Estimates 17 Table 6: CERCLA Balancing Criteria 23

Figure

Figure 1: CERCLA Cleanup Decision Process 4

Page 1 GAO- 01- 441 On- site Waste Disposal

May 25, 2001 The Honorable Spencer Abraham Secretary of Energy

Dear Mr. Secretary: As the Department of Energy (DOE) cleans up its
contaminated nuclear sites, it must decide how best to dispose of huge
volumes of low- level radioactive waste (low- level waste) generated in the
cleanup process. 1 DOE estimates that 26 of its 144 nuclear production sites
will need to dispose of at least 6. 8 million cubic meters of cleanup
wastes- primarily contaminated soil and building debris- over the next 70
years. This amount would fill a space as long and wide as a football field
and almost one mile high. DOE generally allows its sites to select one of
several available disposal locations, including existing on- site
facilities, facilities at other DOE sites , or- depending on costs and other
factors- off- site commercial facilities. Because some sites do not have
existing on- site disposal facilities that can accommodate the projected
volumes of cleanup wastes, they are in the process of developing new on-
site facilities to specifically accommodate disposal of these wastes. Since
1996, three DOE sites- the Fernald Environmental Management Project (Ohio);
the Oak Ridge Reservation (Tennessee); and the Idaho National Engineering
and Environmental Laboratory (INEEL) (Idaho)- have decided to develop new
on- site disposal facilities. To date, only the Fernald facility has
disposed of any actual waste. At least one other site- the Paducah Gaseous
Diffusion Plant (Kentucky)- is considering whether or not to develop an on-
site disposal facility.

As part of ongoing reviews of DOE?s cleanup program, we determined how each
of the three sites decided to dispose of its low- level waste on- site
rather than off- site. Our analysis addressed (1) the extent to which site
officials took into account the comparative costs and risks associated with
off- site waste disposal, and 2) the extent to which site officials
revisited

1 DOE defines low- level waste as any radioactive waste that does not fall
within other classifications, such as high- level waste, spent nuclear fuel,
transuranic waste and uranium mill tailings. As used in this report, low-
level waste may include some mixed waste- that is, waste that contains
hazardous as well as radioactive materials. DOE also manages lowlevel and
other types of wastes generated from its activities relating to developing
materials for nuclear weapons and research.

United States General Accounting Office Washington, DC 20548

Page 2 GAO- 01- 441 On- site Waste Disposal

these cost and risk assessments after reaching their decisions to build
onsite waste disposal facilities.

Officials at the Fernald, Oak Ridge, and INEEL sites decided to develop new
on- site disposal facilities after conducting detailed cost and risk
assessments that compared, on the basis of the information then available,
on- site and off- site disposal. The officials at each site (1) compared the
estimated costs, projected risks to health and the environment, and other
factors, and (2) incorporated comments and suggestions from the public and
other interested parties through an open decision process. In each case,
site officials concluded, and EPA agreed, that the projected cost savings
from on- site disposal outweighed the uncertainties surrounding the long-
term costs and safety risks associated with keeping the wastes on- site.

In the year or more between the decisions to develop on- site disposal
facilities and groundbreaking for these facilities, circumstances at the
three sites have changed in ways that could affect the earlier cost
estimates for each site. For example, the projected volume of waste has
often increased and cleanup schedules have changed. Good business practice
suggests that site officials should reconfirm, on the basis of more current
information and cost comparisons, that on- site disposal remains
advantageous to DOE before constructing new disposal facilities. Recent
federal guidance directing agencies to validate their capital investment
decisions is consistent with this view. Officials at the three sites,
however, made little effort to update and reevaluate their original on- site
and off- site cost comparisons to determine whether on- site disposal
remains the preferred alternative when both costs and risks are taken into
account. Therefore, to ensure that on- site disposal decisions continue to
be advantageous to DOE, this report is recommending that site officials
reevaluate both on- site and off- site disposal options, especially when
project scope or timeframes have changed dramatically, before making major
investments in new on- site disposal facilities. We obtained written
comments on a draft of our report from DOE. The agency supported our
recommendation and noted that it had begun planning to address the issues
raised in the report.

DOE manages the disposal of cleanup wastes that come from remediation,
decontamination, and demolition at sites where operations have been
discontinued. Cleanup wastes are primarily subject to three laws: the Atomic
Energy Act of 1954, as amended; the Resource Conservation and Recovery Act
of 1976 as amended (RCRA); and the Comprehensive Environmental Response,
Compensation, and Liability Act of 1980, as Results in Brief

Background

Page 3 GAO- 01- 441 On- site Waste Disposal

amended (CERCLA). DOE is responsible for the management of its own
radioactive wastes under the Atomic Energy Act. Under RCRA, the
Environmental Protection Agency (EPA) or states with programs authorized by
EPA regulate the hazardous components of mixed wastes. The Congress enacted
CERCLA to clean up the nation?s most severely contaminated hazardous waste
sites. Under CERCLA, which is administered by EPA, the parties responsible
for the contamination are responsible for conducting or paying for the
cleanup. The statute makes federal facilities subject to the same cleanup
requirements as private industry.

For CERCLA projects, EPA has established a decision process designed to
involve the public and EPA in identifying, evaluating, and choosing cleanup
approaches. This process requires parties responsible for the cleanup (in
this case DOE) to consider a range of cleanup alternatives. EPA uses nine
specific criteria, including the estimated costs, feasibility, and risks for
each alternative, to evaluate, compare, and balance tradeoffs among these
alternatives (see fig. 1). Under these criteria, any selected cleanup
alternative must adequately reduce long- term risks to human health and the
environment. The chosen alternative- including the plan for disposing of the
waste- is documented in a Record of Decision (ROD), which EPA must approve.

Page 4 GAO- 01- 441 On- site Waste Disposal

Figure 1: CERCLA Cleanup Decision Process

Identify scope and nature of contamination

DOE performs a Remedial Investigation to determine the nature and extent of
contamination at the site, and to identify additional data needs. Using
appropriate standards, DOE develops a Baseline Risk Assessment to assess
potential threats to human health and the environment.

Activity CERCLA Guidelines

Develop and Evaluate Cleanup Alternatives

Select Preferred Cleanup Alternative

DOE performs a Feasibility Study- often concurrent with the Remedial
Investigation- to set cleanup goals, identify potential cleanup technologies
and processes, develop and screen alternative cleanup approaches, and
perform detailed evaluations of cleanup alternatives using

EPA reviews and approves both the Remedial Investigation and the Feasibility
Study.

Nine CERCLA Criteria Threshold criteria:

1. Overall protection of human health and the environment 2. Compliance with
applicable or relevant and appropriate legal requirements

Balancing criteria:

3. Long- term effectiveness and permanence (risk) 4. Reduction of toxicity,
mobility or volume through treatment 5. Short- term effectiveness (includes
risk) 6. Ease of Implementation (feasibility) 7. Cost

Modifying criteria:

8. Acceptance by the state 9. Acceptance by the community

DOE identifies its Preferred Alternative by balancing trade- offs among the
9 criteria and then prepares a draft Proposed Plan for state and EPA review
as well as public comment. Based on the Proposed Plan and ensuing comments,
DOE then drafts a Record of Decision. After the state, EPA and other
interested parties review and comment on the draft, DOE issues the Final
Record of Decision with any appropriate revisions, and obtains EPA's
signature.

Nine CERCLA criteria.

Page 5 GAO- 01- 441 On- site Waste Disposal

Officials at the three sites we reviewed considered detailed estimates of
the costs and risks associated with on- site and off- site waste disposal.
Among other things, these estimates were based on preliminary determinations
on the extent and type of contamination present at the site. In accordance
with the CERCLA decision process, site officials also assessed how well each
cleanup alternative addressed the nine CERCLA criteria. After balancing the
tradeoffs among the criteria for each alternative, site officials selected
an on- site disposal alternative based, at least in part, on their
estimation that on- site disposal would cost less than off- site disposal
(see table 1). To meet the CERCLA requirement that human health and the
environment be adequately protected, DOE sites adopted accepted strategies,
such as limiting the level of contamination allowed in the disposal
facility, to mitigate long- term risks. DOE, EPA, and other stakeholders
agreed that the benefits of on- site disposal, including cost savings,
outweighed the remaining long- term risks.

Table 1: Comparison of On- site and Off- site Disposal Cost Estimates Used
To Support the Record of Decision at Each Site

Present value in the year of the estimate, dollars in millions

Extent to which off- site disposal cost estimates exceeded on- site
estimates

DOE Site Year of estimate

On- site estimate

Off- site estimate Amount Percent

Fernald Environmental Management Facility a 1995 $578 $772 $194 34 % Oak
Ridge Reservation -- low volume b 1997 100 133 33 33 % Oak Ridge Reservation
-- high volume b 1997 168 450 282 168 % Idaho National Engineering and
Environmental Laboratory (site- wide estimate)

1998 187 605 418 224 % a Fernald estimates included significant costs to
remediate the Great Miami Aquifer. b Oak Ridge compared costs for scenarios
at the low and high ends of their expected volume range.

In accordance with the first steps in EPA?s CERCLA decision process, all
three sites conducted remedial investigations to confirm and to quantify the
nature and extent of contamination. 2 They examined site background and
historical data, and used limited sampling to project the volumes and types
of wastes that could be generated by cleanup activities. Based on

2 For the cleanup decisions we reviewed, the CERCLA decision process and
documentation focused on a cleanup decision for one contaminated area- not
the entire DOE site. For sites like the three we reviewed, officials
typically divide the site into a number of contaminated areas, in part
because not all areas will be cleaned up at the same time. Sites Made
Decisions

Using Preliminary Information About Disposal Needs

Sites Identified the Extent of Cleanup Needed and Developed Alternative
Cleanup Actions

Page 6 GAO- 01- 441 On- site Waste Disposal

this limited information, officials at each site developed a preliminary
model describing the sources of contamination (such as soil or groundwater),
possible ways the contaminants could be released, and whether human exposure
would be likely. Using this model, they assessed the cancer and non- cancer
risks to humans. 3

Officials at each site also prepared a feasibility study that established
cleanup goals, identified possible cleanup technologies and actions, and
analyzed alternative cleanup approaches. For the contaminants of concern,
officials set cleanup goals at contamination levels that posed acceptable
risks according to their exposure model. The waste excavation and disposal
approach, either on- site or off- site, was only one of many approaches
available to officials to meet these cleanup goals. For example, each site
considered leaving at least some waste in place and limiting human exposure
to it by either capping the waste with clean cover materials or restricting
access to the waste areas. Site officials developed their alternative
cleanup approaches using the results of their remedial investigations and
working closely with EPA and state reviewing officials.

Officials in Oak Ridge and Idaho determined that feasible cleanup approaches
were likely to generate more waste than the existing disposal facilities at
those sites could accommodate, and the Fernald site had no existing disposal
facility. Therefore, when conducting their feasibility studies, officials at
each of the three sites considered whether to dispose of their respective
wastes in a new on- site facility or to ship them to an offsite disposal
facility. Specifically, each site used the Envirocare facility in Utah as
its representative off- site disposal facility. DOE and commercial
generators of radioactive waste use this facility, which is located 80 miles
west of Salt Lake City, to dispose of mildly contaminated soils and debris.
In addition to the Envirocare facility, DOE sites that do not have existing
on- site disposal facilities are now authorized to dispose of their low-
level and mixed low- level wastes at DOE?s disposal facilities for these
types of

3 Under CERCLA, EPA uses standards described in the National Contingency
Plan, and accepts levels of contaminants present in sufficient
concentrations to create an excess lifetime cancer risk within, or less
than, the range of 1 chance in 10, 000 (10 -4 ) to 1 chance in one million
(10 -6 ). Non- cancer health effects are assessed in terms of a hazard index
for each contaminant of concern. The calculated hazard index indicates the
potential for the most sensitive individuals, such as children, to be
adversely affected. Hazard indices are compared to a threshold value of 1,
established by EPA as the level above which there is the potential of non-
cancer effects on exposed individuals.

Page 7 GAO- 01- 441 On- site Waste Disposal

wastes at its Hanford Reservation in southeastern Washington and its Nevada
Test Site in southern Nevada. 4

Site officials assessed each proposed cleanup alternative against the
cleanup goals, as well as nine decision criteria specified in EPA?s CERCLA
regulations and guidance. Following this guidance step- by- step, officials
first considered the two threshold criteria, then evaluated qualifying
alternatives against the five balancing criteria, and then applied the two
modifying criteria. CERCLA threshold criteria require a cleanup approach to
(1) achieve overall protection of human health and the environment and (2)
meet all legal requirements, referred to as ?applicable or relevant and
appropriate requirements.? Site officials discarded some alternatives, such
as capping some contaminated areas in place, because they did not meet these
threshold criteria. In some instances, waivers were needed to develop on-
site disposal facilities. For example, the Fernald site obtained a waiver
from the state of Ohio?s prohibition on developing a disposal facility over
a drinking water aquifer. Similarly, the Oak Ridge site obtained a waiver
from EPA? s minimum required distance from the bottom of a landfill that
contained toxic chemicals to the underlying groundwater. Without these
waivers, the sites would not have been able to develop onsite disposal
facilities. (See appendix I.) In both instances, the host states and EPA
agreed with site officials that the proposed facilities could be designed to
meet equivalent safety standards.

After screening cleanup alternatives against the two threshold criteria,
site officials developed more detailed feasibility studies to demonstrate
how well the various surviving alternatives met each of five CERCLA
balancing criteria. The sites used measures that employed varying degrees of
data and subjectivity to evaluate how effectively an alternative met each
criterion (see table 2).

4 There are, however, roadblocks to fully using the disposal facilities at
Hanford and the Nevada Test Site. The two host states may oppose increases
in waste disposal at the sites, and DOE may need to obtain environmental
permits from these states to dispose of out- ofstate mixed wastes. Sites
Used Nine CERCLA

Criteria to Evaluate Cleanup Alternatives

Page 8 GAO- 01- 441 On- site Waste Disposal

Table 2: Evaluation Methods Used to Assess CERCLA Balancing Criteria CERCLA
balancing criteria Evaluation methods used by site officials

Long- term effectiveness and permanence Performed extensive disposal
facility and groundwater modeling to project the potential for disposed
wastes to escape from an engineered facility and migrate, resulting in
exposure risks that exceed CERCLA standards. Reduction of toxicity,
mobility, or volume through treatment This balancing criterion was less
applicable to the decision at all three sites. All three sites

may consider any applicable new treatment technologies. Short- term
effectiveness Considered, and to some extent quantified, risks to people
from accidents and exposure

during cleanup actions, including transportation options. Identified impacts
to the community and environment from activities like excavation and road
building. Ease of implementation Considered whether proposed technologies
were untested or readily available.

Considered logistical and political implications of shipping waste across
the country, or leaving waste on- site. Cost Prepared detailed cost
estimates that included costs for: planning, management,

equipment, any on- site construction, annual operations, transportation,
disposal fees for any waste sent off- site, and, for on- site alternatives
only, facility closure and post closure activities for up to 100 years.

Converted estimated costs to present worth dollars for comparisons, as
required by CERCLA. Noted that costs contained significant uncertainties
thought to approximate CERCLA guidelines of a -30 to a + 50 percent margin
of error.

All three sites then used these evaluations to balance the criteria for
onsite and off- site disposal alternatives. Generally, each alternative
approach had strengths and weaknesses in some of the criteria, and the sites
had to make tradeoffs according to their unique conditions and priorities.
Table 3 lists the key tradeoffs each site cited in its comparative analysis
between on- site and off- site alternatives of similar cleanup scope. (See
appendix 1 for a description of each site?s comparative analysis.)

Page 9 GAO- 01- 441 On- site Waste Disposal

Table 3: Key Tradeoffs in Each Site?s Analysis of CERCLA Balancing Criteria
DOE Site Comparative analysis of balancing criteria for on- site and off-
site disposal alternatives

Fernald Off- site disposal provided the most long- term protection at the
Fernald site. However, on- site disposal of qualifying waste was selected
because of reduced cross- country transportation risk, a slightly lower
cost, and less dependence on the future availability of off- site disposal
facilities. Oak Ridge Off- site disposal in an arid location provided better
long- term protection of groundwater. However, an on- site disposal

facility would not contribute substantially more to potential exposure,
because other portions of the site are expected to remain contaminated. On-
site disposal selected because of reduced risk from cross- country
transportation, a lower costs for the highest estimated waste volume, and
less dependence on the future availability of off- site disposal facilities.
INEEL Both on- site and off- site disposal would be effective in the long
term- provided the on- site facility remained protective

of the groundwater. Off- site disposal would be more difficult to implement
because it required long distance waste shipments and depended on the
availability of off- site disposal capacity. (The Feasibility Study,
however, provided minimal support for this conclusion.) On- site disposal
selected because of lower costs for the projected waste types and volumes,
as well as minor reductions in transportation risks. a

a the increased risks of physical accidents and injuries- not radiological
exposure- was the differentiating factor

Officials at each site applied the two CERCLA modifying criteria- state and
community acceptance- to its preferred alternative of on- site disposal for
most of its waste. Each site involved state and community stakeholders early
in the decision process. State environmental agencies participated in
preliminary reviews and informal discussions from the start of the Remedial
Investigation throughout the final cleanup decision. Generally, by the time
site officials issued their Final Record of Decision, they had addressed, or
had a plan to address, environmental concerns raised by the states. For
example, the Ohio Environmental Protection Agency supported the development
of an on- site facility in Fernald contingent upon specific restrictions on
the source of and radioactivity in any waste accepted for disposal. The
Record of Decision incorporated an approach to meet these restrictions.

Site officials also involved and informed community stakeholders, and
received support for their decisions from groups such as the Fernald
Citizens Task Force, the Oak Ridge Reservation Environmental Management Site
Specific Advisory Board, and the INEEL Citizens Advisory Board. The Fernald
Citizens Task Force, which is comprised of individuals with diverse
interests in the future of the site, convened in 1993 to provide focused
input on central cleanup issues at Fernald. A Task Force report issued in
1995 included recommendations on the site?s future use, waste disposal
options, and cleanup objectives and priorities. DOE?s selected alternative
mirrored these recommendations.

All three sites held public hearings on their Proposed Plan and ROD, and
accepted comments for the time periods required under CERCLA. The

Page 10 GAO- 01- 441 On- site Waste Disposal

resulting comments and DOE responses were incorporated in the decision
documents. In each case, host state environmental agencies concurred with
the proposed decisions. Each DOE site and its respective EPA region then
signed a Final ROD that documented the decision for an on- site disposal
facility. This final decision allowed DOE sites to move forward with
planning for site excavation and construction.

After deciding to build new on- site disposal facilities, site officials
continued to refine disposal needs and develop specific plans for these
facilities for one or more years. During this time, significant changes
occurred in site assumptions regarding the types and volume of wastes
needing disposal, detailed design of on- site facilities, duration of the
cleanup, and cost of off- site transportation and disposal. Under such
circumstances, good business practice suggests that earlier cost estimates
should be confirmed before construction begins. Likewise, 1997 guidance
issued by the Office of Management and Budget (OMB) states that agencies
should validate their earlier planning decisions with updated information
before finalizing capital investments. However, the three sites conducted
little further evaluation of off- site disposal options, despite changed
circumstances that could narrow the cost difference between onsite and off-
site disposal. At Oak Ridge, for example, a simple update of the projected
waste volumes, transportation rates, and costs for off- site disposal of
some types of waste effectively reduced the difference between on- site and
off- site cost estimates by 51 percent. Such changes in relative costs could
also affect the balancing of costs and other factors considered while making
cleanup decisions. In particular, uncertainties about long- term stewardship
needs become more significant as cost differences narrow. The elapsed time
between the preparation of the initial cost estimates that were used to
support the disposal decision and the commencement of construction of on-
site disposal facilities argues for validating the initial cost comparisons
before committing funds to construction of new facilities. DOE has not taken
advantage of this time to update their cost comparisons at the three sites.

A year or more can elapse between the time the costs are estimated and the
commencement of actual cleanup activities. During this period, officials at
the three sites we reviewed continued to determine the extent and nature of
contamination needing cleanup, and often changed their assumptions about
waste volumes, waste types, cleanup duration, and the type of disposal
facility needed. Although such changes can have major implications on cost
estimates for both on- site and off- site disposal, officials at the sites
applied the CERCLA process in a manner that DOE Has Not Used

Updated Information to Reassess Disposal Decisions Before Making Major
Investments in On- site Facilities

Assumptions Changed as Sites Refined Cleanup Plans

Page 11 GAO- 01- 441 On- site Waste Disposal

discouraged re- examination of costs for alternatives other than their
previously selected approaches.

At all three sites, the waste volumes used to compare on- site and off- site
disposal costs were significantly less than the waste volume currently
projected for on- site disposal. At two of the sites, site- wide cleanup
plans and waste projections were not well defined when the cost estimates
were prepared. Officials at those sites now expect to dispose of much more
waste. Officials at the Fernald site noted that, although the site?s cost
estimate was based on 1. 4 million cubic meters of waste from one operable
unit, the overall decision making process was based on the sitewide estimate
of 1. 9 million cubic meters. (See table 4.) As the volume of waste grows,
the potential need to construct additional disposal capacity to accommodate
the waste also grows. At the time of our review, Oak Ridge officials stated
that they would need to obtain further geologic surveys and regulatory
approval before expanding the disposal facility to accommodate the larger
waste volume now projected. Because the cost comparisons were largely
limited to an earlier set of assumptions about waste volumes, without
preparing updated cost estimates DOE is not in a position to assess whether
these changes will have a substantial effect on the comparative costs of on-
site and off- site disposal.

Table 4: Projected Waste Volumes for Disposal Used for the ROD and Current
Estimate

Amounts in cubic meters

DOE Site Waste volume used for comparing on- and off- site costs Current
projection of waste

volume for disposal

Fernald 1, 400, 000 1,900, 000 a Oak Ridge 170,000 - 840, 000 2,200, 000
INEEL b (Waste Area Group 3) 63,000 228,000 INEEL b (site- wide) 356,000
369, 000 c a In its proposed plan, the Fernald site projected that
approximately 1. 9 million cubic meters of lowlevel waste generated site-
wide would be consolidated in the disposal cell; however, the cost estimates
were based on approximately 1.4 million cubic meters of waste projected
under the ROD for Operable Unit 5. b The cost estimate in the ROD for the
INEEL on- site disposal facility was based on just 63, 000 cubic meters of
waste projected from cleanup of Waste Area Group 3. Altogether the INEEL
site has 10 Waste Area Groups throughout the site. In their 1998 feasibility
study, officials estimated costs for disposal of 356,000 cubic meters of
waste projected site- wide. c In addition, INEEL officials said that at
least 312, 000 more cubic meters of low- level soils waste will need
remediation when the adjacent chemical plant and tank farms are dismantled
after the year 2035- well after the planned on- site disposal cell is
scheduled for closure.

Source: DOE

Waste Volume and Types Have Changed

Page 12 GAO- 01- 441 On- site Waste Disposal

Further investigation of the contaminated areas at the sites also changed
assumptions about the types of waste that will be generated. This is
especially important because the disposal requirements- and therefore, the
cleanup costs- vary by waste type. For example, mixed waste- waste that is
radioactive and also contains hazardous substances- must be disposed of in
facilities that meet more stringent RCRA standards. Because meeting RCRA
standards increases disposal costs, the proportion of mixed waste in cleanup
waste will affect overall cost estimates. Disposal fees at the Envirocare
facility, for example, are much higher for mixed waste than for low- level
waste. Also, cost estimates can be affected by how much of the waste is
building debris, such as concrete or metal, and how much is soil. Building
debris can cost more for disposal due to its awkward sizes and shapes. Sites
may also need to obtain additional fill material to properly dispose of
debris, or they may need to adjust their disposal schedules to ensure a
proper mix of the two types of waste. 5 Onsite facilities that need to
increase their disposal capacity, purchase additional fill, or adjust
disposal schedules will probably face higher costs than originally
estimated.

Since developing their cost comparisons, the three sites have continued to
change their assumptions about the length of the cleanup. After finalizing
their cleanup decisions and selecting on- site disposal, site officials
revised their on- site cost estimates to provide justification for their
annual budget proposals over the next few years. These revisions often
resulted in changed assumptions about the time needed for cleanup
operations. The revised on- site disposal estimates reflected project life
cycles that accelerated cleanup schedules according to DOE?s 1998 plan to
complete cleanup at most of its sites by 2006. 6 The abbreviated schedules
assumed that facilities would operate for fewer years, tending to reduce the
original on- site estimates. For example, since preparing their first cost
estimates, Oak Ridge officials have shortened their projected schedule for
on- site

5 Disposal cells are designed to have enough soil around the debris to fill
in any voids to reduce the potential for subsidence (sinking) of the cell
contents, which can weaken the cell?s permanent cap. If enough contaminated
soil is not available, the facility will need to obtain fill material to put
around it. Oak Ridge is currently projecting a waste volume that may include
as much as 35 percent clean fill; however, studies are underway to identify
means of reducing or eliminating clean fill requirements. Similarly, waste
disposal shipments need to be scheduled so that enough soil is available to
fill around shipments of debris. Officials at the Fernald site noted that it
has been necessary to coordinate decontamination and demolition projects
around the site to have the debris ready for emplacement at the proper time.
6 Accelerating Cleanup: Paths to Closure; DOE/ EM- 0362; June 1998 Cleanup
Schedules Remain in

Flux

Page 13 GAO- 01- 441 On- site Waste Disposal

disposal from about 30 years to about 10 years and officials at Fernald
decreased their operating schedule from about 20 years to 13 years.
Officials at these sites did not update comparable estimates for off- site
disposal because they no longer considered off- site disposal to be a viable
option.

The sites? cleanup schedules remain in flux. The current operating schedules
and related disposal cost estimates appear optimistic. Fernald officials,
for example, state that funding constraints are already forcing a slowdown.
In fiscal year 2001, Fernald plans to dispose of 60, 000 cubic yards (after
compaction), or 36 percent, of the 168, 000 cubic yards called for in the
project?s baseline. Schedules at Oak Ridge and INEEL could face similar
pressures. For example, the INEEL site estimated the operating costs for on-
site disposal of site- wide cleanup wastes for approximately 10 years, even
though site cleanup could be much longer, because cleanup schedules had not
been finalized for all waste areas around the site. If current schedules
prove unworkable, then the costs for on- site disposal will change. However,
there will be no comparable analysis for off- site disposal.

As on- site and off- site cost comparisons were originally made, plans for
on- site facilities were purely conceptual: design details, engineering
drawings, and even the exact locations of the facilities were still being
determined. Concurrent with improving information on the projected waste
volume and types following their on- site disposal decisions, officials at
the three sites also developed and refined engineering designs for their
respective planned facilities. These refinements reflected changes in
assumptions about such things as geologic features at the proposed facility
location and the exact nature and level of contamination the disposal
facility could safely accept. For example, additional geological surveys
were needed at INEEL to determine how deep the cell could be built without
hitting bedrock. Ultimately the cell depth will affect the area of land
covered by the facility and thus the amount of material needed for the final
cap. Another facility design feature that continues to evolve is the proper
soil to debris ratio that was discussed above. DOE officials? opinions on
the optimal ratio have varied from 1: 1 to 8: 1, and the final ratio will
depend upon the physical condition of the debris.

As disposal facility plans become better defined, the resulting decisions
are likely to have cost implications. For example, when INEEL developed its
cost estimate, the tentative plans did not include a facility for sizing,
sorting or treating the wastes. INEEL officials have since added plans to
construct an on- site treatment facility, which they currently estimate will
Facility Designs Are Still Being

Developed

Page 14 GAO- 01- 441 On- site Waste Disposal

cost $15 million. Similarly, since Fernald developed its on- site estimate,
the site has added considerable costs to implement waste acceptance
oversight activities, in response to stakeholder concerns. These increases
in on- site disposal costs cannot be compared to any rigorous analysis of
off- site disposal costs, however, because the sites dismissed off- site
disposal alternatives several years ago.

Since the three sites made their cost comparisons, some off- site disposal
fees have decreased and volume discounts might be available for the higher
waste volumes now projected. The three sites relied upon the best
available-- though preliminary-- information and assumptions in preparing
their original off- site cost estimates. For off- site disposal fees, the
sites relied on historical rates, such as those in DOE?s existing contract
with Envirocare. Their estimates for off- site disposal ranged from $242 to
$312 per cubic meter of waste disposed. 7 Such fees change over time, and
the sites? estimates now appear unrealistically high, when compared with
current fees for off- site disposal at Envirocare. That company now prices
disposal of bulk rail shipments of soils classified as low- level wastes for
as low as $180 per cubic meter. In addition, DOE?s year 2000 contract with
Envirocare provides for significant discounts- a price drop from $519 to
$176 per cubic meter- for disposal of specified shipments of debris.
Envirocare officials told us that, because the historical DOE contract rates
for disposal of soils and debris had been negotiated for relatively small
waste volumes, additional volume discounts might be available for the larger
volumes of soil and debris now projected by the sites.

For their off- site cost estimates, site officials also used rail
transportation rates that appeared high in some cases, but they have not
revisited transportation options. DOE had little historical data on rail
costs for lowlevel radioactive waste shipments, and each site used a
different approach to estimate these costs. Because of the preliminary
nature of the cost estimates, site officials made simplified assumptions
about shipping configurations and rates. However, once they had better
information regarding the amounts and timetables for waste disposal,
officials did not fully reconsider alternative configurations or schedules
to determine whether rail costs could be reduced. For example, they did not
attempt to adjust rail costs for possible use of ?dedicated? trains. At
Fernald, dedicated trains now carry waste that is not qualified for on- site
disposal directly to the Envirocare facility. These trains make fewer stops
and

7 For comparison, historical rates were converted to year 2000 dollars. Off-
Site Disposal Costs Could

Decrease

Page 15 GAO- 01- 441 On- site Waste Disposal

complete the trip in much less time. If DOE rents rail cars by the day, the
overall cost for a train dedicated to low- level cleanup waste could be
considerably less. Envirocare officials suggested that further savings were
possible if DOE would consider proposals that bundle the rail transportation
and disposal services into one package agreement. These officials stated
that they have negotiated similar agreements with other customers.

Good business practice suggests that early cost comparisons that are
susceptible to uncertainties should be updated before major capital
investments are made. This concept is embedded in recent OMB guidance that
advocates such revalidation of planning estimates for capital investment
decisions. OMB seeks to improve agency planning, budgeting for, and
acquiring capital assets through guidance issued in Circular A- 11, Part 3.
This guidance states that agencies should make effective use of competition
and consider alternative solutions. In this instance, the competition is
between disposal options as well as potential contractors. For these sites,
competition between on- site and off- site disposal options could provide
several incentives. First, it provides an incentive to keep on- site
disposal costs as low as possible. If off- site disposal is eliminated
completely as an option, sites have less incentive to ensure that on- site
disposal plans are as economical as possible. Second, it provides incentives
for off- site disposal facility contractors to reduce rates and create more
competition with on- site disposal.

OMB?s 1997 supplement to Part 3 of Circular A- 11, the Capital Programming
Guide, provides even more definitive guidance. It states that once a capital
project has been funded, an agency?s first action is to validate that the
planning phase decision is still appropriate. It further states that,
because a year or more can elapse between the planning decision and
commitment, agencies should review their needs and the capabilities of the
market. DOE?s own order implementing this guidance, issued in October 2000,
calls for independent review of cost estimates and verification of mission
need prior to final approval for construction funding. However, the order
does not require the sites to re- validate, using independent reviews, the
cost comparisons between on- site and off- site disposal alternatives.

Once site officials have refined their disposal project scope to the point
where they can request contract proposals for construction, it appears
reasonable for them to consider ways that the off- site disposal services
market could compete with the on- site proposals. The CERCLA process allows
for selection of acceptable alternatives when the business Good Business
Practice

and Federal Guidance Suggest Reevaluation of Disposal Options

Page 16 GAO- 01- 441 On- site Waste Disposal

environment changes, as long as these alternatives satisfy the regulatory
standards for the cleanup. Moreover, the three sites left open the
possibility for changes in their selected remedies. For example, the INEEL
ROD calls for further evaluation of cost effectiveness of on- site or off-
site disposal prior to excavation of contaminated areas, but does not
specify that this should occur prior to major construction phases. EPA?s
CERCLA guidelines specifically address how agencies need to document changes
they make from the alternative selected in the ROD. In some of EPA?s
examples, the guidelines suggest that large increases in the waste volumes,
disposal costs, or a change in disposal location from on- site to off- site,
should be documented in an Explanation of Significant Difference. EPA?s
guidelines state that more fundamental changes, such as the discovery that
additional costly waste treatment will be needed prior to disposal, may
require an amendment to the ROD that must reconsider the nine criteria and
invite public comments. Both examples show that the built- in flexibility of
the CERCLA process accommodates more cost- effective business decisions as
well as improved cleanup technologies.

Changes in both on- site and off- site cost assumptions mean that the
balance of costs and risks at each site may now be much different than when
the comparisons were made. As a result, updated comparisons may show that,
on a cost basis alone, off- site disposal is now a much more competitive
alternative. However, because cost is only one factor that is considered
when making disposal decisions, off- site disposal costs do not necessarily
need to drop below on- site disposal costs for off- site disposal to emerge
as the better alternative. To determine the relative advantages of the two
alternatives, officials must also assess their respective longterm risks,
the stewardship activities that will address these risks, and the estimated
costs of these activities. These long- term stewardship risks are highly
uncertain. As the gap between on- site and off- site disposal costs narrows,
this uncertainty becomes relatively more significant to the balancing among
CERCLA criteria. The elapsed time from the ROD until bidding and
construction of an on- site disposal facility argues for DOE sites to use
current information and ensure that the balance of cost and long- term risk
remains favorable.

Changes in cost assumptions for off- site disposal indicate considerable
potential for narrowing the cost gap between these disposal alternatives. Of
the three sites, only Oak Ridge has updated its off- site cost analysis to
reflect more recent circumstances or volume discounts, and even this
estimate has been superceded by additional developments. Table 5 shows how
much the gap between on- site and off- site disposal closed when offsite
estimates were adjusted to reflect changes in commercial prices for Changes
in Cost Could

Greatly Affect Earlier Balance of Costs and Risks

Comparison Updates Substantially Narrow Cost Gap

Page 17 GAO- 01- 441 On- site Waste Disposal

some off- site disposal fees and transportation costs, and in one case,
changes in waste- type.

Table 5: The Effect of Current Prices for Low- level Waste Disposal on DOE?s
Off- site Estimates Estimates made by the site Off- site update made by GAO

Site Estimate for on- site disposal Estimate for off- site

disposal Gap between onsite

and off- site estimates Amount of

estimate a Decrease in gap

Percentage decrease in gap

Fernald b $849 million $1,126 million $277 million (33% higher)

$1,026 million $100 million 36 % Oak Ridge c $294 per cubic yard e $770 per
cubic yard $476 (162 %

higher ) $526 per cubic yard $244 per

cubic yard 51 % INEEL d $236 million $713 million $477 million

(202% higher) $610 million $103 million 22 % a GAO recomputed DOE?s off-
site estimates using the sites? own cost formulas, and substituting DOE?s
current contract price for commercial disposal of low- level bulk soil. For
INEEL, rail transportation rates were also updated. For Oak Ridge, in
addition to updating low- level disposal rates and transportation rates, the
proportion of hazardous waste was lowered to reflect the site?s current
estimate. b Amounts are in constant 1995 dollars.

c Amounts are in constant 1999 dollars. d Amounts are in constant 1998
dollars. e Unit prices are from Oak Ridge?s high volume estimates, which
represent the lowest unit costs.

When on- site and off- site disposal costs become more comparable, other
factors begin to assume increased significance. Among these factors is the
issue of retaining the waste on site, where it will pose a potential threat
to human health and the environment, for all practical purposes, forever.
The sites have attempted to incorporate the costs of long- term stewardship
into their on- site estimates, but these cost estimates are based on
extremely limited information.

Expected long- term stewardship costs are uncertain for several reasons.
First, the sites develop these estimates before specific plans are drawn up
for protecting the waste. Second, there is little historical information on
which to base the preliminary estimates, because DOE has closed very few
sites. Finally, the preliminary estimates at the three sites did not appear
to provide any contingency amounts for non- routine problems that might
arise, and some long- term issues are open- ended. For example, the
postclosure plan for the Fernald site, issued in May 1997, states that the
postclosure leachate collection and monitoring must continue until leachate
is Long- term Risk and Cost

Uncertainties Become More Significant

Page 18 GAO- 01- 441 On- site Waste Disposal

no longer detected or ceases to pose a threat, with no mention of how long
that might be.

These limitations are likely to persist. In its October 2000 report on
longterm stewardship, DOE states: ?Given the limitations of available data,
considerable uncertainty will be associated with any long- term stewardship
cost estimates.? In another recent study, the National Research Council
noted that long- term stewardship cost estimates have significant
uncertainties due to controversies over such matters as discount rates and
hidden costs. DOE is in the process of developing standardized guidance for
estimating long- term stewardship costs, and anticipates that sites will
include such estimates in their fiscal year 2003 budget process.

DOE is also examining alternative financing approaches for long- term
stewardship. However, these approaches may not adequately cover the
potentially high costs associated with any disposal facility failure and the
consequent release of contamination into the environment. Furthermore,
alternative financing may not be sufficient to cover all of the estimated
post- closure costs. For example, according to site officials, the Oak Ridge
site and the Tennessee Department of Environment and Conservation entered
into an administrative agreement (Consent Order) to establish the Tennessee
Perpetual Care Investment Fund. The Consent Order requires DOE to annually
deposit $1 million into the fund for 14 years. The state will use fund
income to cover costs of annual post- closure surveillance and maintenance
of the disposal facility. Site officials had previously estimated these
annual costs would range from about $684, 000 to about $922,000 in year 2000
dollars. To generate income in this range, the fund principal- which is
equivalent to about $11.3 million in year 2000 dollars- will need to earn an
average return of roughly 6 to 8 percent annually. Considering that the
average real treasury rate over the past decade was about 3. 6 percent, the
fund may not generate enough income to cover estimated post- closure costs.

Site officials pointed out that uncertainties surrounding long- term
stewardship costs also affect the Envirocare facility. Envirocare maintains
a trust fund, as required by Utah state rules implementing Nuclear
Regulatory Commission requirements, to cover future closure and longterm
stewardship costs in case the firm goes out of business. Under CERCLA,
according to site officials, the federal government, which disposes of large
quantities of waste at the Envirocare facility, would probably be liable in
the event that these funds were insufficient. In our view, however, this
point does not diminish the importance of evaluating

Page 19 GAO- 01- 441 On- site Waste Disposal

the risk for on- site disposal. For several reasons, potential increases in
stewardship costs to DOE at the Envirocare facility are less likely than at
the planned on- site disposal facilities, especially those in wetter
climates. First, the Envirocare facility is located in a dry climate, which
would restrict movement of contaminants from the facility to the underlying
groundwater. Second, the groundwater beneath the site is not suitable for
human consumption or even for watering livestock because of its high mineral
content. Finally, the facility is in a location that is remote from
population centers.

The CERCLA decision process, culminating with the ROD, represents planning
and agreement for remediation activities at the three sites. After the ROD
is signed, project assumptions and timeframes are subject to change for an
extended period, allowing DOE sites time to confirm their earlier
conclusions that on- site disposal remains advantageous despite long- term
cost and risk uncertainties. DOE sites could validate the early cost
comparisons by re- estimating the off- site disposal costs using current
disposal and transportation prices combined with baseline assumptions (about
waste volumes and characteristics, for example) for the proposed on- site
disposal facility. Another approach would be to solicit proposals for off-
site disposal along with proposals requested for construction of an on- site
facility. Generally, DOE sites plan to award several contracts over the life
of the disposal project, each covering a specific construction phase. For
example, Fernald site officials expect the final disposal facility to
consist of 6 to 8 sub- units called cells. As of November 2000, the site has
awarded three separate construction contracts covering construction for
various phases of three cells. At Oak Ridge, the baseline budget for the on-
site facility calls for two construction phases, with the second phase
proceeding in six expansion steps. INEEL officials have stated that their
planned on- site disposal facility may be expanded in a second phase to
accommodate the large quantity of waste generated after its chemical plant-
located adjacent to the on- site facility- is dismantled after 2035. Site
officials stated that they will re- evaluate cost effectiveness at that time
in accordance with ROD requirements. When sufficient time elapses between
such contract phases, DOE could benefit from reevaluating the market for
off- site disposal at each phase. Such competition could provide incentives
for both on- site and off- site proposals to be as economical as possible.
Once the DOE sites have these ?real world? estimates in hand, they would be
in a better position to evaluate the extent to which cost savings for on-
site disposal continue to balance the longterm uncertainties. DOE Should Use
Current

Information to Validate Planning Decisions

Page 20 GAO- 01- 441 On- site Waste Disposal

Unless DOE revisits its disposal needs and its current options for disposing
of wastes off- site, it could miss opportunities to reduce cleanup costs at
the three sites and at other sites, such as Paducah, that might propose the
development of new on- site facilities. Building in a decision checkpoint
before major investment decisions are finalized could identify instances
when the use of off- site disposal would be less expensive, or when the cost
difference no longer outweighs the long- term risks associated with on- site
disposal. Such validation of the cost comparison is especially important in
instances where DOE is aware that the scope or timeframe of the cleanup
effort has changed dramatically. Remaining open to new proposals for off-
site disposal would also inject an element of competition into this process.
Thus, even if the validation did nothing more than confirm the original
decision to dispose of the wastes on- site, it has the potential to ensure
that costs are kept to a minimum.

We recommend that, before constructing new or expanding existing facilities
for disposal of cleanup waste at the Fernald, INEEL, and Oak Ridge sites,
the Secretary of Energy revisit the cost comparisons for onsite and off-
site disposal to determine if the cost estimates used to support the ROD
remain valid. If cost advantages for on- site disposal have decreased, the
Secretary should reassess whether expected cost savings from on- site
disposal facilities outweigh the long- term risks associated with these
proposed disposal facilities. We also recommend that DOE validate cost
comparisons at any other sites that may decide to develop an on- site
disposal facility.

We provided a draft of this report to DOE for review and comment. DOE
generally agreed with the report?s conclusion and recommendation that
assumptions used to select on- site disposal need to be re- validated before
constructing or expanding on- site disposal facilities. DOE pointed out that
reassessments are already planned for the disposal cell at the INEEL site in
Idaho, which is currently in an early design phase. The Department also
stated that it will consider whether to revisit plans to proceed with
expansion of existing or construction of new disposal facilities as part of
a comprehensive assessment of its Environmental Management program.

Appendix III presents DOE?s comments on the report. DOE also suggested
several technical clarifications which we have incorporated into the report
as appropriate. DOE?s technical comments included the observation that
another factor to be considered when evaluating off- site disposal is the
receiving facility?s capacity to accommodate incoming waste volumes. GAO
agrees that the coordination of multiple waste shipments to an off
Conclusion

Recommendation Agency Comments

Page 21 GAO- 01- 441 On- site Waste Disposal

site facility would be a challenge that would need to be addressed during
any contract negotiations.

We performed our review at DOE?s Fernald, INEEL, and Oak Ridge sites. We
interviewed DOE and contractor officials at each site who are familiar with
the sites? decisions to develop on- site disposal facilities. To understand
how site officials evaluated disposal alternatives, we reviewed each site?s
Record of Decision, Feasibility Study and other supporting documentation. To
determine the extent of EPA and state participation in the decision process,
we interviewed officials from regional EPA offices and state environmental
agencies that reviewed and concurred with DOE?s decision at each site. We
also reviewed pertinent legislation and implementing regulations and
guidance on disposal of radioactive and hazardous wastes, including planning
for capital investments in new disposal facilities, and discussed waste
disposal issues with officials at DOE headquarters and at the Defense
Nuclear Facility Safety Board. To evaluate off- site disposal alternatives
used for comparison at each site, we obtained and reviewed information on
DOE?s use of the Envirocare commercial disposal facility, and interviewed
officials of that company to assess the availability of commercial
facilities that dispose of low- level radioactive wastes. We also determined
the extent to which DOE?s cost comparisons depended upon the rates assumed
for off- site transportation and commercial disposal fees. (See app. II for
a further discussion of our scope and methodology.) We conducted our review
from May 2000 through May 2001 in accordance with generally accepted
government auditing standards.

This report contains a recommendation to you. As you know, 31 U. S. C. 720
requires the head of a federal agency to submit a written statement of the
actions taken on our recommendations to the Senate Committee on Governmental
Affairs and to the House Committee on Government Reform not later than 60
days from the date of this letter, and to the House and Senate Committees on
Appropriations with the Agency?s first request for appropriations made more
than 60 days after the date of this letter. Scope and

Methodology

Page 22 GAO- 01- 441 On- site Waste Disposal

Copies of this report are available on request. If you or your staff have
any questions on this report, please call me at (202) 512- 3841. Key
contributors to this report are listed in appendix IV.

Sincerely yours, (Ms.) Gary L. Jones Director, Natural Resources

and Environment

Appendix I: DOE Sites? Analyses of the Primary Tradeoffs Between On- Site
and OffSite Disposal Alternatives

Page 23 GAO- 01- 441 On- site Waste Disposal

To select a cleanup alternative, officials at the Fernald, Oak Ridge and
INEEL sites weighed the various cleanup approaches and made tradeoffs
according to the site?s unique conditions and priorities. Using CERCLA?s
five balancing criteria (see Table 6), site officials compared the
advantages and disadvantages of their on- site and off- site disposal
alternatives. Their analyses relied on site- specific information developed
in their feasibility studies, and varied in depth according to the
availability of data and the importance of each criterion at the site. Each
site issued a Proposed Plan that summarized the comparative analysis and
designated on- site disposal as the preferred alternative for the cleanup
approach. After considering public comments on the Proposed Plan, each of
the three sites issued a Record of Decision selecting an on- site disposal
approach. The following brief summaries describe each site?s analysis of the
primary tradeoffs that were considered between its on- site and off- site
alternatives of similar cleanup scope.

Table 6: CERCLA Balancing Criteria CERCLA Balancing Criteria Description

Long- term effectiveness and permanence Risk to human health and the
environment from exposure to contaminants remaining after site closure
Reduction of toxicity, mobility or volume through treatment Extent to which
the alternative uses treatment technologies or processes

where possible Short- term effectiveness Duration of site cleanup and risks
to human health and the environment

from exposure to contaminants or from cleanup activities Ease of
Implementation Technical, logistical and administrative ease or difficulty
of construction

and operations Cost Cost effectiveness with life cycle costs estimated at
-30 to +50 percent

According to the Fernald site?s 1995 Proposed Plan, officials preferred the
on- site disposal alternative after determining that this approach: 1) was
reliable over the long term, 2) offered the lowest overall short- term
risks, 3) was less costly in comparison to other alternatives, and 4)
employed technologies that could be implemented. Although officials
concluded that on- site disposal was reliable over the long term, their
comparative analysis showed that the off- site alternative held an advantage
for longterm effectiveness. This analysis pointed out that off- site
disposal left the least amount of contamination at the site and did not
require engineering and institutional controls to be reliable over the long
term. In contrast, any on- site disposal facility at Fernald would need a
design that ensured protection of the Great Miami Aquifer for thousands of
years. Furthermore, Ohio?s solid waste disposal restrictions prohibit
building such a landfill over the aquifer, which was designated as a sole
source Appendix I: DOE Sites? Analyses of the

Primary Tradeoffs Between On- Site and OffSite Disposal Alternatives

Fernald

Appendix I: DOE Sites? Analyses of the Primary Tradeoffs Between On- Site
and OffSite Disposal Alternatives

Page 24 GAO- 01- 441 On- site Waste Disposal

aquifer under the Safe Drinking Water Act. EPA and the Ohio EPA agreed to
waive this restriction if the proposed on- site facility could be designed
to meet equivalent safety standards.

To apply CERCLA criteria to the Fernald site, officials weighed the longterm
advantage of disposing of all waste off- site against disadvantages of this
approach, some of which were of significant concern to various stakeholders.
These disadvantages appeared under three CERCLA criteria:

 Site officials judged short- term risks for the off- site disposal option
to be higher overall based on increased risks associated with shipping large
quantities of waste by rail across country. Officials quantified the
increased transportation risks for the comparable off- site alternative in
their site?s feasibility study as approximately 10 injuries and 3 fatalities
(for approximately 20, 000 rail cars travelling to Utah and back).

 The site?s comparison of life cycle costs showed that cleanup approaches
depending mainly on off- site disposal were more expensive than approaches
with an on- site disposal facility. Its detailed comparison of alternatives
showed that, for disposal of similar waste volumes, the estimated cost for
off- site disposal was 34 percent more than the on- site estimate. In their
proposed plan, site officials noted that the accuracy of the cost estimates
typically varied between -30 to +50 percent because of underlying
uncertainties in the available information used to develop them. Site
officials stated that other criteria, particularly the plan?s
implementability and community concerns about off- site rail transportation,
played a more significant role in the site?s final decision.

 Site officials questioned whether the off- site alternative could be
successfully implemented if off- site disposal facilities became unavailable
over the projected 22- year duration of the cleanup. Furthermore, they
feared that opposition to shipping large volumes of radioactive waste to
western states could hinder Fernald?s access to off- site disposal for its
more concentrated wastes, which cannot safely remain at the Fernald
location.

The Oak Ridge site?s proposed plan (January 1999) stated a preference for
the on- site disposal alternative after showing that on- site disposal
offered comparable protection at lower cost and less transportation risk
than its off- site alternative. The plan noted that the cost advantage was
only significant for estimates that used the high end of the projected range
of the anticipated waste volumes requiring disposal. Similar to the Fernald
plan, the Oak Ridge plan also notes that concerns by states receiving the
Oak Ridge

Appendix I: DOE Sites? Analyses of the Primary Tradeoffs Between On- Site
and OffSite Disposal Alternatives

Page 25 GAO- 01- 441 On- site Waste Disposal

wastes for off- site disposal could hinder access to off- site disposal for
large volumes of waste from the Oak Ridge Reservation.

Site officials concluded that an on- site disposal facility would provide
adequate long- term protection if engineering barriers were designed to
contain waste indefinitely. To ensure the long- term integrity of the
facility, they adopted the following three strategies to: 1) design the
disposal facility to meet or exceed long term safety requirements, 2) limit
the level of contamination allowed in the facility so that any leaks would
pose no unacceptable risks, and 3) provide for long- term monitoring and
facility maintenance. The facility?s design also addresses the need to
provide groundwater protection equivalent to that required for landfills
under the Toxic Substances Control Act of 1976. That act, as implemented by
federal regulations, requires the bottom of a landfill liner to be 50 feet
above the historical high groundwater table. Based on the protection
afforded by the facility?s location and design (predominantly aboveground),
EPA agreed to waive this technical requirement.

Unlike the Fernald site, where the cleanup is expected to render most of the
site accessible to the public, the Oak Ridge Reservation expects to restrict
public access to many areas indefinitely and leave significant contamination
on the site, including areas near the proposed on- site facility location.
For various technical and safety reasons, DOE does not plan to excavate
these areas. As a result, some contaminated areas around the Oak Ridge site
will pose long- term risks regardless of whether an onsite disposal facility
is constructed. Site officials performed a site- wide (composite) analysis
of health risks, and estimated that the radiation from the proposed on- site
facility would amount to approximately 1. 1 millirem per year (after 1000
years). This amount represents roughly one- quarter of the estimated
radiation dose from all sources within Bear Creek Valley after remediation,
and according to site officials, is well within the established values for
protection of human health and the environment.

Along with their conclusion that on- site disposal provided comparable
protection to the off- site alternative, site officials found that two other
CERCLA criteria gave the advantage to the on- site alternative:

 The comparison of estimated costs for on- site and off- site disposal
showed that on- site disposal cost significantly less only under the high
volume scenario. This high volume scenario envisioned more extensive site-
wide cleanup at the Oak Ridge Reservation than DOE?s baseline assumptions.
By the time the ROD was issued in November 1999, site officials considered
the high volume scenario to be the most realistic and selected

Appendix I: DOE Sites? Analyses of the Primary Tradeoffs Between On- Site
and OffSite Disposal Alternatives

Page 26 GAO- 01- 441 On- site Waste Disposal

the on- site disposal alternative based, in part, on cost comparisons
estimated for the higher waste volumes.

 Based on calculations in their feasibility study, site officials concluded
that the on- site disposal alternative had significantly less transportation
risk than the off- site disposal alternative. The feasibility study reported
that the risk of transportation accident- related injuries or fatalities was
highest for off- site scenarios that used trucks (111 injuries and 10
fatalities). For rail transport of the high- end waste volume to the off-
site facility in Utah, the risks were 8. 2 injuries and .07 fatalities,
compared to 0.41 injuries and 0.003 fatalities for the small number of rail
shipments required for the onsite alternative. According to the study, the
risks from radiological exposure during transportation were very small for
either alternative.

The INEEL proposed plan (October 1998) proposed on- site disposal as the
preferred alternative, stating that the on- site approach ensures long- term
protection of human health and the environment, complies with applicable
legal requirements, and is a permanent and cost- effective solution.
According to the summary comparative analysis, three criteria differentiated
between on- site and off- site disposal alternatives: short term
effectiveness, implementability, and cost. The proposed plan does not
differentiate between the long- term effectiveness for on- site and off-
site disposal. It concludes that, when compared to alternatives that capped
waste in place, the two cleanup approaches provided equivalent long- term
protection because each excavated contaminated soils and disposed of them in
an engineered disposal facility- regardless of the facility?s location. The
plan, and the subsequent ROD issued one year later, further noted that the
on- site disposal facility would be designed to protect groundwater quality
in the subterranean Snake River Plain Aquifer, as well as to prevent
external exposure to radiation. Similar to the analysis by Oak Ridge
officials, INEEL officials relied upon the adequacy of the facility?s
design, as well as other strategies intended to maintain protectiveness over
the long- term, to reach its conclusion that on- site disposal is as
protective as off- site disposal in the long- term.

When site officials evaluated three other CERCLA criteria, they found that
the off- site disposal alternative had the following disadvantages when
compared with the on- site alternative:

 In the short term, officials found that both on- site and off- site
disposal alternatives posed minor risks to workers or the environment, and
that the off- site alternative posed an additional minor risk to
communities. The site?s feasibility study stated more specifically that the
off- site alternative would pose some increased risk to communities from
transport and INEEL

Appendix I: DOE Sites? Analyses of the Primary Tradeoffs Between On- Site
and OffSite Disposal Alternatives

Page 27 GAO- 01- 441 On- site Waste Disposal

potential railroad accidents. However, the study further noted that the rail
lines passed through very rural communities, and stated that potential risk
should be minimal.

 In the proposed plan, site officials concluded that the off- site disposal
alternative would be the most difficult to implement because it would
require the transport of ?large volumes of contaminated soils great
distances and depends upon the availability of off- site disposal
capability.? The feasibility study did not provide support for this concern,
and stated that ?off- site disposal? has been previously performed;
therefore this alternative should be administratively feasible.?

 In their proposed plan INEEL officials concluded that the off- site
disposal alternative was the most expensive. They compared the estimated
costs for excavation and disposal of 63, 000 cubic meters of waste projected
for the cleanup area under the Record of Decision. The off- site estimate
was $221 million, 160 percent more costly than on- site estimate of $85
million. In the proposed plan, officials also noted that the on- site
disposal facility would be constructed to accept contaminated cleanup
materials from sites located throughout the INEEL site. They estimated that
off- site disposal for the projected 356,000 cubic meters of site- wide
waste would cost 224 percent more than an on- site alternative ($ 605
million versus $187 million). Site officials stated that they developed
their site- wide cost estimates by modifying the original estimates for
63,000 cubic meters.

Appendix II: Scope and Methodology Page 28 GAO- 01- 441 On- site Waste
Disposal

In February 2000, DOE adopted a new policy allowing all DOE sites to dispose
of low- level and mixed radioactive wastes at its facilities located at the
Nevada Test Site and the Hanford Reservation. Sites can also use commercial
off- site disposal facilities under certain circumstances. DOE?s policy was
aimed at containing low- level and mixed wastes generated from its past or
ongoing operations. However, the Department expects to generate
significantly larger quantities of low- level and mixed wastes from its
cleanup operations. In 1996, the Hanford site opened a facility for disposal
of its on- site cleanup wastes under the CERLA program. Since 1996, three
other DOE sites have made decisions to develop new, on- site disposal
facilities for their low- level cleanup wastes governed by CERCLA, and are
in various stages of planning, constructing, and filling these facilities.
These sites are: the Fernald Environmental Management Project (Ohio); the
Oak Ridge Reservation (Tennessee); and the Idaho National Engineering and
Environmental Laboratory (INEEL) (Idaho). Plans for the new facilities at
these sites entail permanent on- site disposal of significant quantities of
wastes that would otherwise qualify for disposal off- site under DOE?s
policies. 1 We reviewed the sites? decisions to determine (1) the extent
that site officials considered the comparative costs and risks of off- site
disposal options and (2) the extent that site officials revisited these cost
and risk assessments as circumstances warranted. In addition, at least one
other site, the Paducah Gaseous Diffusion Plant (Kentucky), is currently
considering proposals to develop a new on- site facility. Our review covered
the decisions already made at Fernald, Oak Ridge, and INEEL. We did not
review the decision at Hanford because DOE?s recent policy designates
Hanford as one of two preferred sites for acceptance of DOE- wide low- level
wastes.

We visited the three sites to observe the locations of the new disposal
facilities and to determine what alternatives, if any, each site considered
for disposal of their cleanup wastes. We interviewed site officials and
reviewed decision documents to determine the factors that each site
considered, including risks and costs of various disposal alternatives. We
also interviewed officials from the state and Environmental Protection
Agency offices that reviewed and concurred with DOE?s decision at each site.

1 DOE?s recent policy allows the Oak Ridge and INEEL sites to continue
disposal of their low- level wastes on- site to the extent practical.
However, existing facilities at the two sites have no capacity for the
projected quantities of cleanup wastes. Appendix II: Scope and Methodology

Appendix II: Scope and Methodology Page 29 GAO- 01- 441 On- site Waste
Disposal

To understand Departmental and legal influences for the sites? waste
disposal decisions, we consulted legislative and executive guidance on
radioactive waste disposal and capital investment planning. We also
interviewed federal officials at DOE headquarters as well as the Defense
Nuclear Facility Safety Board.

In order to determine current off- site disposal prices for low- level
radioactive wastes, we reviewed information on recent uses of commercial
disposal by various DOE sites. We also reviewed DOE?s disposal contracts
with Envirocare and interviewed company officials.

We conducted a limited analysis to determine the extent that each site?s
cost comparison depended upon the rates used for off- site transportation
and commercial disposal fees. To illustrate how much the gap between on-
site and off- site disposal estimates can close when off- site rates are
adjusted to reflect changes in commercial prices (and in one case, changes
in projected waste- type), we adjusted off- site costs as follows:

 For Fernald and INEEL, we substituted the latest contract prices for
disposing of low- level bulk soil waste off- site in place of the rates used
by the sites? for low- level waste in their original estimates for cost
comparison. (Neither Fernald nor INEEL had an updated version of the off-
site estimate that we could have used to compare to current on- site
estimates.) For INEEL, we also substituted transportation rates that were
more in line with current prices. This exercise decreased the difference
between on- site and off- site disposal costs by 36 percent at Fernald and
22 percent at INEEL.

 For Oak Ridge, we used the site?s most recent cost comparison analysis,
and substituted updated estimates of the type of wastes, as well as current
prices for low- level waste disposal and commercial estimates of
transportation rates. When Oak Ridge officials prepared their most recent
off- site estimate in 1999, they assumed that 44 percent of the waste would
be classified as hazardous for off- site disposal. They have since revised
the figure to less than 1 percent. The combined effect of reducing the
proportion of hazardous waste and applying the lower contract and
transportation prices decreased the gap between on- site and off- site
disposal cost estimates by 51 percent.

We conducted our review from May 2000 through May 2001 in accordance with
generally accepted government auditing standards.

Appendix III: Comments From the Department of Energy

Page 30 GAO- 01- 441 On- site Waste Disposal

Appendix III: Comments From the Department of Energy

Appendix IV: GAO Contacts and Staff Acknowledgements

Page 31 GAO- 01- 441 On- site Waste Disposal

Dwayne Weigel, (202) 512- 6876 Patricia Rennie, (206) 287- 4843

In addition to those named above, John Cass, Linda Chu, Christine Colburn,
Daniel Feehan, Hova Risen- Robertson and Stan Stenerson made key
contributions to this report. Appendix IV: GAO Contacts and Staff

Acknowledgements GAO Contacts Acknowledgements

Related GAO Products Page 32 GAO- 01- 441 On- site Waste Disposal

Low- Level Radioactive Wastes: Department of Energy Has Opportunities to
Reduce Disposal Costs (GAO/ RCED- 00- 64 , Apr. 12, 2000).

Low- Level Radioactive Wastes: States Are Not Developing Disposal Facilities
(GAO/ RCED- 99- 238, Sept. 17, 1999).

Nuclear Waste: DOE?s Accelerated Cleanup Strategy Has Benefits but Faces
Uncertainties (GAO/ RCED- 99- 129, Apr. 30, 1999).

Nuclear Waste: Corps of Engineers? Progress in Cleaning Up 22 Nuclear Sites
(GAO/ RCED- 99- 48, Feb. 26, 1999).

Department of Energy: Alternative Financing and Contracting Strategies for
Cleanup Projects (GAO/ RCED- 98- 169, May 29, 1998).

Radioactive Waste: Interior?s Continuing Review of the Proposed Transfer of
the Ward Valley Waste Site (GAO/ RCED- 97- 184, July 15, 1997).

Department of Energy: Management and Oversight of Cleanup Activities at
Fernald (GAO/ RCED- 97- 63, Mar. 14, 1997).

Radioactive Waste: Status of Commercial Low- Level Waste Facilities (GAO/
RCED- 95- 67, May 5, 1995). Related GAO Products

(141454)

The first copy of each GAO report is free. Additional copies of reports are
$2 each. A check or money order should be made out to the Superintendent of
Documents. VISA and MasterCard credit cards are also accepted.

Orders for 100 or more copies to be mailed to a single address are
discounted 25 percent.

Orders by mail:

U. S. General Accounting Office P. O. Box 37050 Washington, DC 20013

Orders by visiting:

Room 1100 700 4 th St., NW (corner of 4 th and G Sts. NW) Washington, DC
20013

Orders by phone:

(202) 512- 6000 fax: (202) 512- 6061 TDD (202) 512- 2537

Each day, GAO issues a list of newly available reports and testimony. To
receive facsimile copies of the daily list or any list from the past 30
days, please call (202) 512- 6000 using a touchtone phone. A recorded menu
will provide information on how to obtain these lists.

Orders by Internet

For information on how to access GAO reports on the Internet, send an email
message with ?info? in the body to:

Info@ www. gao. gov or visit GAO?s World Wide Web home page at: http:// www.
gao. gov

Contact one:

 Web site: http:// www. gao. gov/ fraudnet/ fraudnet. htm

 E- mail: fraudnet@ gao. gov

 1- 800- 424- 5454 (automated answering system) Ordering Information

To Report Fraud, Waste, and Abuse in Federal Programs
*** End of document. ***