Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 15
Subsurface Contamination in the
DOE Complex
Over the last five decades, the United States has created a massive
industrial complex to develop, test, manufacture, and maintain nuclear
weapons for national security purposes. The U.S. Army Corps of Engi-
neers, Manhattan Engineering District, started constructing the complex
during the Second World War. The complex was expanded during the
ensuing Cold War by the Atomic Energy Commission, the Energy
Research and Development Authority, and starting in 1 977, the Depart-
ment of Energy (DOE). The DOE complex, as it has come to be known,
encompasses 1 34 distinct geographic sites in 31 states and one territory
with a total area of over two million acres (DOE, 1 998a). The individual
sites range in size from several hundred square miles to less than one
square mile; these sites host a variety of defense-related activities rang-
ing from uranium mining and milling to nuclear weapons testing (see
Figure 2.1 ).
The production and testing of nuclear weapons has created a legacy
of significant environmental contamination, as described in some detail
later in this chapter. In 1 989, Congress created the Office of Environ-
mental Management (EM) in DOE to reduce threats to health and safety
posed by the environmental contamination at DOE sites. To meet this
objective, EM has undertaken a major cleanup effort, which, according
to DOE, is the largest environmental cleanup in the world. This is cer-
tainly true from a cost standpoint: EM is now spending about $5.8 bil-
lion per year on its cleanup program and has spent over $50 billion
since 1990. It expects to spend another $147 billion between 1997 and
2070 (DOE, 1 998a), but this estimate is uncertain because the magni-
tude of contamination and the level of cleanup effort required at some
sites are still poorly understood.
In this chapter, the committee provides an overview of the subsur-
face contamination problems around the DOE complex and shows by
C h a p t e r 2
OCR for page 16
~-
-~
owl ~
~ =
' I ~
-I
DO
\~ I
~ · ~ 1 141
_ —~
~ · ~
FIGURE2.1 Location ofDOE example how advances in scientific and engineering knowledge can
complex sites.The major improve cleanup effectiveness. The chapter is organized into three sec-
sites are labeled by name on tions. The fi rst provides an overview of the DO E com p I ex and its m is-
the figure. The locations of sion and describes the legacy of contamination from weapons produc-
other sites are indicated by tion and re I ated activities. The second section i I I ustrates the range of
closed cimles.SOURCE: DOE subsurface problems that exist across the complex today and what DOE
is doing to correct them. The examples are taken from the six largest
DOE sites: Hanford, Idaho, Nevada, Oak Ridge, Rocky Flats, and
Savannah River (see Sidebar 2.1~. In the third section, the committee
discusses how scientific and engineering research can improve the
effectiveness of DOE's mission to tackle these contamination problems.
S U B S U R F A C E S C ~ E N C E
OCR for page 17
-
- ~
\ ·
\
~ - \
/ -
- ~ . ~
W:~
J ~
y
J .
1 1
-in
rim
_ ~
/
C3
/
-
r
\
L/
This discussion will be used to support the recommendations in
Chapters 5 and 6.
Past Practices and Consequences
Nuclear weapons production during the Cold War was a highly
industrialized enterprise that involved a vast complex of mines and
industrial sites across the United States. The front end of the process
was focused on the production of uranium, which was then used to
produce other weapons materials, particularly plutonium and tritium.
C h a p t e r 2
\
/
OCR for page 18
SIDEBAR2.1 THE DOE COMPLEX
Although the DOE complex encompasses over 100 distinct sites, much of the major defense-related
activities were conducted at the six largest DOE sites (see Figure 2.1) described below.
The Hanford Site is located in southeastern Washington state and covers an area of about 1,450 square
kilometers (560 square miles). Production of materials for nuclear weapons took place here from the
1940s until mid-1989.The site contains several production reactors,chemical separations plants,and
solid and liquid waste storage sites.
The Idaho National Engineering and Environmental Laboratory, first established as the Nuclear Reactor
Testing Station and then the Idaho National Engineering Laboratory, occupies 2,300 square kilometers
(890 square miles) in a remote desert area along the western edge of the upper Snake River plain.The
site was established as a building, testing, and operating station for various types of nuclear reactors
and propulsion systems, and the site also manages spent fuel from the naval reactor program.
The Nevada Test Site, which occupies about 3,500 square kilometers (1,350 square miles) in southern
Nevada, was the primary location for atmospheric and underground testing of the nation's nuclear
weapons starting in 1951.
The Oak Ridge Reservation covers an area of approximately 155 square kilometers (60 square miles)
and is located about 10 kilometers (6 miles) west of Knoxville,Tennessee.The reservation has three
major operating facilities: the Oak Ridge National Laboratory, the Y-12 Plant, and the K-25 Plant.The
laboratory was originally constructed as a research and development facility to support plutonium
production technology. The Y-12 Plant was built to produce highly enriched uranium by electromag-
netic separation; and the K-25 Plant, formerly known as the Oak Ridge Gaseous Diffusion Plant, also
was created to produce highly enriched uranium for nuclear weapons.
The Rocky Flats Environmental Technology Site is situated on about 140 hectares (~350 acres) near
Denver, Colorado, and has more than 400 manufacturing, chemical processing, laboratory, and support
facilities that were used to produce nuclear weapons components. Production activities once included
metalworking, fabrication and component assembly, and plutonium recovery and purification.
Operations at the site ceased in 1989.
The Savannah River Site, located near Aiken, South Carolina, covers an area of about 800 square kilo-
meters (300 square miles).The site was established in 1950 to produce special radioactive isotopes
(e.g., plutonium and tritium) for use in the production of nuclear weapons.The site contains produc-
tion reactors, chemical processing plants, and solid and liquid waste storage sites.
The back end was focused on the fabrication and testing of nuclear
devices. The major production steps and waste byproducts are
described in Sidebar 2.2.
The United States is no longer produc
ing plutonium and tritium1 for
1The secretary of energy has announced that DOE may produce tritium in the
future to replenish current stocks of nuclear weapons.
S U B S U R F A C E S C ~ E N C E
18
OCR for page 19
TABLE 2.1 Principal Dense Non-Aqueous Phase Liquic (DNAPL), Metal, anc
Racionuc~ice Contaminants in the DOE Complex
DNAPLs Metals Radionuclides
Trichloroethylene Lead Plutonium
Dichloroethylene Chromium (Vl) Strontium-90
Tetrachloroethylene Mercury Cesium-137
Perchloroethylene Zinc Uranium (various isotopes)
Chloroform Beryllium Tritium
Dichloromethane Arsenic Thorium
Polychlorinated Biphenyls Cadmium Technetium-99
Copper
Radium
iodine-1 29
SOURCE: EPA (1977); INEEL (1997); Riley and Zachara (1992).
nuclear weapons, and a large part of the DOE complex has been shut
down or placed on standby. All of DOE's production reactors have been
shut down, and only two reprocessing facilities (the F and H canyons at
Savannah River) continue to operate. These are scheduled to be phased
out during the next decade. The weapons design and assembly facilities
also continue to operate, but their mission now includes the disassem-
bly of surplus nuclear weapons. The Nevada Test Site remains open, but
only subcritical nuclear tests have been conducted there since 1992.
During the last decade, a large part of the DOE complex, including
some of the sites d iscussed i n Sidebar 2. 1, have taken on a new m is-
sion: namely, remediation of the environmental contamination resulting
from weapons production. This mission is formidable, because it
involves cleanup of a wide variety of hazardous chemicals and radioac-
tive materials introduced into the environment during five decades of
weapons production and testing (see Sidebar 2.3~. The contaminants
include dense non-aqueous phase liquids (DNAPLs; see Sidebar 2.4~;
toxic metals such as lead, chromium, and mercury; and radionuclides
such as plutonium, cesium, strontium, and tritium (see Table 2.1~.
These contaminants were introduced into the environment through
a variety of pathways, including intentional disposal into the ground
through injection wells, disposal pits, and settling ponds; and through
accidental spills and leaks from storage tanks and waste transfer lines.
In some cases, there is little information available on either the timing
or magnitude of contaminant releases to the environment, or the fate of
contaminants in the subsurface after release. Moreover, DOE sites are
C h a p t e r 2
19
OCR for page 20
SIDEBAR 2.2 NUCLEAR FUEL CYCLE AND NUCLEAR WEAPONS PRODUCTION
The production of nuclear
weapons is a technically com-
plex and highly industrialized
process.The major production
steps and waste byproducts
of this process are described
below.
Mining and milling. Uranium ore
was mined at over 400 sites in
the United States and processed
in mills to produce uranium
oxide.These processes produced
large volumes of mine and mill tailings that contained heavy metals and radioactive radium and thori-
um.This waste is being managed through the Uranium Mill Tailings Radiation Control Act program.
Nuclear Weapons Production
Uranium
Minina and Uranium Uranium Uranium
Milling Refining Enrichment Foundry
Fuel and Plutonium
Target Production
Fabrication Reactors
Uranium is mined and Uranium is processed into low-
Uranium gas is Uranium metal
retinediromore enrichment uranium highly convertedinto isformedinto
enriched uranium, an] depleted metal fuel and target
uranium elements for
reactors
Uranium enrichment Elaborate chemical processes were used to concentrate the fissile isotope uranium-
235 from the milled ore. Uranium enrichment facilities were built at Oak Ridge (Y-12 and K-25 Plants),
Ohio (Portsmouth Plant), and Kentucky (Paducah Plant).The waste streams from the enrichment process
include depleted uranium (i.e., depleted in U-235 relative to U-238), uranium-contaminated scrap metal,
polychlorinated biphenyl-contaminated waste, and a variety of organic solvents. Separation of lithium
isotopes at the Oak Ridge Y-12 plant also produced large amounts of mercury waste.
Fuel and target fabrication. The enriched uranium was converted to metal at the Fernald Plant in Ohio
and then fabricated into reactor fuel or targets for plutonium production at Hanford and Savannah
River.These processes produced uranium dust and a variety of chemical wastes.
Plutonium production. The United States produced about 100 metric tons of plutonium between 1944
and 1988 at 14 reactors at the Hanford and the Savannah River sites.The reactors at Savannah River
also produced tritium.Thousands of tons of uranium fuel were processed through the reactors during
their four decades of operation.The waste streams from these operations include solid and liquid
located in a variety of climatic zones and have complex subsurface
characteristics (see Table 2.2), which makes it difficult to predict the
location, transport, and fate of contaminants once they are released into
the environment. As discussed in some detail in other National
Research Counci I reports (N RC, 1 997a, 1 999), technologies to effec-
tively remediate many subsurface DNAPL, metal, and radionuclide con-
tamination problems are either lacking or are unproven for large-scale
site remediation.
Although subsurface contamination is generally acknowledged to be
a significant problem across the DOE complex, estimates of the magni-
S U B S U R F A C E S C ~ E N C E
20
OCR for page 21
Weapons
Design Testing
Nonnuclear
Components
Reprocessing
to Separate
Plutonium Nuclear
Components
fin_
I_ Assembly and
dismantlement
Department
—~~\\~—\ of Defense
Warhead Triggers/ neutron
generators, and other electrical and
mechanical components are
assembled into complete warheads
Uranium slugs are irradiated
to create plutonium metal
and chemical separation is
used to extract it
Uranium and plutonium
are further processed for
warhead triggers
Figure Source: DOE
radioactive waste, acids, and solvents.The cooling water from the reactors contained some radionu-
clides, most notably tritium.
Plutonium Separation. Plutonium and other special isotopes were separated from the irradiated fuel by
a variety of chemical processes. Chemical separations plants were located at the Hanford, Savannah
River, and Idaho sites. Operation of the separations plants produced significant volumes of highly
radioactive and hazardous chemical waste and water containing low levels of radionuclides and haz-
ardous chemicals.
Weapons design, fabrication, and assembly. Weapons design was the responsibility of the Los Alamos
and Lawrence Livermore National Laboratories.Weapons components were produced at several sites
in the United States, and final assembly took place at the Pantex Plant in Texas.The fabrication process
produced several waste streams, including scrap uranium and plutonium metal and solvents.
Weapons testing. The United States has conducted more than a thousand nuclear weapons tests in the
atmosphere, under water, and underground, and most have occurred at the Nevada Test Site. This test-
ing resulted in the contamination of surface and subsurface sites with radioactive materials, including
tritium, plutonium, and fission products.
tude of the problem vary considerably, as shown in Table 2.3. Accord-
ing to recent DOE estimates (DOE, 1 998a) there are about 6.4 hi l l ion
cubic meters (226 billion cubic feet) of contaminated soil, groundwater,
and related environmental media at its sites.2 Most of this contamina-
2The subsurface contamination estimates provided in this chapter are compiled
from various DOE documents. The committee cannot evaluate the accuracy of any
of these estimates, but believes based on the briefings and documents it received
during the course of this study that the estimates are likely to have very large
uncertainties.
C h a p t e r 2
21
OCR for page 22
tion is at two sites, the Hanford Site in eastern Washington and the
Idaho National Engineering and Environmental Laboratory in south-
central Idaho (see Figure 2.1~. At these two sites alone, EM cleanup is
not expected to be completed before 2050, and after cleanup is "com-
plete" EM does not know how much contamination will remain in the
ground to be managed through surveillance and containment.
EM's current cleanup plans, which also are given in the Paths to
Closure report (DOE, 1 998a), anticipate expenditures on the order of
$57 hi I I ion between 1 997 and 2006 to complete cleanup at al I but 1 0
TABLE 2.2 Geologic anc C~imato~ogic Variability Across the DOE Weapons Complex
DOE Site Climate Geology and Hydrogeology Surface Waters Depth to
Groundwater (m)
Savannah River Humid, subtropical Atlantic Coastal Plain with clay soils. Savannah River 0-38a
Site The strata are deeply dissected by
creeks, and most groundwater
eventually seeps into and is diluted
by the creeks.
Hanford Site Arid,cool;mild Alluvial plain of bedded sediments Columbia River 60-90b
winters and warm with sands and gravels. Groundwater
summers;average flows toward the Columbia River.
annual rainfall
16 cm (6.3 in.)
Oak Ridge Humid,typical of Valley and ridge province bordering Clinch River 6-37'
Reservation the southern the Cumberland Plateau. Primary
Appalachian region; porosity is low, but fracture porosity
average annual is present. High clay content.
precipitation Shallow water table.
138 cm (54.4 in.)
Rocky Flats Temperate, semiarid, Colorado Piedmont section of the Several streams 0-93
Environmental and continental Plains physiographic province. occur on or near
Technology temperatures; average Alluvial deposits cover the site. the facility
Site annual rainfall just
under 40 cm (1 5 in.)
Idaho National Semiarid with Near the northern margin of the Big Lost River and 60-240
Engineering and sagebrush-steppe Eastern Snake River plain, a low-lying other ephemeral
Environmental characteristics located area of late Tertiary and Quaternary streams
Laboratory in a belt of prevailing volcanism and sedimentation.
western winds; Basalt covers three-quarters of its
average annual rainfall surface.
22 cm (8.5 in.)
Michelle Ewart, SRS, personal communication, 2000.
bGephart and Lundgren (1998).
'Grover Chamberlain, DOE-HQ, personal communication, 2000.
Christine Gel les, DOE-HQ, personal communication, 2000.
SOURCE: Adapted from Sandia National Laboratories (1996),except where noted.
S U B S U R F A C E S C I E N C E
22
OCR for page 23
of its sites, including the major sites shown in Table 2.3. DOE expects
an additional expenditure of $79 billion to clean up those remaining 10
sites between 2007 and 2070. About $14 billion will be incurred for
SIDEBAR 2.3 A PRIMER ON RADIOACTIVE WASTE
Radioactive wastes are the unwanted byproducts of the nuclear fuel cycle (see Sidebar 2.2) and may
contain both radioactive isotopes and hazardous chemicals. In the United States, radioactive waste is
classified and managed by its source of production rather than by its physical, chemical, or radioactive
properties. Consequently, different classes of waste can contain many of the same radioactive isotopes,
and even"low-level"waste can contain certain long-lived radioactive isotopes.
In general, nuclear fuel cycle wastes are grouped into the following broad classes for purposes of man-
agement and disposal:
· Mill tailings are wastes resulting from the processing of ore to extract uranium and thorium.
· Spent nuclear fuel is fuel that has been irradiated in a nuclear reactor, and for the purposes of dis-
posal may include cladding and other structural components.
· High-level waste is the primary waste produced from chemical processing of spent nuclear fuel.This
waste is usually liquid in form and contains a wide range of radioactive and chemical constituents.
Spent nuclear fuel is often referred to as high-level waste in nuclear waste management terminolo-
gy although it is defined differently in the regulations.
· Transuranic waste excludes high-level waste as defined above and includes waste that contains
alpha-emitting transuranium (i.e., atomic number greater than 92) isotopes with half lives greater
than 20 years and concentrations greater than 100 nanocuries per gram. DOE also includes U-233
in its definition of transuranic waste.This waste usually consists of contaminated materials like
clothing and tools resulting from the manufacture of nuclear weapons.
· Low-level waste is radioactive waste that does not meet one of the definitions given previously.
There are two other classes of materials that DOE sometimes manages as waste:
· Nuclear materials, such as plutonium and special-use isotopes, that may be declared as surplus and
disposed of as waste.
· Contaminated environmental media, such as contaminated soil and groundwater, that may fall
under the Environmental Protection Agency's Comprehensive Environmental Response,
Compensation and Liability Act.The cleanup of this contamination may generate additional
radioactive and chemical waste streams that must be treated and managed.
In the United States, the federal government regulates the management and disposal of most types of
radioactive waste. Federal regulations seek to reduce to reasonably achievable levels the exposure of
workers and other members of the public to this waste.The guiding philosophy for waste management
is sequestration, that is, to isolate the waste from human populations and the environment, either
through long-term storage or disposal in an underground facility until it no longer poses a hazard.
C h a p t e r 2
23
OCR for page 24
remedial action, which is defined by DOE as the characterization and
cleanup of sites where contaminants or contaminated materials were
released into the environment. The cleanup of these sites will involve
the recovery and treatment of abandoned materials; remediation of soil,
grou ndwater and su rface water; and man itori ng where contam i nation
cannot be cleaned up to unrestricted release standards.
According to EM, site cleanup will be considered "complete" when,
among other things, releases to the environment have been cleaned up
in accordance with agreed standards and groundwater contamination
TABLE 2.3 Projectec Magnituce,Timing, anc Cost of DOE Cleanup Activities
DOE Site Projected Completion Soil, Pre-2006 Post-2006 Residual Residual
End State(s)a Date of Groundwater, Life-Cycle Life-Cycle Conta- Conta-
Planned end Other Costs Costs minants minants
Cleanup Media Requiring (1998 $B) (1998 $B)b in Soil in Water
Projects Remedial Action
(1 o6 ma)
u u
vie ~ ,,, —
~ ~ ~ a
Hanford IM, other TBD 2046 1,400 13 37.4
Idaho UR, RR, IM 2050 4,700 5.1 11.3
Nevada RR, IM, 2014 3.1 d 0.92 1.3
Test Site otherTBD
& Other
Associated
Sites'
Oak Ridge & UR, RR, IM 2013 31 5.4 7.7
Associated
Sitese
Rocky Flats UR, RR, IM 2006-2010 0.79 5.3 0.96
Savannah IM, other TBD 2038 1 72 1 2 1 7.7
River
Other Sites UR, RR, IM, 1999-2038 120 7.8 2.8
other TBD
Totals 6,400 50 79
aUR = unrestricted release; RR = restricted release; IM = long-term institutional management;TBD = to be determined.
bPost-2006 cost estimates include some but not all costs for long-term institutional management.
'Includes the Nevada Test Site and eight off-site locations in five states (Alaska, Colorado, Mississippi, Nevada, and New Mexico) where under-
ground nuclear tests were conducted.
Estimate does not include groundwater contaminated by nuclear testing.
elncludes the Oak Ridge Reservation, the Paducah and Portsmith Gaseous Diffusion Plants in Kentucky and Ohio, respectively, and the Weldon
Spring Site in Missouri.
SOURCE: Compiled from DOE (1 998a, 1999).
S U B S U R F A C E S C I E N C E
OCR for page 25
has been contained or long-term treatment or monitoring has been put
in place (DOE, 1 998a, p. 1-7~. In other words, even after EM has com-
pleted its cleanup projects there will still be contaminants left in the
subsurface and in surface land-disposal facilities that will require long-
term management and possibly future actions to prevent further spread.
Examples of Subsurface Contamination
Problems at Ma jor DOE Sites
The committee received several briefings on soil and groundwater
contamination problems and remediation activities at five of the six
major DOE sites (see Sidebar 2.1~: Idaho, Hanford, Nevada, Oak Ridge,
SIDEBAR 2.4 NON-AQUEOUS PHASE LIQUIDS IN HETEROGENEOUS FORMATIONS
Non-aqueous phase liquids (NAPLs) are a common class of subsurface contaminants at many DOE
sites. Dense non-aqueous phase liquids (or DNAPLs) are organic chemicals such as trichloroethylene,
tetrachloroethylene, and polychlorinated biphenyls that have densities greater than water (i.e., ~ 1.0
gram per cubic centimeter) at standard temperature and pressure and have low solubilities.Their rela-
tively high density causes them to migrate downward through soils and groundwater under the influ-
ence of gravity.When they encounter a low-permeability layer, they may pool or move laterally.
Because of their low solubilities, NAPLs remain as a separate phase and may provide a long-term
source of groundwater contamination.
The detection, characterization, and remediation of DNAPL contamination is generally difficult for a
number of reasons, including geological heterogeneity; complex physical, chemical, and biological
interactions; lack of efficient and cost effective field characterization techniques; and limitations and
unavailability of properly validated modeling tools for the design and evaluation of remediation tech-
niques. Experimental studies (e.g.,Schwille, 1988; Keeper and Fried, 1991; lilangasekare and others,
1995) have shown that geologic heterogeneity can cause lateral spreading, preferential flow, and
DNAPL pooling. In fact, such heterogeneities may be the major factor in controlling the entrapment
distribution of DNAPLs in the subsurface.The DNAPL may exist as discontinuous, stable pore-scale
masses trapped in soils under capillary forces, but it may also exist as an immobile continuous phase
trapped by various heterogeneity features.
Researchers (e.g., Pfannkuch, 1984; Schwille, 1988) have identified two geometries associated with
subsurface DNAPL contamination: (1 ) cylinders or fingers, and (2) pools on impermeable layers or
bedrock.The experimental work by Illangasekare and others (1995) and the conceptual studies by
Hunt and others (1 986a,b) demonstrate that other geometries are possible as well, including zones of
high saturation trapped in coarse lenses below the water table; thin pools trapped in coarse sand lay-
ers; and suspended pools trapped on top of fine sand or clay layers.
C h a p t e r 2
25
OCR for page 36
SIDEBAR 2.7 EFFECTS OF SUBSURFACE HETEROGENEITY ON FATE AND TRANSPORT MODELING AND
REM EDIATI ON
Lawrence Livermore National Laboratory, a DOE facility in California, overlies groundwater contami-
nated with volatile organic chemicals originating from land disposal of chemicals when the site was
used as a naval airfield in the 1940s.There are multiple contamination zones corresponding to differ-
ent disposal locations, consisting primarily of dissolved trichloroethylene and perchloroethylene
groundwater contaminant plumes.The western-most plume stretches for over a mile and is of concern
because it is migrating slowly toward municipal water supply wells in the city of Livermore. For over 10
years the site has been subject to intensive hydrogeologic investigation and remedial action (Thorpe
and others, 1990). As a result, hundreds of monitoring wells have been installed to provide for geologic
characterization of the site, monitor the composition and flow of groundwater, and support the design
and implementation of remediation technologies.
To more clearly understand the role and effects of geologic heterogeneity on remediation,Tompson
and others (1998) used hydraulic conductivity data from 240 of these monitoring wells to construct a
statistical distribution depicting the heterogeneous aquifer beneath the site. For a given realization of
this distribution, together with various boundary conditions used to reflect remedial (associated with
a remedial pumping well) or ambient conditions, groundwater flow paths can be produced using a
finite difference flow model.
To illustrate the effects of the fine-scale heterogeneity on contaminant transport and remedial recov-
ery, hypothetical contaminant pulses were released in each model realization to evaluate plausible
migration scenarios over 40 years of ambient conditions and then over 200 additional years of remedi-
al pumping from a well located 1,000 meters from the original source. Model runs indicated a wide
range of possible outcomes from one realization to the next.When the total pumping time was
allowed to run for 200 years, in some cases most of the contaminant mass was recovered from the
model domain,whereas in other realizations as little as one-third of the input mass was recovered.This
indicates the drastic effect that spatial variability of aquifer materials the exact distribution of which
is never known in precise detail can have on predictions of contaminant transport.The variation in
the results is indicative of the real uncertainty that would be expected for the behavior of a natural
system.
Significant uncertainties in understanding of the inventory, distribu-
tion, and movement of contaminants in the unsaturated zone exist at
Hanford. Further, attempts to model contaminant fate and transport
there have met with mixed success. Inaccurate models can have disas-
trous consequences when they mislead treatment or containment strate-
gies. Therefore, improved models for predicting contaminant migration
are needed to evaluate the impact of such releases into the environ-
ment. These models must be based on a good understanding of the sub-
surface features that control contaminant fate and transport (e.g., see
Sidebar 2.7), as well as important transport processes.
S U B S U R F A C E S C ~ E N C E
36
OCR for page 37
Metal and Radionuclide Contamination at Idaho
An important mission at the Idaho site was chemical processing of
spent fuel from research and naval reactor programs. After chemical
processing, the high-level liquid waste was stored in underground
tanks. Idaho managers recognized early on that tank storage space
wou Id be insufficient, so the site developed a faci I ity to convert the
waste into a powdered ceramic, or calcine, that could be more safely
handled and stored. Consequently, Idaho was able to avoid the inten-
tional discharge of high-level liquid wastes into the subsurface.
There have nevertheless been several releases of radionuclides and
metals from the single tank farm that supported the site's chemical pro-
cessing facility. An underground waste transfer line was accidentally
ruptured by drilling, and up to 13,700 liters (~3,600 gallons) of high-
level waste with a total activity of over 32,000 curies was released into
the unsaturated zone between 1956 and 1974. In 1972, another leak in
the tank farm released about 52,900 1 iters (~1 4,000 gal ions) with a total
activity of about 28,000 curies. The major contaminants include
chromium, mercury, cesium, strontium, plutonium, and iodine. Some of
this waste is located in a perched water zone beneath the tank farm,
but the extent of waste migration is poorly known.
The Idaho site is characterized by a thick unsaturated zone (see
Table 2.2), but this zone overlies one of the largest aquifers in the west-
ern United States, the Snake River aquifer, which covers an area of
about 26,000 square kilometers (10,000 square miles). This aquifer sup-
plies water to most of central Idaho and provides a major source of
recharge to the Snake River. Protection of the aquifer and the river is a
high priority at the Idaho site and is driving many of the site's remedia-
tion decisions. Decisions about remediation of the radionuclide conta-
mination beneath the tank farms is hampered by a lack of information
about the distribution of contamination, as well as the physical and
chemical characteristics of the unsaturated zone.6
Contamination in the Saturated Zone
The saturated zone is defined as that part of the subsurface where
pore spaces are filled with water. In unconfined aquifers, the top of the
saturated zone defines the groundwater table. The principal saturated
zone contamination problem across the DOE complex are contaminated
groundwater plumes (i.e., large volumes of groundwater contaminated
with dissolved and complexed chemicals, metals, and radionuclides).
6The committee was told that the least expensive remediation alternative would
cost about $600 million and would involve removal of the perched water zone
and pump-and-treat remediation of the underlying aquifer.
C h a p t e r 2
37
OCR for page 38
These plumes have been formed by the injection or migration of waste
into moving groundwater and have length scales on the order of kilo-
meters to tens of kilometers, depending on the nature of the source and
the rate and direction of groundwater movement.
All of the major DOE sites contain contaminated groundwater
plumes, and in some cases these plumes have migrated off site or are
discharging into surface waters. The following examples from the
Savannah River, Nevada, Hanford, and Idaho sites are illustrative of
plume-related problems across the DOE complex.
DNAPt Plumes at Savannah River
The Savannah River Site contains dozens of groundwater plumes
containing DNAPLs, metals, and radionuclides, but the DNAPL plume
in the Administrative and Materials Manufacturing Area is perhaps most
interesting because of its size and location. That area comprises about
140 hectares (350 acres) in the northern portion of the Savannah River
Site and is located less than a mile from the site boundary. Currently a
research and development center, the area was first established for the
manufacture of production reactor components, including target assem-
blies and fuel rods (Westinghouse Savannah River Co., 1995~.
From the 1 950s through the early 1 980s, contaminated wastewater
from fuel and target manufacturing was pumped through an under-
ground line into a settling basin, which had a capacity of about 30 mil-
lion liters (8 million gallons). The basin overflowed periodically into a
natural seepage area and a shallow depression known as Lost Lake and
released approximately 1.6 million kilograms (3.5 million pounds) of
solvents (principally trichloroethylene and tetrachloroethylene) and
heavy metals to the environment. DOE believes that most of the heavy
metals were trapped in the soil and about half of the solvents evaporat-
ed, while the remainder migrated downward from the seepage areas
into the groundwater (Westinghouse Savannah River Co., 1995~. In this
part of the site the groundwater moves at rates ranging from a few cen-
timeters to about 90 meters per year.
DOE has instal led some 400 monitoring wel Is since 1 981 to track
the spread of contamination, and based on these monitoring data and
model ing studies, scientists at the Savannah River Technology Center
have created a three-dimensional representation of the plume. DOE has
installed a pump-and-treat system at the downstream toe of the plume
to halt its further spread. DOE has been unable to locate or remove the
DNAPL sources that are feeding this plume or to apply effective reme-
diation technologies to the plume itself; it therefore faces the prospect
of long-term institutional management of this contamination, including
pump-and-treat remediation.
S U B S U R F A C E S C ~ E N C E
OCR for page 39
.
Radionuclide Contamination at the Nevada Test Site
Over 925 nuclear tests were conducted at the Nevada Test Site
between 1 951 and 1 992 and resu Ited i n the emplacement i nto the sub-
surface of several hundred million curies of radioactivity, including
significant quantities of tritium, plutonium, and fission products (see
Table 2.4~. Many of these tests were conducted at or below the ground-
water table. Nevada officials contend that the site contains more conta-
minated media than any other site in the DOE complex (Walker and
Liebendorfer, 1 998~. DOE notes in Paths to Closure (DOE, 1 998a,
p. E-56) that it has no plans to remediate the subsurface in and around
the underground tests because "cost-effective remediation technologies
have not yet been demonstrated."
TABLE 2.4 Isotope Inventories from Uncergrounc Testing at the Nevaca Test Site
Location Isotope Inventory (1 o6 curies)
(Numbers are rounded)
Pahute Mesaa Tritium 69.9
Cesium-1 37 1.95
Strontium-90 1.56
Krypton-85 0.13
Plutonium-241 0.09
Samarium-1 51 0.07
Europium-1 52 0.03
Plutonium-239 0.02
Europium-1 54 0.02
Others (34 isotopes) 0.05
Total Pahute Mesa 73.8
Non-Pahute Mesa Tritium 30.7
Potassium-40 24.7
Cesium-1 37 1.48
Strontium-90 1.19
Plutonium-241 0.10
Krypton-85 0.09
Europium-1 52 0.06
Samarium-1 51 0.05
Europium-1 54 0.05
Plutonium-238 0.03
Plutonium-239 0.01
Others (32 isotopes) 0.04
Total Non-Pahute Mesa 58.5
aSee Figure 2.5 for locations.
SOURCE: Presentation to the committee by Robert Bangerter, DOE-Nevada Operations Office,
December 15, 1998.
C h a p t e r 2
39
OCR for page 40
SIDEBAR 2.8 PLUTONIUM MIGRATION AT NEVADA TEST SITE?
A potentially significant example of the deficiency in understanding subsurface radionuclide transport
processes was provided by Karsting and others (1999), who reported that they had detected plutoni-
um in groundwater at the Nevada Test Site. The plutonium was detected in water collected from moni-
toring wells on Pahute Mesa, near the northwestern border of the test site (Figure 2.5).The plutonium
was apparently being carried on colloids.The origin of the colloids and the plutonium geochemistry is
still uncertain.
Karsting and others were able to trace the plutonium to the Benham Test, which was detonated in 1968
in zeolitized bedded tuft at a depth below the surface of about 1,400 meters.This test is located about
1.3 km laterally and up to 600 meters below the monitoring wells.The origin of the plutonium was
identified from its 240Pu/239Pu isotopic ratio,which is distinctive for each underground test.The pluto-
nium ratio is recorded in the melt-glass collected from the underground test cavities. No evidence was
found for migration of plutonium from other nearby tests.
The suggested transport of plutonium at the test site has potentially significant implications for DOE's
plans to passively manage contaminants there, especially if plutonium transport proves to be more
pervasive than is currently recognized.This discovery also has potentially significant implications for
the underground disposal of nuclear waste. Conventional wisdom suggests that plutonium is relatively
immobile in oxidizing subsurface environments like at the test site and has strong sorbing tendencies.
Indeed, underground tests at the test site were believed to demonstrate the effective fixation of pluto-
nium in subsurface environments.The work by Karsting and others has demonstrated that the concep-
tual models for plutonium migration are incomplete; it also suggests that additional basic research on
the geochemical behavior of plutonium is required.
Tritium is very mobile in groundwater, and large plumes of tritium
have been detected from many of the underground tests. It has long
been argued that most other radionuclides, and especially plutonium,
are relatively immobile due to their low solubilities in groundwater and
strong sorption onto mineral surfaces. As discussed in Sidebar 2.8, how-
ever, recently published work challenges this conventional view.
Mixed Contaminant Plumes at Test Area North
Test Area North at the Idaho N ational Engi neeri ng and Envi ran mental
Laboratory covers about 50 hectares (125 acres) in the northern part of
the site and was used to support the Aircraft Nuclear Propulsion
Program between 1954 and 1961. From 1960 through the 1 970s, the
area housed the Loss-of-Fluid Test Facility, which was used for reactor
safety testing and behavior studies. The primary source of the contami-
nated groundwater plume is the Technical Support Faci I ity injection
wel 1, which was used from 1 953 to 1 972 to inject I iquid wastes directly
into the Snake River plain aquifer. The contaminants included raw
sewage, trichloroethylene, tritium, strontium-90, and cesium-137.
S U B S U R F A C E S C ~ E N C E
40
OCR for page 41
Although the source area for this plume the injection wel I is
known, the source term is not. Moreover, the subsurface in this region
consists of highly fractured rock, which makes it difficult to locate and
characterize the contamination. Characterization of the extent of conta-
mination began in 1988, and recent data suggest that most of the con-
tamination probably occurred as entrained sludge in two major fracture
zones (see Figure 2.6~.
Contaminant Plumes at the Hanford Site
DOE estimates that groundwater under more than 220 square kilo-
meters (85 square miles) of the Hanford Site is contaminated above cur-
rent standards, mostly from operations in the 100 and 200 Areas (Plate
3~. The 100 Area is located on about 6,900 hectares (17,000 acres) in
the northern section of the Hanford site and contains nine production
reactors and several waste burial sites (DOE, 1996~. The main sources
of subsurface contamination in the 100 Area are from radionuclide
(mainly tritium) contaminated reactor cooling water and metal and
DNAPL contaminants from operations and disposal. Contamination in
the 200 Area was discussed in the section on the unsaturated zone ear-
lier in this chapter.
Disposal of supernatant liquids into the ground and leaks from the
high-level waste tanks have produced significant contamination of the
saturated zone in the 200 Area (Gephart and Lundgren, 1998~.
Groundwater plumes of the following contaminants exist at levels
exceeding current drinking water standards at the 200 Area: tritium,
strontium-90, technetium-99, iodine-1 29, carbon tetrachloride, chromi
um, and uranium. The plumes are flowing northeast toward the
Columbia River at several tens of meters per year (see Figure 2.7~.
Nevada
\ Test Site
Repute Mesh
\L . ~ .' ." ~-
I Nevada
~ ,i~ ~
1\ 1\
.~
. . _ ~ _
Nevada\t
Test
Site
· = Underground I
nuclear test
~ 1 ~
l l l
0 15km
C h a p t e r 2
\
Benham to
Molbo \
~ \O
Tybo l
XER-20-5
\ ~ Belmont
1 ~ 1
0 1200 2400 m
, _
1~
~ ID
1 In
103
FIGURE 2.5 Plan view of
Pabute Mesa with location
of the Benham Test and
groundwater collection well
cluster ER-20-5. SOURCE:
Karsting and others (1999).
OCR for page 42
FIGURE2.6 Conceptual
model for subsurface conta-
mination at TestArea North
at the Idaho Site. Dense
non-aqueous phase liquids
(DNAPLs) may be entrained
in fractures and perched on
dense basalt flows and sedi-
mentary interbeds. SOURCE:
Idaho Engineering and
Environmental Laboratory.
Not to Scale
1 ...... ~.2 ~ ......... .
in. ~
TSF-05 injection well
s 'A ~2su'd'-c'ie'l so-
!_ & ~ ~
I Inlerlayereu casalt Claw
~1 1 1 ~~-
~ 1-
P-Q Inlerbed _
! ~ —
inter, !!~
_1 ~ _
_ _
=t
of I OW!
·:- l.
i..~
.. ....
· .-
· · ·:
. ...
....
. . .
.. _ .
. ;'
,
s
..
_. ~ .
~ 11
~ ~1~
It - ~
Possible pooled DNAPL on
_ _ 7`inlerbed moving info fractures | _
~ 1~-
_ ~
t
rlPt~llnclwntpr Girl - rlir~r'lir~n _
J .. . _ . :
~ -
| i; ~ ga~ TCE p bole
; ~ ., 1 ..
I -: ;1
.::;.. 1
Possible DNAPL Pooled
Din fract~,,: fractured and fissured flow margins
~ dense central portion of basalt fInw
Q-R Inierbed
DOE has established an extensive network of monitoring wells to
track the movement of the groundwater plumes, but very little remedia-
tion work is being done at present. DOE has established a groundwater
extraction well network to intercept a chromium plume in the 100
Area. The ch rom i u m is extracted usi ng ion exchange and the treated
water is returned to the aquifer. Pump-and-treat systems also have been
established in the 200 Area to contain the highest concentrations of a
uranium and technetium-99 plume and a carbon tetrachloride plume
(DOE, 1 998b).
DOE has a very poor understanding of the source areas, amounts,
and timing of contaminant discharges into the subsurface at Hanford.
~ ~ . . . .
UL)t IS beginning to support "forensic" investigations of past waste
releases to the subsurface (e.g., Agnew and others, 1997), but additional
work will be needed to improve the knowledge of the extent and mag-
nitude of subsurface contamination at the Hanford site. Improvements
in understanding and modeling fate and transport processes in the sub-
surface is also needed to provide long-term predictive capabilities.
S U B S U R F A C E S C ~ E N C E
OCR for page 43
Conclusions
The examples provided in this chapter illustrate that subsurface con-
tamination is an enormously difficult cleanup problem as well as a sig-
nificant challenge to science. Much of the subsurface contamination at
DOE sites is poorly characterized and widely dispersed in the environ-
ment, making it very expensive or technically impractical to treat effec-
`,3 2; so ~ ~ i:-i3s3sss¢~$
?~ ~ ~~ ~i.
4 ~ Is; ~ ~ is ~:~-
~ ~ off
FIGURE 2.7 Plan view show-
ing the fast spread of tritium
plumes from the 200 East
Area at the Hanford Site to
the Columbia River. SOURCE:
Richland Operations Office.
sow- z ~:03~Li~:d B~333,3t fi3Xz~=
Liz' ''I.
otiose ~~#tS~ C;33
.,.. ~~ o~.~
2 4 ~ ~ kzinm~s.~:
~'~1
~ ~ ~N ~
$ 2 4 ~ zoom ~ ~ - ` ~$
~
C h a p t e r 2
), ~ SSS 3: A_ :~x
§~3is33sis - ~ it;
3.\J~ ~
OCR for page 44
tively with current technologies. Moreover, the contamination that can-
not be removed or effectively isolated from the environment will
require long-term management, which represents a potentially large
future mortgage for the nation.
· ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ..
SIDEBAR 2.9 BASIC SCIENCE CAN IMPROVE ENVIRONMENTAL MANAGEMENT
Basic scientific research can provide several benefits to waste management efforts if it is properly
focused on difficult cleanup problems (see Chapter 6). Basic research can produce new scientific knowl-
edge and engineering tools to improve the effectiveness of cleanup efforts, lower cleanup costs,
reduce risks to worker and public health, and improve environmental quality. Equally important, basic
research can help improve current waste management practices and thereby reduce the likelihood of
future environmental insults. Scientific studies in the 200 Area at Hanford provide a simple yet com-
pelling illustration of the potential benefits for environmental management.
The 200 Area is comprised of two major operating zones (200 East and 200 West) that contain a variety
of waste disposal and waste storage facilities (see Plate 3).These facilities,which include drainage
cribs, settling basins, and underground tanks, are major contributors to the site's groundwater contam-
ination. As discussed elsewhere in this chapter, groundwater contaminant plumes have formed
beneath both areas, but the plumes originating from the 200 East Area are significantly larger in size,
extending some 15 kilometers (9 miles) to the Columbia River (see Figure 2.7).
Basic geological research conducted at Hanford (see Reidel and others [1992] and DOE [1998b] for a
summary of the Hanford geology) suggests that plume size is controlled to a large extent by the physi-
cal and chemical properties of the geological formations underlying the 200 Area.The 200 East Area is
underlain by the Hanford Formation, which is comprised of permeable sands and gravels that provide
relatively direct pathways to the groundwater some 100 meters below the surface. The 200 West Area,
on the other hand, is underlain by the Ringold Formation, which consists of less permeable sands, grav-
els, and clays that provide a barrier to widespread contaminant migration.
These findings provide a compelling demonstration that~geology counts" in waste management and
site remediation, and that locating disposal facilities must take account of subsurface properties as
part of a defense-in-depth waste containment strategy.' DOE is constructing and operating several
facilities in the 200 Area to dispose of a variety of cleanup and defense wastes. It recently sited a large
land disposal facility (the Environmental Restoration Disposal Facility) in 200 West to manage certain
types of chemically and radioactively hazardous cleanup wastes from other parts of the Hanford Site.
At least two other disposal facilities have been constructed or are planned for the 200 East Area: the
Naval Reactor Disposal Facility, which contains nuclear reactors from decommissioned U.S. Navy sub-
marines, and the planned Immobilized Low-Activity Waste Disposal Facility,which will take low-activity
waste generated during processing of high-level waste from the Hanford tanks. If the past is a guide to
the future, the disposal facilities in the 200 East Area may create new site contamination problems that
will require additional remediation efforts.
HA defense-in-depth waste containment strategy uses multiple artificial or natural barriers to improve the
long-term performance of the containment system.
S U B S U R F A C E S C ~ E N C E
44
OCR for page 45
The committee believes that this future mortgage could be reduced
significantly through the development of new and improved technolo-
gies to locate, remove or contain, and monitor subsurface contamina-
tion at DOE sites. However, the development of such technologies will
require advances in basic understanding of the complex natural systems
at DOE sites and also in understanding the nature of contaminant
"insults" to those systems. The report of the NRC Committee on
Building an Effective Environmental Management Science Program
(NRC, 1 997b, p. 22) concluded that "new technologies are required to
deal with EM's most difficult problems, and new technologies require
new science." The present committee agrees with this statement and
notes that, given the long-term nature of the cleanup mission and its
projected cost (see Chapter 1), DOE has necessary cause and time to
do the required basic research to support the development of these
needed technologies (see Sidebar 2.9~.
C h a p t e r 2
45
OCR for page 46
S U B S U R F A C E S C I E N C E
46
Representative terms from entire chapter:
savannah river
