DOE's second rebuttal witness was DR. NEIL WILLIAMS who testified March
24-25, 1999. Dr. Williams testified on direct during DOE's presentation on March 1, 1999.
TESTIMONY OF DR. NEIL WILLIAMS - March 24-25, 1999.
Dr. Williams referred to a color cross-section poster showing the three
areas in and near the WIPP site: the Land Withdrawal Area, the Margin
Area, and Nash Draw. Each area has separate hydrological characteristics,
including water locations and rates of migration. Conditions found in one
area are not necessarily found in another.
LAND WITHDRAWAL AREA (WIPP FACILITY)
This area is often moisture efficient. First rain would increase the
moisture of the soils and then there might be vertical migration. The
Santa Rosa Formation is cemented in this area. There are sandstones,
siltstones, claystones and conglomerates with varying conductivities. The
Dewey Lake Redbeds are finer grain sandstones with silt and clay in this
area. Conductivities are less than in the Santa Rosa Formation. The Dewey
Lake is fractured in much of this area. The fractures are filled with
gypsum and anhydrite, which lowers the conductivity so the water perches on
top of the Dewey Lake at the WIPP site. There is no water in the Dewey Lake
except right at the top where the fractures are not so filled.
The DOE never said there was no moisture migration through other members of
the Rustler besides the dolomites (the Magenta and the Culebra). It just
takes a long time. The dolomites have high conductivities. The dolomites
are highly fractured and highly transmissive and extend across the site.
The Salado is halite, impure halite, anhydrite and clay layers. All have
very low conductivities and are barriers to liquid migration.
Water migrates through soils, then through the Santa Rosa Formation and
accumulates at the low permeability barrier of the Dewey Lake Redbeds in
the weathered or fractured top of the Dewey Lake where it perches. The
WIPP facility itself causes this. This horizon was not saturated before
construction was started. Then the DOE excavated basins for water in the
caliche. The DOE could get rid of this water by lining the ponds, pumping
water from the formation, or in other ways. The vertical conductivity of
this area is low. The horizontal conductivity in this area is higher so
the water will spread laterally.
The Margin Area is between Nash Draw and the site. The primary feature in
this area is that the Santa Rosa Formation is absent. This fact reduces
the confining stress on top of the Dewey Lake Redbeds that impacts the area
where the cracks are filled. In the top, the fractures are open and the
conductivity is much higher. About 180 feet bgs (below ground surface) is
the boundary between the fractures filled with gypsum (above) and the
fractures filled with anhydrite (below). The anhydrite has even lower
conductivity. The halite member in the Rustler has undergone dissolution
in the Margin Area. This appears to have occurred when the formation was
deposited. There is some dissolution along the boundary that continues
today, but this is only within a foot or two of the deposition boundary.
In 200,000,000 years the boundary will have moved only a few feet at most
and will not move toward the repository at a very fast rate.
Completing WIPP-33 would have no useful purpose. We are concerned with what
happens within the WIPP site and out to the compliance point. More than
adequate work has been done to show that the DOE will find contamination if
the contamination leaves the repository.
NASH DRAW AREA
The Nash Draw Area is different from the other two areas. There is
significant dissolution of the upper Salado within the Nash Draw Area.
This has left a residue. Overlying units have vertical fractures, higher
infiltration rates, solution cavities, and sinkholes, etc. But the
conditions in the Nash Draw Area have no impact on water flow in the Land
The first primary water-bearing zone above the repository is the Culebra.
It is very well understood and is the first primary pathway for lateral
migration away from the repository. Though there is some water at the
interface between the Rustler and the Salado, the conductivity is low. The
interface area is not a water-bearing zone that could be a primary pathway.
The Culebra is also the most productive water-bearing zone above the
repository. The shafts are connections from the repository to the surface.
The shafts do not constitute a pathway to the Culebra either before or
after they are closed. To be able to migrate to the Culebra, the shafts
would have to be filled with water. There is not enough water to do this.
The materials in the shaft seals are chosen for both short- and long-term
Dr. Williams compared the most conventional lining systems. The Salado is
great for containment. Dr. Phillips said pressurized brine could be a
driving force to push contaminants out of the repository. Dr. Williams
stated that there is not enough brine coming into the repository to
saturate the waste. Even Dr. Bredehoeft said the brine level would not
reach up to the waste material for at least 2,000 years. The repository
has been open for 10 years and the observed brine inflow rate is much lower
than was anticipated. The brine comes from the consolidation of clay in
the anhydrite layers. Also, the EPA looked at water injection as a breach
scenario and concluded that it was a very low probability phenomenon.
Potentiometric data show that there is a southerly flow, not westerly as
Dr. Phillips claims. All units in Nash Draw could contain water because of
the fracturing, dissolution, etc. in that area. The Nash Draw, however, is
separate from the WIPP site conditions. There is very low conductivity in
the confining units at the site and the fractures are filled there.
The primary pathway out of the repository would be vertical to the Culebra.
Over time, contamination would move only 2.4 meters in 80 years and just
over 4 meters in 100 years out of the repository. The anhydrite layers
above and below the repository are not pathways because there are no
receptors at the end of the anhydrite beds. The anhydrite layers have very
low conductivities outside of the Disturbed Rock Zone (DRZ). The DRZ
extends out only 6 to 16 1/2 feet and changes as a function of time.
Marker Bed 139 has a very low impact on constituent migration. Marker Bed
139 is about 6 feet below the repository and has higher conductivities only
in the DRZ. Otherwise, the conductivities of Marker Bed 139 are about 10-9
or 10-14/sec. with an inward hydraulic gradient so it is a barrier, not a
There is water in the Santa Rosa Formation/Dewey Lake Redbeds interface.
This is not, however, the primary pathway. The primary pathway is from the
repository to the Culebra to the facility boundary, or the point of
compliance. For this reason, the Santa Rosa Formation does not need to be
There are no karst features in the Land Withdrawal Area. There are karst
features in the Margin Area and Nash Draw Area. These karst features are
too far away to affect the repository and do not need to be considered
Dr. Phillips' model of the regional water balance is flawed. There are a
number of assumptions in the model that are not consistent with conditions
in the WIPP site or even in the surrounding area. Dr. Phillips neglects
important components such as soil moisture, groundwater flow into and out
of the catch basin, etc. so the model is very inaccurate. Water balance is
not performed. Also, Dr. Phillips made mistakes in his calculations. Most
significantly, when he calculated the brine evaporation rate, he multiplied
1.17 times the fresh water evaporation rate. Dr. Phillips should have
divided. This leads to a 34% error in the brine evaporation rate. Dr.
Phillips also used an average annual precipitation rate of 14.21
inches/year when he should have used 12.67 inches/year. This is a 12%
difference. However, these two errors are small compared to neglecting
many important flow components.
Rainwater recharge is different in the WIPP site and in the Nash Draw Area.
In Nash Draw, there is rapid infiltration of almost fresh water. This does
not exist at the WIPP site, which is shown by the filled gypsum fractures
in the Dewey Lake Redbeds.
HIGHLIGHTS OF CROSS EXAMINATION OF DR. WILLIAMS
Migration out of the repository will only be a few meters during the RCRA
The basis for the distinction between the Nash Draw Area and the Land
Withdrawal Area is that the hydrology is different and the properties of
the materials are different
The Culebra is continuous so one would expect a response between WIPP-13,
WIPP-25 and WIPP-30 during the multi-well pump tests.
ERDA-9 is not drilled vertically so it could be closer than 93 feet from
the repository footprint at the bottom. It is possible that ERDA-9 is a
potential pathway before the WIPP site is closed.
The Santa Rosa Formation/Dewey Lake Redbeds water is a perched
water-bearing zone or a saturated zone above an unsaturated zone.
Unconfined potentiometric surface rises above the confining layer. It
would be better to call it a confined zone because perched is isolated.
Dr. Chaturvedi mischaracterized what the DOE said when he claimed the DOE
identified the water table in the Dewey Lake Redbeds for the EPA.
The source of water in the exhaust shaft is immaterial because the amount
of liquid coming down the shaft is 0 to 100 gallons/week. A maximum of
1,300 gallons/week could be managed in the repository. Water has been
coming in since 1995 when it was discovered. More boreholes are needed to
assess the extent of the 80 acres. No one has done a water balance study
to test the hypothesis that the water is a local occurrence.
Dr. Williams studied RCRA compliance for the WIPP site for about one year.
The Idaho National Engineering and Environmental Laboratory (INEEL) opened
9,000 drums of waste in order to inspect for liquids. The waste was
dispositioned as WIPP certified waste. The report on these wastes does not
state that the waste drums were opened, but that they were "examined." Dr.
Williams interprets that to mean that the waste drums were opened and that
the DOE measured the average liquid content.
Dr. Williams has looked at deep geologic disposal of mine ash in coalmines.
WIPP is the first deep geologic disposal facility in the United States for
mixed transuranic (TRU) waste. Dr. Williams has not evaluated anything at
WIPP other than what is at the site. Dr. Williams stated that WIPP is the
best containment of all the facilities he knows of. Dr. Williams has not
observed RCRA disposal of TRU mixed waste elsewhere because WIPP is the
first disposal facility for TRU mixed waste.
Early on the DOE identified about 17 potential deep geologic disposal sites
around the country. Dr. Williams has not compared the containment at WIPP
with any of these 17 potential sites.
Most mixed waste facilities have fingerprinting, but it depends on the
facility. The Fernald Mixed Waste Facility is on-site so the facility is
where the wastes are already located. [The DOE's Fernald site processed
various uranium compounds into uranium metal. The Fernald site is located
20 miles northwest of Cincinnati, Ohio.] If the waste is outside of
Fernald's Waste Acceptance Criteria (WAC), however, the waste has to go
Dr. Williams would think that pyrophoric materials would be segregated out
and disposed of separately. [Pyrophoric materials are materials that
ignite spontaneously upon exposure to air.]
The Santa Rosa Formation is not present in Nash Draw or the Margin Area.
The Dewey Lake Redbeds is present in all three locations. The conductivity
of the Dewey Lake Redbeds has been measured at other places than WQSP-6A,
but Dr. Williams did not know where.
Dr. Williams' color cross-section poster shows halite in the lower un-named
member across the whole site. The isopach map from Sandia Report 82-1069,
which Dr. Powers co-authored, does not this. Dr. Williams says he cannot
say why because the isopach map does not show the Land Withdrawal boundary.
When the witness was shown a map by Dr. Phillips that had the Land
Withdrawal boundary mapped, Dr. Williams said the boundary on his document
was clearly different from that on Dr. Phillips' map. The halite boundary
is close to the western boundary of the Land Withdrawal Area. Dr.
Williams' cross-section is idealized and he didn't want to quibble about a
few feet. Dr. Phillips said that the difference was a mile or more. Dr.
Williams then said the cross-section was only demonstrative and was not
It is difficult to look at a picture of core to identify the open
conditions that could have been caused by the drilling itself. It would be
necessary to look at the geotechnical logs and hydraulic data. It is
possible to get into real trouble by just looking at pictures of core. It
is necessary to look at the core itself. The core is located at the WIPP
Dissolution of halite occurred in 2 time frames. First, there was some
dissolution during or shortly after the deposition of the formation. More
recently, there was dissolution at the halite boundary that is of only
limited extent (a few feet). Some dissolution residues that were described
in the earlier work were reconsidered since the DOE drilled the shafts and
saw the exposed beds and logged them in detail. Where the DOE thought that
before they had dissolution residues, they found that this was not the
case. What the DOE looks for is evidence of recent dissolution along the
contact boundaries. This is only limited.
Turning WIPP-33 into a test well now would have no value for RCRA since it
is located outside of the compliance point boundary. It is necessary to
accept that WIPP is an operating facility and get on with it.
Dr. Williams was asked, "Since the caverns in WIPP-33 are not found in the
Culebra, from where does the water flow to these caverns?" The primary
pathway is not to WIPP-33. The Magenta is immaterial for the primary
pathway, which is southerly. Dr. Williams has not seen a flow path for the
Magenta but stated that the probability of the Magenta being a pathway is
so low that it does not warrant consideration. If fluids migrate through
the shaft, they will meet the Culebra which would be a hydraulic barrier
from the Magenta. The Culebra is the most transmissive formation. The
Culebra is first, so everything will go into it first. It is not
impossible though that the Magenta could be a pathway.
The brine aquifer is absent in the Land Withdrawal Area. It is possible,
however, that there is a thin layer of dissolution residue extending into
the Margin Area that shows up in H3, H6, P6, P12, P13, P14, and WIPP-33,
Low Total Dissolved Solids (TDS) in the water in the Land Withdrawal Area
could cause dissolution within the Rustler. However, there are completely
different conditions in the site area and Nash Draw. The Dewey Lake
Redbeds area a very tight formation with gypsum filling in fractures, low
vertical conductivity, and very low lateral conductivity. There is no
direct recharge of the Rustler where there is a tight Dewey Lake formation
in the Land Withdrawal Area.
Dr. Williams referred to brine inflow into the repository as the driving
force described by Dr. Phillips. Dr. Philips was actually describing brine
from the pressurized brine reservoir beneath the repository.
Dr. Williams said that the surface water divide is not consistent with the
ground water divide, although Dr. Phillips' described them as about the
same in his dissertation. Dr. Phillips' model is inconsistent, but Dr.
Williams did not know where the groundwater divide was in all locations.
There is southerly flow through the catch basin. Also there are open
contours in the Culebra within the catch basin. The Program Plan for
Hydrologic Investigation for the Los Medaņos Area states that a divide to
the east is suspected and that holes should be drilled from P-18. Dr.
Williams did not know why the wells were not drilled there.
If a correction were made to Dr. Phillips' calculation error, there would
be .73 times as much inflow. If 12.67 inches/year of rainfall were used in
the calculations, there would be .89 times as much rainfall as is stated in
Dr. Phillips' dissertation. That would mean the amount of water in Laguna
Grande would come from a smaller amount of rainfall. Dr. Williams
disagreed that this would mean the infiltration rate would increase as a
percentage of rainfall because, as Dr. Williams stated, Dr. Phillips' whole
model is completely wrong. Dr. Phillips never evaluated the percolation of
rainfall through arid soils, soil moisture, and changes in soil moisture
DOE's Comment No. 108 states that the point of compliance should be at the
Land Withdrawal boundary, not where the monitoring wells are going to be
located. Alternatively, the DOE would like the point of compliance to be
at the locations of the current monitoring wells. The point of compliance
would actually only be the down gradient WQSP wells.
WIPP-33 has caverns in the Magenta. It is hard to describe the direction
of groundwater flow in the Magenta at WIPP-33. The orientation of the
caverns would be the best indication, but Dr. Williams was not sure what