This is the accessible text file for GAO report number GAO-03-930T
entitled 'Nuclear Waste: Challenges and Savings Opportunities in DOE's
High-Level Waste Cleanup Program' which was released on July 17, 2003.
This text file was formatted by the U.S. General Accounting Office
(GAO) to be accessible to users with visual impairments, as part of a
longer term project to improve GAO products' accessibility. Every
attempt has been made to maintain the structural and data integrity of
the original printed product. Accessibility features, such as text
descriptions of tables, consecutively numbered footnotes placed at the
end of the file, and the text of agency comment letters, are provided
but may not exactly duplicate the presentation or format of the printed
version. The portable document format (PDF) file is an exact electronic
replica of the printed version. We welcome your feedback. Please E-mail
your comments regarding the contents or accessibility features of this
document to Webmaster@gao.gov.
This is a work of the U.S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed
in its entirety without further permission from GAO. Because this work
may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this
material separately.
Testimony:
Before the Subcommittee on Oversight and Investigations, Committee on
Energy and Commerce, House of Representatives:
United States General Accounting Office:
GAO:
For Release on Delivery Expected at 9:30 a.m. EDT,
Thursday, July 17, 2003:
NUCLEAR WASTE:
Challenges and Savings Opportunities in DOE's High-Level Waste Cleanup
Program:
Statement of Robin M. Nazzaro, Director:
Natural Resources and Environment:
High-Level Waste Cleanup:
GAO-03-930T:
GAO Highlights:
Highlights of GAO-03-930T, a testimony before the Subcommittee on
Oversight and Investigations, Committee on Energy and Commerce, House
of Representatives
Why GAO Did This Study:
The Department of Energy (DOE) oversees the treatment and disposal of
94 million gallons of highly radioactive nuclear waste from the
nation’s nuclear weapons program, currently at DOE sites in
Washington, Idaho, and South Carolina. In 2002, DOE began an
initiative to reduce the estimated $105-billion cost and 70-year time
frame of this cleanup. GAO was asked to testify on the status of this
initiative, the legal and technical challenges DOE faces in
implementation, and any further opportunities to reduce costs or
improve program management. GAO’s testimony is based on a report
(GAO-03-593) released at the hearing.
What GAO Found:
DOE’s initiative for reducing the costs and time required for cleanup
of high-level wastes is still evolving. DOE’s main strategy for
treating high-level waste continues to include separating and
concentrating much of the radioactivity into a smaller volume for
disposal in a geologic repository. Under the initiative, DOE sites are
evaluating other approaches, such as disposing of more waste on site.
DOE’s current savings estimate for these approaches is $29 billion,
but the estimate may not be reliable or complete. For example, the
savings estimate does not adequately reflect uncertainties or take
into account the timing of when savings will be realized.
DOE faces significant legal and technical challenges to realize these
savings. A key legal challenge involves DOE’s process for deciding
that some waste with relatively low concentrations of radioactivity
can be treated and disposed of on-site. A recent court ruling
invalidated this process, putting the accelerated schedule and
potential savings in jeopardy. A key technical challenge is that DOE’s
approach relies on laboratory testing to confirm separation of the
waste into high-level and low-activity portions. At the Hanford Site
in Washington State, DOE plans to build a facility before conducting
integrated testing of the waste separation technology—an approach that
failed on a prior major project.
DOE is exploring proposals, such as increasing the amount of
high-level waste in each disposal canister, that if successful could
save billions of dollars more than the current $29 billion estimate.
However, considerable evaluation remains to be done. DOE also has
opportunities to improve program management by fully addressing
recurring weaknesses GAO has identified in DOE’s management of cleanup
projects, including the practice of incorporating technology into
projects before it is sufficiently tested.
What GAO Recommends:
GAO made recommendations in the report on which this testimony is
based. In commenting on the report, DOE agreed to consider seeking
clarification from the Congress about its authority to decide that
waste with low concentrations of radioactivity could be treated and
disposed of on-site. DOE disagreed with the need to conduct integrated
testing of the Hanford waste separation technology and argued that its
existing actions are sufficient to support decision making with
rigorous analysis, test new technology before incorporating it into
projects, and pursue concurrent design/construction of complex
facilities. GAO disagrees and continues to believe that its
recommendations are warranted.
www.gao.gov/cgi-bin/getrpt?GAO-03-930T
To view the full product, click on the link above. For more
information, contact Robin M. Nazzaro at (202) 512-3841 or
nazzaror@gao.gov.
[End of section]
Mr. Chairman and Members of the Subcommittee:
We are pleased to be here today to discuss the Department of Energy's
(DOE) high-level waste cleanup program. DOE has about 94 million
gallons of highly radioactive nuclear waste from the nation's nuclear
weapons program. This waste is currently in temporary storage at DOE
sites in Washington, South Carolina, and Idaho. After investing more
than 20 years and about $18 billion, DOE acknowledged in February 2002
that the program to clean up its high-level waste was far behind
schedule, far over budget, and in need of major change. In 2002, DOE
began an initiative to reduce the program's nearly $105-billion
estimated cost and 70-year time frame to finish permanent disposal of
this waste. Our testimony, based on work included in the report being
released by the Subcommittee today,[Footnote 1] discusses (1) the
components of DOE's high-level waste and the process involved in
preparing the waste for disposal, (2) the status of DOE's accelerated
cleanup initiative for high-level waste, (3) legal and technical
challenges DOE faces in implementing the initiative, and (4) further
opportunities to reduce costs beyond those identified in DOE's current
cost-savings proposal or to improve program management.
In summary, we found the following:
* DOE's high-level waste has many components, ranging from radioactive
isotopes and corrosive chemicals to the water in which much of this
material was initially discharged. The radioactive components vary
greatly; a small portion will remain dangerously radioactive for
millions of years, while the vast majority will lose much of their
radioactivity more quickly, so that more than 90 percent of the current
radioactivity will be gone within 100 years. To prepare the waste for
permanent disposal, DOE plans to separate the waste into two waste
streams: one with high levels of radioactivity and the other with lower
concentrations of radioactivity. DOE expects that this process will
concentrate at least 90 percent of the radioactivity into a volume that
is significantly smaller than the current total volume of waste. DOE
plans to immobilize and bury the highly radioactive portion in a
permanent underground repository. The remaining waste will be
immobilized and disposed of at the location where it is currently
stored or at some other location.
* DOE's initiative to accelerate the cleanup is evolving, and while its
savings estimates are changing accordingly, we have ongoing concerns
about the reliability of those estimates. As of April 2003, DOE
estimated it could shorten the waste cleanup schedule by 20-35 years
and save up to $29 billion. To help achieve these schedule and cost
reductions, DOE has identified alternative treatment and disposal
strategies, such as developing ways to permanently dispose of more of
the radioactive waste at current sites rather than moving it to the
planned underground repository. However, our assessment of DOE's
savings estimate indicates that it may not be reliable. For example,
the savings analysis does not take into account all costs associated
with alternative treatment strategies. Also, the estimate of savings
does not compare costs on the basis of "present value," where dollars
to be saved in future years are discounted to a common year to reflect
the time value of money. At DOE's Savannah River Site in South
Carolina, such an adjustment would lower the savings estimate for
accelerated waste processing by $2.6 billion--from $5.4 billion to
$2.8 billion (in 2003 dollars).
* DOE faces significant legal and technical challenges to realize the
estimated savings. A key legal challenge involves DOE's authority to
apply a designation other than high-level waste to some waste with
relatively low concentrations of radioactivity, so that this portion
can be treated less expensively than highly radioactive waste. A recent
court ruling invalidated this redesignation process, thus precluding
DOE from proceeding with this element of its accelerated initiative. If
DOE cannot meet its accelerated schedules, then potential savings are
in jeopardy. A key technical challenge is that DOE's approach relies
primarily on laboratory testing to confirm that separating waste into
high-level and low-activity portions will be successful. At the Hanford
Site in Washington State, DOE is planning to construct full-scale
facilities before fully testing the technologies on an integrated
basis--an approach that has failed on another project in the past,
resulting in significant cost increases and schedule delays.
* DOE is exploring additional cost savings beyond those identified in
its current cost-saving proposals. The proposals that offer significant
potential are being developed by the Hanford and Savannah River sites.
These proposals call for increasing the amount of waste that can be
concentrated into the canisters destined for the permanent underground
repository. DOE's data indicates that these proposals, if successful,
could save several billion dollars. Considerable evaluation of these
proposals remains to be done and cost-saving estimates have not yet
been fully developed, according to DOE officials. DOE also has
opportunities to improve its management of the cleanup program by
addressing management weaknesses that we and others have identified in
the past. Although DOE has taken steps to improve program management,
we have continuing concerns about management weaknesses in several
areas. These include making key decisions without rigorous supporting
analysis, incorporating technology into projects before it is
sufficiently tested, and pursuing a "fast-track" approach of launching
into facility construction before completing sufficient design work. It
does not appear that DOE's current management efforts will fully
address these weaknesses.
Our report makes several recommendations to DOE that, if implemented,
will help to manage or reduce legal and technical risks to the program,
avoid costly delays, and strengthen overall program management. DOE
agreed to consider our recommendation to seek clarification from the
Congress regarding its authority to determine that some waste can be
treated and disposed of as other than high-level waste. However,
regarding our recommendations that the department conduct integrated
pilot testing of its waste separation processes at Hanford, and take
steps to improve the management of high-level waste projects, such as
by conducting more rigorous analyses to support key project decisions,
DOE believes that its current approach is adequate. We do not agree
with DOE's views and continue to believe that all of our
recommendations are warranted.
Background:
DOE has a vast complex of sites across the nation dedicated to the
nuclear weapons program. DOE largely ceased production of plutonium and
enriched uranium by 1992, but the waste remains at the sites. Most of
the tanks in which the waste is stored have already exceeded their
design life. For example, many of Hanford's and Savannah River's tanks
were built in the 1940s to 1960s and were designed to last 10-40 years.
Leaks from some of these tanks were first detected at Hanford in 1956
and at Savannah River in 1959. Given the age and deteriorating
condition of some of the tanks, there is concern that some of them will
leak additional waste into the soil, where it may migrate to the
water table and, in the case of the Hanford Site, to the Columbia
River.
Responsibility for the high-level waste produced at DOE facilities is
governed primarily by federal laws, including the Atomic Energy Act of
1954. These laws established responsibility for the regulatory control
of radioactive materials including DOE's high-level waste and assigned
the Nuclear Regulatory Commission (NRC) the function of licensing
facilities that are expressly authorized for long-term storage of
high-level radioactive waste generated by DOE. In addition, the Nuclear
Waste Policy Act of 1982 defined high-level radioactive waste. Various
other federal laws, including the Resource Conservation and Recovery
Act of 1976, guide how DOE must carry out its cleanup program. The
high-level waste cleanup program is under the leadership of the
Assistant Secretary for Environmental Management. It involves
consultation with a variety of stakeholders, including the
Environmental Protection Agency, state environmental agencies where DOE
sites are located, county and local governmental agencies, citizen
groups, advisory groups, and Native American tribes.
DOE's High-Level Waste Is a Complex Mixture That Requires a Multi-Step
Process to Prepare for Disposal:
The waste in the tanks at the Hanford and Savannah River sites and the
Idaho National Laboratory near Idaho Falls is a complex mixture of
radioactive and hazardous components. DOE's process for preparing it
for disposal is designed to separate much of the radioactive material
from other waste components.
Much of the Radioactivity Declines Relatively Quickly:
Nearly all the radioactivity in the waste originates from radionuclides
with half-lives[Footnote 2] of about 30 years or less. The relatively
short half-lives of most of the radionuclides in the waste means that
within 30 years, about 50 percent of the current radioactivity will
have decayed away, and within 100 years this figure will rise to more
than 90 percent. Figure 1 shows the pattern of decay, using 2002 to
2102 as the 100-year period. Extending the analysis beyond the 100-year
period shown in the figure, in 300 years, 99.8 percent of the
radioactivity will have decayed, leaving 0.2 percent of the current
radioactivity remaining.
Figure 1: Natural Decay of Radionuclides in DOE's Untreated High-Level
Waste from 2002 to 2102:
[See PDF for image]
Note: Radioactivity is measured in a unit called a curie. One curie
equals 37 billion atomic disintegrations per second.
[End of figure]
Despite the relatively rapid decay of most of the current
radioactivity, some radionuclides have half-lives in the hundreds of
thousands of years and will remain dangerously radioactive for millions
of years. Some of these long-lived radionuclides are potentially very
mobile in the environment and therefore must remain permanently
isolated. If these highly mobile radionuclides leak out or are released
into the environment, they can contaminate the soil and water.
Processing Can Concentrate the Radioactivity into a Much Smaller Volume
of Waste:
DOE plans to isolate the radioactive components and prepare the waste
for disposal through a multi-step treatment process. DOE expects this
process to concentrate at least 90 percent of the radioactivity into a
much smaller volume that can be permanently isolated for at least
10,000 years in a geologic repository. The portion of the waste not
sent to the geologic repository will have relatively small amounts of
radioactivity and long-lived radionuclides. Based on current disposal
standards used by the NRC, if the radioactivity of this remaining waste
is sufficiently low, it can be disposed of on site near the surface of
the ground, using less complex and expensive techniques than those
required for the highly radioactive portion. DOE plans to dispose of
this waste on site in vaults or canisters, or at other designated
disposal facilities.
DOE has successfully applied this process in a demonstration project at
the West Valley site in New York State. At West Valley, separation of
the low-activity portion from the high-level portion of the waste
reduced by 90 percent the quantity of waste requiring permanent
isolation and disposal at a geologic repository. The high-level portion
was stabilized in a glass material (vitrified) and remains stored at
the site pending completion of the high-level waste geologic repository
and resolution of other issues associated with disposal costs.[Footnote
3] The remaining low-activity portion was mixed with cement-forming
materials, poured into drums where it solidified into grout (a cement-
like material), and remains stored on site, awaiting shipment to an
off-site disposal facility.
DOE's Initiative for Accelerating Cleanup Is Still Evolving, with the
Extent of Savings Uncertain:
DOE's new initiative, implemented in 2002, attempts to address the
schedule delays and increasing costs DOE has encountered in its efforts
to treat and dispose of high-level waste. This initiative is still
evolving. As of April 2003, DOE had identified several strategies to
help reduce the time needed to treat and dispose of the waste. Based on
these strategies, DOE estimated that it could reduce the waste cleanup
schedule by about 20 to 35 years at its high-level waste sites and save
about $29 billion compared to the existing program baseline.[Footnote
4] While some degree of savings is likely if the strategies are
successfully implemented, the extent of the savings is still uncertain.
Initiative Centers on Ways to Speed Disposal and Save Money:
Many of DOE's proposals to speed cleanup and reduce environmental risk
involve ways to do one or more of the following:
* Deal with some tank waste as low-level or transuranic[Footnote 5]
waste, rather than as high-level waste. Doing so would eliminate the
need to vitrify the waste for off-site disposal in the geologic
repository for high-level waste.
* Complete the waste treatment more quickly by using additional or
supplemental technologies. For example, DOE's Hanford Site is
considering using up to four supplemental technologies, in addition to
vitrification, to process its low-activity waste. DOE believes these
technologies are needed to help it meet a schedule milestone date of
2028 agreed to with regulators to complete waste processing. Without
these technologies, DOE believes waste treatment would not be completed
before 2048.
* Segregate the waste more fully than initially planned and tailor
waste treatment to each of the waste types. By doing so, DOE plans to
apply less costly treatment methods to waste with lower concentrations
of radioactivity.
* Close waste storage tanks earlier than expected, thereby avoiding the
operating costs involved in maintaining the tanks and monitoring the
wastes.
Table 1 summarizes the estimated cost savings for each DOE site if
accelerated proposals for cleaning up high-level waste are successfully
implemented.
Table 1: DOE's Estimated Cost Savings from Proposals to Accelerate
Cleanup of High-Level Waste:
Amounts are in billions of current dollars, fiscal year 2003 to the end
of cleanup:
Site: Idaho National Laboratory; Current baseline lifecycle cost
estimate: $10.07; Accelerated lifecycle cost estimate: $ 3.10;
Estimated savings from accelerated initiatives: $ 6.97.
Site: Hanford; Current baseline lifecycle cost estimate: 56.19;
Accelerated lifecycle cost estimate: 41.67; Estimated savings from
accelerated initiatives: 14.52.
Site: Savannah River; Current baseline lifecycle cost estimate: 18.82;
Accelerated lifecycle cost estimate: 11.49; Estimated savings from
accelerated initiatives: 7.33.
Site: Totals; Current baseline lifecycle cost estimate: $85.08;
Accelerated lifecycle cost estimate: $56.26; Estimated savings from
accelerated initiatives: $28.82.
Source: DOE.
Note: West Valley is not included in this table because high-level
waste cleanup at the site was essentially completed in September 2002.
[End of table]
Savings Estimate May Not Be Reliable:
Our review indicates that DOE's current estimate of $29 billion may not
yet be reliable and that the actual amount to be saved if DOE
successfully implements the alternative waste treatment and disposal
strategies may be substantially different from what DOE is projecting.
We have several concerns about the reliability and completeness of the
estimate. These concerns include the accuracy of baseline cost
estimates from which savings are calculated, whether all appropriate
costs are included in the analysis, and whether the savings estimates
properly reflect the timing of the savings or uncertainties.
Baseline Costs Are Not Fully Reliable:
DOE's current lifecycle cost baseline is used as the base cost from
which potential savings associated with any improvements are measured.
However, in recent years, we and others have raised concerns about the
reliability of DOE's baseline cost estimates. In a 1999 report, we
noted that DOE lacked a standard methodology for sites to use in
developing their lifecycle cost baseline, raising a concern about the
reliability of data used to develop these cost estimates.[Footnote 6]
DOE's Office of Inspector General also raised a concern in a 1999
review of DOE project estimates, noting that several project cost
estimates examined were not supported or complete. DOE acknowledged in
its February 2002 review of the cleanup program that baseline cost
estimates do not provide a reliable picture of project costs.[Footnote
7]
Estimates of Project Costs May Be Incomplete:
Some of DOE's savings may be based on incomplete estimates of the costs
for the accelerated proposals. According to Office of Management and
Budget (OMB) guidance on developing cost estimates, agencies should
ensure that all appropriate costs are addressed in the estimate.
However, DOE has not always done so. For example, the Idaho National
Laboratory's estimated savings of up to $7 billion is based, in large
part, on eliminating the need to build a vitrification facility to
treat its waste. However, the waste may have to undergo an alternative
treatment method before it can be accepted at a geological repository,
and the Idaho National Laboratory is considering four different
technologies for doing so. Nevertheless, DOE's current savings estimate
reflects the potential cost of only one of those technologies. DOE has
not yet developed the costs of using any of the other waste treatment
approaches. DOE noted that the accelerated lifecycle estimate could
likely change depending on which one of the technologies is selected
and the associated costs of treating the waste are developed.
Savings Estimates Do Not Reflect Timing, Uncertainty, or Nonbudgetary
Impacts:
According to OMB guidance, agencies should ensure that the timing of
when the savings will occur is accounted for, that uncertainties are
recognized and quantified where possible, and that nonbudgetary
impacts, such as a change in the level of risk to workers, are
quantified, or at least described. We found problems in all three
areas.
* Regarding the time value of money, applying OMB guidance would mean
that estimates of savings in DOE's accelerated plans should reflect a
comparison of its baseline cost estimate with the alternative,
expressed in a "present value," where the dollars are discounted to a
common year to reflect the time value of money. Instead, DOE's savings
estimates generally measure savings by comparing dollars in
different years. For example, the Savannah River Site estimates a
savings of nearly $5.4 billion by reducing by 8 years (from 2027 to
2019) the time required to process its high-level waste. Adjusting the
savings estimate to present value in 2003 results in a savings of
$2.8 billion in 2003 dollars.
* Regarding uncertainties, in contrast to OMB guidance, the DOE savings
estimates generally do not consider uncertainties. For example, the
savings projected in the Idaho National Laboratory's accelerated plan
reflect the proposal to no longer build the vitrification facility and
an associated reduction in operations costs. However, the savings do
not account for uncertainties such as whether alternatives to
vitrification will succeed and at what cost. Rather than reflecting
uncertainties by providing a range of savings, DOE's savings estimate
is a single point estimate of $7 billion.
* Regarding nonbudgetary impacts, DOE's savings estimates generally
do not fully assess the value of potential nonbudgetary impacts, such
as a change in the level of risk to workers or potential effects on the
environment. OMB guidelines recommend identification and, where
possible, quantification of other expected benefits and costs to
society when evaluating alternative plans. For example, the Idaho
National Laboratory's accelerated plan does not assess potential
increases in environmental risk, if any, from disposing of the waste
without stabilizing it into a vitrified form. By not assessing these
benefits and risks to workers and the environment, DOE leaves unclear
how important these risks and trade-offs are to choosing an alternative
treatment approach.
Key Legal and Technical Challenges Could Limit Potential Savings from
DOE's Accelerated Cleanup Initiative:
DOE faces significant legal and technical challenges in achieving the
cost and schedule reductions proposed in its new initiative. On the
legal side, DOE's proposals depend heavily on the agency's authority to
apply a designation other than "high-level waste" to the low-activity
portion of the waste stream, so that this low-activity portion does not
have to be disposed of more expensively as high-level waste. The
portion of DOE's order setting out criteria for making such
determinations has been invalidated in a recent court ruling. On the
technical side, DOE's proposals rest heavily on the successful
application of waste separation methods that are still under
development and will not be fully tested before being put in place.
DOE's track record in this regard has not been strong; it has had to
abandon past projects that were also based on promising--but not fully
tested--technologies. Either or both of these challenges could limit
the potential savings from DOE's accelerated cleanup initiative.
DOE's Accelerated Initiative Relies on a Process for Reclassifying
Waste That the Court Has Ruled Invalid:
DOE has traditionally managed all of the wastes in its tanks as
high-level waste because the waste resulted primarily from the
reprocessing of spent nuclear fuel and contains significant amounts of
radioactivity. However, by separating the waste into high-level and
low-activity portions and managing the low-activity portion as
something other than high-level waste, DOE could use less costly and
less complicated treatment approaches. DOE has developed guidelines for
deciding when waste in the tanks should not be considered high-level
waste. In 1999, under Order 435.1, DOE formalized its process for
determining which waste is incidental to reprocessing ("incidental
waste"), not high level waste, and therefore will not be sent to a
geological repository for high-level waste disposal. This process
provides a basis for DOE to treat and dispose of some portion of its
wastes less expensively as low-level or transuranic wastes.
DOE's ability to define some waste as incidental to reprocessing, and
to then follow a different set of treatment and disposal requirements
for that waste, is central to its overall strategy for addressing its
tank waste. For example, DOE planned to use its incidental waste
process to manage about 90 percent of its 54 million gallons of tank
waste at the Hanford Site as low-level waste, rather than process it
through a high-level waste vitrification facility. Using that approach,
most of the waste would be eligible for treatment and disposal on site.
Such an approach would save billions compared to treating all of the
waste as high-level waste and sending it for disposal in a high-level
waste geologic repository.
A recent court ruling precludes DOE from reclassifying some of its
waste as other than high-level waste. In March 2002, the Natural
Resources Defense Council and others filed a lawsuit challenging DOE's
authority to manage its wastes through its incidental waste
process.[Footnote 8] The plaintiffs alleged that DOE arbitrarily
established the incidental waste determination process without proper
regard for the law or properly establishing a justification for this
process. A primary concern of the plaintiffs was that DOE would use its
incidental waste process to permanently leave intensely radioactive
waste sediments in the tanks with only minimal treatment. The lawsuit
alleged that DOE's incidental waste process improperly allows DOE to
reclassify high-level waste as incidental waste that does not need to
be treated in the same way as high-level waste. According to the
plaintiffs, the Nuclear Waste Policy Act defines all waste originating
from a given source--that is, from reprocessing of spent nuclear fuel-
-as high-level waste and requires that such waste be managed as
high-level waste, yet DOE has chosen to differentiate its wastes
according to the level of radioactivity and manage them accordingly. In
a July 3, 2003 ruling on the lawsuit, the court agreed with the
plaintiffs, stating that the portion of DOE's Order 435.1 setting out
its incidental waste determination process violates the Nuclear Waste
Policy Act and thus is invalid.
The court's ruling could seriously hinder DOE's efforts to implement
its accelerated treatment and disposal strategies. Under the ruling,
DOE's incidental waste determinations cannot be implemented. Since the
start of the lawsuit, DOE had not implemented any of its approved
incidental waste determinations and had not yet decided whether to
defer or proceed with its pending incidental waste determinations--such
as those for closing tanks at the Savannah River Site and Idaho
National Laboratory.
If DOE appeals the court ruling, a lengthy legal process could follow.
A lengthy legal process will also likely delay treatment plans for this
waste and delay closing tanks on an accelerated schedule. For example,
the Idaho National Laboratory planned to begin closing tanks in the
spring of 2003, pending approval of an incidental waste determination
that would allow DOE to close the tanks by managing tank waste
residuals as low-level waste.[Footnote 9] A DOE official at the Idaho
National Laboratory told us that while a delay of several months would
not immediately threaten schedule dates, a delay beyond 24 months would
seriously affect the site's ability to meet its accelerated 2012 date
to close all of the tanks.
If the court's ruling invalidating DOE's incidental waste determination
process is upheld, DOE may need to find an alternative that would allow
it to treat waste with lower concentrations of radioactivity less
expensively. Searching for such an alternative could delay progress at
all three of DOE's high-level waste sites that rely on incidental waste
determinations. If DOE cannot meet its accelerated schedules, then
potential savings are in jeopardy. At this point, the department does
not appear to have a strategy to avoid the potential effects of
challenges to its incidental waste determination authority, either from
the current court ruling or future challenges. At the time of our
report, DOE officials told us that they believed the department would
prevail in the legal challenge. DOE believed it would be premature to
explore alternative strategies to overcome potentially significant
delays to the program that could result from a protracted legal
conflict or from an adverse decision. Such strategies could range from
exploring alternative approaches for establishing an incidental waste
regulation to asking that the Congress provide legislative authority
for DOE to implement an incidental waste policy.
Accelerated Initiative Also Relies on Waste Separation Approaches That
Will Not Be Fully Tested:
Like the ability to determine that some waste is incidental to
reprocessing, the ability to separate the waste components is important
to meet waste cleanup schedule and cost goals. If the waste is not
separated, all of it--about 94 million gallons--may have to be treated
as high-level waste and disposed of in the geological repository. Doing
so would require a much larger repository than currently planned, and
drive up disposal costs by billions of dollars. Successful separation
will substantially reduce the volume of waste needing disposal at the
planned repository, as well as the time and cost required to prepare it
for disposal, and allow less expensive methods to be used in treating
and disposing of the remaining low-activity waste. The waste separation
process is complicated, difficult, and unique in scope at each site.
The waste differs among sites not only in volume but also in the way it
has been generated, managed, and stored over the years.
The challenge to successfully separate the waste is significant at the
Hanford Site, where DOE intends to build a facility for separating the
waste before fully testing the separation processes that will be used.
The planned laboratory testing includes a combination of pilot-scale
testing of major individual processes and use of operational data for
certain of those processes for which DOE officials said they had
extensive experience. However, integrated testing will not be performed
until full-scale facilities are constructed. DOE plans to fully test
the processes for the first time during the operational tests of the
newly constructed facilities.
This approach does not fully reflect DOE guidance, which calls for
ensuring that new or complex technology is mature before integrating it
into a project. Specifically, DOE's Project Management Order 413.3
requires DOE to assess the risks associated with technology at various
phases of a project's development. For projects with significant
technical uncertainties that could affect cost and schedule, corrective
action plans to address these uncertainties are required before the
projects can proceed. In addition, DOE's supplementary project
management guidance suggests that technologies be developed to a
reasonable level of maturity before a project progresses to full
implementation to reduce risks and avoid cost increases and schedule
delays. The guidance suggests that DOE avoid the risk of designing
facilities concurrently with technology development.
The laboratories working to develop Hanford's waste separation process
have identified several technical uncertainties, which they are working
to address. These uncertainties or critical technology risks include
problems with separating waste solids through an elaborate filtration
system, problems associated with mixing the waste during separation
processes, and various problems associated with the low-activity waste
evaporator.
Given these and other uncertainties, Hanford's construction
contractor and outside experts have seen Hanford's approach as
having high technical risk and have proposed integrated testing during
project development. However, DOE and the construction contractor
eventually decided not to construct an integrated pilot facility and
instead to accept a higher-risk approach. DOE officials said they
wanted to avoid increasing project costs and schedule delays, which
they believe will result from building a testing facility. Instead,
Hanford officials said that they will continue to conduct pilot-scale
tests of major separation processes. DOE officials said they believe
this testing will provide assurance that the separation processes will
function in an integrated manner. After the full-scale treatment
facilities are constructed, DOE plans to fully test and demonstrate the
separation process during facility startup operations.
The consequences of not adhering to sound technology development
guidelines can be severe. At the Savannah River Site, for example,
DOE invested nearly $500 million over nearly 15 years to develop a
waste separation process, called in-tank precipitation, to treat
Savannah River's high-level waste. While laboratory tests of this
process were viewed as successful, DOE did not adequately test the
components until it started full-scale operations. DOE followed this
approach, in part, because the technology was commercially available
and considered "mature." However, when DOE started full-scale
operations, major problems occurred. Benzene, a dangerously flammable
byproduct, was produced in large quantities. Operations were stopped
after DOE spent about $500 million because experts could not explain
how or why benzene was being produced and could not determine how to
economically reconfigure the facility to minimize it. Consequences of
this technology failure included significant cost increases, schedule
delays, a full-scale waste separation process that did not work, and a
less-than-optimum waste treatment operation. Savannah River is now
developing and implementing a new separation technology at an
additional cost of about $1.8 billion and a delay of about
7 years.[Footnote 10]
Subsequent assessments of the problems that developed at Savannah River
found that DOE (1) relied on laboratory-scale tests to demonstrate
separation processes, (2) believed that technical problems could be
resolved later during facility construction and startup, and
(3) decided to scale up the technology from lab tests to full-scale
without the benefit of using additional testing facilities to confirm
that processes would work at a larger scale. Officials at Hanford are
following a similar approach. Several experts with whom we talked
cautioned that if separation processes at Hanford do not work as
planned, facilities will have to be retrofitted, and potential cost
increases and schedule delays would be much greater than any associated
with integrated process testing in a pilot facility.
Opportunities Exist to Explore Additional Cost Savings and to
Strengthen Program Management:
In addition to the potential cost savings identified in the accelerated
site cleanup plans, DOE continues to develop and evaluate other
proposals to reduce costs but is still assessing them. Although the
potential cost savings have not been fully developed, they could be in
the range of several billion dollars, if the proposals are successfully
implemented. At the Savannah River and Hanford sites, for example, DOE
is identifying ways to increase the amount of waste that can be placed
in its high-level waste canisters to reduce treatment and disposal
costs. DOE also has a number of initiatives under way to improve
overall program management. However, we are concerned that the
initiatives may not be adequate. In our examinations of problems that
have plagued DOE's project management over the years, three
contributing factors often emerged--making key project decisions
without rigorous analysis, incorporating new technology before it has
received sufficient testing, and using a "fast-track" approach
(concurrent design and construction) on complex projects. Ensuring that
these weaknesses are addressed as part of its program management
initiatives would further improve the management of the program and
increase the chances for success.
DOE Is Considering Additional Potential Opportunities to Reduce Costs:
DOE is continuing to identify other proposals for reducing costs under
its accelerated cleanup initiative. Among the proposals that DOE is
considering, the ones that appear to offer significant cost savings
opportunities would increase the amount of waste placed in each
disposal canister. The amount of waste that can be placed into a
canister depends on a complex set of factors, including the specific
mix of radioactive material combined with other chemicals in the waste,
such as chromium and sulfate, that affect the processing and quality of
the immobilized product. These factors affect the percentage of waste
than can be placed in each canister because they indicate the
likelihood that radioactive constituents could move out of the
immobilizing glass medium and into the environment. The greater the
potential for the waste to become mobile, the lower the
allowable percentage of waste and the higher the percentage of glass
material that must be used.
Savannah River officials believe they can increase the amount of waste
loaded in each canister from 28 percent to about 35 percent, and for at
least one waste batch, to nearly 50 percent. In June 2003, Savannah
River began to implement this new process to increase the amount of
waste in each canister. If successful, Savannah River's improved
approach could reduce the number of canisters needed by about 1,000
canisters and save about $2.7 billion, based on preliminary estimates.
Other efforts to increase waste loading of the canisters are also under
way that, if successful, may permit further cost savings of about $1.7
billion. The Hanford Site is also exploring ways to decrease the
numbers of waste canisters that will be needed by using waste forms
other than the standard borosilicate glass. This effort is in a very
early stage of development and cost-savings estimates have not been
fully developed.
DOE Has Opportunities to Improve Management of the Program by
Addressing Previously Identified Weaknesses:
In addition to site-specific proposals for saving time and money, DOE
is also undertaking management improvements using teams to study
individual issues. Nine teams are currently in place, while other teams
to address issues such as improving the environmental review process to
better support decision making have not yet been formed. Each team has
a disciplined management process to follow,[Footnote 11] and even after
the teams' work is completed, any implementation will take time. These
efforts are in the early stages, and therefore it is unclear if they
will correct the performance problems DOE and others have identified.
We are concerned that these management reforms may not go far enough in
addressing performance problems with the high-level waste program. Our
concerns stem from our review of initiatives under way in the
management teams, our discussions with DOE officials, and our past and
current work, as well as work by others inside and outside DOE. We have
identified three recurring weaknesses in DOE's management of cleanup
projects that we believe need to be addressed as part of DOE's overall
review. These weaknesses cut across the various issues that the teams
are working on and are often at the center of problems that have been
identified. Two of these weaknesses have been raised earlier in this
testimony--lack of rigor in the analysis supporting key decisions, and
incorporating technology into projects before it is sufficiently
mature. The final area of weakness involves using "fast-track" methods
to begin construction of complex facilities before sufficient planning
and design have taken place.
Key Decisions Not Always Supported by Rigorous Current Analysis:
DOE's project management guidance emphasizes the importance of rigorous
and current analysis to support decision making during the development
of DOE projects. Similarly, OMB guidance states that agencies should
validate earlier planning decisions with updated information before
finalizing decisions to construct facilities. This validation is
particularly important where early cost comparisons are susceptible to
uncertainties and change.
DOE does not always follow this guidance, yet no DOE management team
appears to be addressing this weakness. Proceeding without rigorous
review has been a recurring cause of many of the problems we have
identified in past DOE projects. For example, the decision at Hanford
to construct a vitrification plant to treat Hanford's low-activity
waste has not been validated with updated information. Hanford's
primary analysis justifying the cost of this approach was prepared in
1999 and was based on technical performance data, disposal assumptions,
and cost data developed in the early to mid-1990s--conditions that are
no longer applicable. Subsequent analyses have continued to rely on
this data. However, since that time conditions have changed, including
the performance capabilities of alternative technologies such as grout,
the relative cost of different technologies, and the amount of waste
DOE intends to process through a vitrification facility.
DOE officials disagree with our assessment of their analysis, stating
that a comprehensive analysis was conducted in the spring of 2003.
However, DOE's high-level waste project team agreed that the DOE
officials at Hanford had not performed a current, rigorous analysis of
low-activity waste treatment options including the use of grout as an
alternative to vitrification, and the team encouraged the Hanford site
to update its analysis based on current waste treatment and disposal
assumptions. DOE officials at Hanford told us they do not plan to
reassess the decision to construct a low-activity vitrification
facility because their compliance agreement with the state of
Washington calls for vitrification of this waste. They also stated that
vitrification is a technology needed for destroying hazardous
constituents in a portion of the waste.
New Technology Is Incorporated before It Is Sufficiently Mature:
Our work on Department of Defense acquisitions has documented a set of
"best practices" used by industry for integrating new technology into
major projects. We reported in July 1999 that the maturity of a
technology at the start of a project is an important determinant of
success.[Footnote 12] As technology develops from preconceptual design
through preliminary design and testing, the maturity of the technology
increases and the risks associated with incorporating that technology
into a project decrease. Waiting until technology is well-developed and
tested before integrating it into a project will greatly increase the
chances of meeting cost, schedule, and technical baselines. On the
other hand, integrating technology that is not fully mature into a
project greatly increases the risk of cost increases and schedule
delays. According to industry experts, correcting problems after a
project has begun can cost 10 times as much as resolving technology
problems beforehand.
DOE's project management guidance issued in October 2000 is consistent
with these best practices. The guidance discusses technology
development and sets out suggested steps to ensure that new technology
is brought to a sufficient level of maturity at each decision point in
a project. For example, during the conceptual design phase of a
project, "proof of concept" testing should be performed before approval
to proceed to the preliminary design phase. Furthermore, the guidance
states that attempting to concurrently develop the technology and
design the facility for a project poses ill-defined risks to the
project.
Nevertheless, as we discussed earlier, DOE sites continue to integrate
immature technologies into their projects. For example, as discussed
earlier, DOE is constructing a facility at the Hanford Site to separate
high-level waste components, although integrated testing of the many
steps in the separations process has not occurred and will not occur
until after the facility is completed. DOE, trying to keep the project
on schedule and within budget, has decided the risks associated with
this approach are acceptable. However, there are many projects for
which this approach created schedule delays and unexpected costs. The
continued reliance on this approach in the face of so many past
problems is a signal of an area that needs careful attention as DOE
proceeds with its management reform efforts. At present, no DOE
management team is addressing this issue.
Facility Construction Starts before Design Is Sufficiently Developed:
Finally, we have concerns about DOE's practice of launching into
construction of complex, one-of-a-kind facilities well before their
final design is sufficiently developed, again in an effort to save time
and money. Both DOE guidance and external reviews stress the importance
of adequate upfront planning before beginning project construction.
DOE's project management guidance identifies a series of well-defined
steps before construction begins and suggests that complex projects
with treatment processes that have never before been combined into a
facility do not lend themselves to being expedited. However, DOE
guidance does not explicitly prohibit a fast-track--or concurrent
design and construction--approach to complex, one-of-a-kind projects,
and DOE often follows this approach. For example, at the Hanford Site,
DOE is concurrently designing and constructing facilities for the
largest, most complex environmental cleanup job in the United States.
Problems are already surfacing. Only 24 months after the contract was
awarded, the project was 10 months behind schedule dates, construction
activities have outpaced design work causing inefficient work
sequencing, and DOE has withheld performance fee from the design/
construction contractor because of these problems.
DOE experienced similar problems in concurrent design and construction
activities on other waste treatment facilities. Both the spent nuclear
fuel project at Hanford and the waste separations facility at the
Savannah River Site encountered schedule delays and cost increases in
part because the concurrent approach led to mistakes and rework, and
required extra time and money to address the problems.[Footnote 13] In
its 2001 follow-up report on DOE project management, the National
Research Council noted that inadequate pre-construction planning and
definition of project scope led to cost and schedule overruns on DOE's
cleanup projects.[Footnote 14] The Council reported that research
studies suggest that inadequate project definition accounts for
50 percent of the cost increases for environmental
remediation projects. Again, no DOE team is specifically examining the
"fast-track" approach, yet it frequently contributed to past problems
and DOE continues to use this approach.
Conclusions:
DOE's efforts to improve its high-level waste cleanup program and to
rein in the uncontrolled growth in project costs and schedules are
important and necessary. The accelerated cleanup initiative represents
at least the hope of treating and disposing of the waste in a more
economical and timely way, although the actual savings are unknown at
this time. Furthermore, specific components of this initiative face key
legal and technical challenges. Much of the potential for success
rested on DOE's ability to dispose of large quantities of waste with
relatively low concentrations of radioactivity on site by applying its
incidental waste process. Recently, a court ruled that the portion of
DOE's order setting out its incidental waste determination process
violates the Nuclear Waste Policy Act and is invalid. Thus, DOE is
precluded from implementing this element of its accelerated initiative.
Success in accelerating cleanup also rests on DOE's ability to obtain
successful technical performance from its as-yet unproven waste
separation processes. Any technical problems with these processes will
likely result in costly delays. At DOE's Hanford Site, we believe the
potential for such problems warrants reconsidering the need for more
thorough testing of the processes, before completing construction of
the full-scale waste separation facility.
DOE's accelerated cleanup initiative should mark the beginning, not
the end, of DOE's efforts to identify other opportunities to improve
the program by accomplishing the work more quickly, more effectively,
or at less cost. As DOE continues to pursue other management
improvements, it should reassess certain aspects of its current
management approach, including the quality of the analysis underlying
key decisions, the adequacy of its approach to incorporating new
technologies into projects, and the merits of a fast-track approach to
designing and building complex nuclear facilities. Although the
challenges are great, the opportunities for program improvements are
even greater. Therefore, DOE must continue its efforts to clean up its
high-level waste while demonstrating tangible, measurable program
improvements.
In the report being released today, we made several recommendations to
help DOE manage or reduce the legal and technical risks faced by the
program as well as to strengthen DOE's overall program management. DOE
agreed to consider seeking clarification from Congress regarding its
authority to define some waste as incidental to reprocessing, if the
legal challenge to its authority significantly affected DOE's ability
to achieve savings under the accelerated initiative. Regarding our
recommendations to conduct integrated pilot-scale testing of the
separations facility at Hanford before construction is completed, and
to make other management improvements to address the weaknesses I just
discussed, DOE's position is that it has already taken appropriate
steps to manage the technology risks and strengthen its management
practices. We disagree and believe that implementing all of our
recommendations would help reduce the risk of costly delays and improve
overall management of DOE's entire high-level waste program.
Thank you, Mr. Chairman and Members of the Subcommittee. That concludes
my testimony. I would be pleased to respond to any questions that you
may have.
Contacts and Acknowledgements:
For further information on this testimony, please contact Ms. Robin
Nazzaro at (202) 512-3841. Individuals making key contributions to this
testimony included Carole Blackwell, Robert Crystal, Doreen Feldman,
Chris Hatscher, George Hinman, Gary Jones, Nancy Kintner-Meyer, Avani
Locke, Mehrzad Nadji, Cynthia Norris, Tom Perry, Stan Stenersen, and
Bill Swick.
FOOTNOTES
[1] U.S. General Accounting Office, Nuclear Waste: Challenges to
Achieving Potential Savings in DOE's High-Level Waste Cleanup Program,
GAO-03-593 (Washington, D.C.: June 17, 2003).
[2] Each radioactive component, or radionuclide, in high-level waste
loses its radioactivity at a rate that differs for each component. This
rate of decay, which cannot be changed, is measured in "half-lives"--
that is, the length of time required for half of the unstable atoms to
decay and release their radiation.
[3] At Savannah River, high-level sludge from the tanks has also been
stabilized in glass material and is currently stored on site pending
completion of the geologic repository. As of August 30, 2002, Savannah
River had produced 1,331 canisters of this stabilized waste.
[4] Unless otherwise noted, all dollar estimates are as reported by DOE
and are in current dollars.
[5] Low-level radioactive waste is defined as radioactive material that
is not high-level radioactive waste, spent nuclear fuel, transuranic
waste, or certain by-product material (the tailings or wastes produced
by the extraction or concentration or uranium or thorium from any ore
processed primarily for its source material content). 42 U.S.C.
10101(16). Transuranic wastes come primarily from reprocessing of spent
nuclear fuel and from fabrication of nuclear weapons. Transuranic waste
is defined as waste with radionuclides with atomic numbers greater than
92 (that is, uranium) and having half-lives greater than 20 years in
concentrations greater than 100 nanocuries per gram.
[6] U.S. General Accounting Office, Nuclear Waste: DOE's Accelerated
Cleanup Strategy Has Benefits but Faces Uncertainties, GAO/RCED-99-129
(Washington, D.C.: Apr. 30, 1999).
[7] U.S. Department of Energy, A Review of the Environmental Management
Program (Washington, D.C.: Feb. 4, 2002).
[8] Natural Resources Defense Council, Inc. v. Abraham, No. 01-CV-413
(D. Idaho, filed Mar. 5, 2002). The lawsuit was originally filed in
January 2000 in the 9th Circuit Court of Appeals and was subsequently
transferred to the federal district court in Idaho. The other parties
to the lawsuit are the Snake River Alliance, the Confederated Tribes
and Bands of the Yakama Nation, and the Shoshone Bannock Tribes. In
addition, the states of Washington, Idaho, Oregon and South Carolina
are participating as amicus curiae.
[9] Tank closure at the Idaho National Laboratory is also pending
completion of its National Environmental Policy Act process.
[10] U.S. General Accounting Office, Nuclear Waste: Process to Remove
Radioactive Waste From Savannah River Tanks Fails to Work, GAO/RCED-99-
69 (Washington, D.C.: Apr. 30, 1999).
[11] Under DOE's project management principles, for example, teams must
define project requirements, conduct preliminary risk assessments, and
prepare a risk mitigation plan prior to developing a baseline cost
estimate of proposed alternatives.
[12] U.S. General Accounting Office, Best Practices: Better Management
of Technology Development Can Improve Weapon System Outcomes, GAO/
NSIAD-99-162 (Washington, D.C.: July 30, 1999).
[13] For a discussion of the problems associated with the fast track
design/build approach on these projects, see U.S. General Accounting
Office, Nuclear Waste: DOE's Hanford Spent Nuclear Fuel Storage
Project--Cost, Schedule, and Management Issues, GAO/RCED-99-267
(Washington, D.C.: Sept. 20, 1999) and Nuclear Waste: Process to Remove
Radioactive Waste From Savannah River Tanks Fails to Work, GAO-RCED-99-
69 (Washington, D.C.: Apr. 30, 1999).
[14] National Research Council, Progress in Improving Project
Management at the Department of Energy (Washington, D.C.: Nov. 2001).