ML20057B710
| ML20057B710 | |
| Person / Time | |
|---|---|
| Issue date: | 09/30/1993 |
| From: | NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS) |
| To: | |
| Shared Package | |
| ML20057B708 | List: |
| References | |
| REF-WM-61 NUDOCS 9309230170 | |
| Download: ML20057B710 (77) | |
Text
-
v i
.s i
1 FINAL TECHNICAL EVALUATION REPORT FOR THE PROPOSED REMEDIAL ACTION AT THE GUNNISON, COLORADO URANIUM MILL TAILINGS SITE SEPTEMBER 1993 1
DIVISION OF LOW LEVEL WASTE MANAGEMENT AND DECOMMISSIONING U.S. NUCLEAR REGULATORY COMMISSION
)
9309230170 930916 P
PDR WASTE MM-b1 ppg
,a i
J
1 ABSTRACT This final Technical Evaluation Report summarizes the U.S. Nuclear Regulatory Commission staff's review of the proposed remedial action for the Gunnison uranium mill tailings site. The sections of the report are arranged by technical discipline to correspond to the Environmental Protection Agency's proposed standards in Title 40 of the Code of Federal Regulations, Part 192, Subparts A through C.
The NRC staff review of the preliminary final Remedial Action Plan (RAP) identified significant open issues in geology, geotechnical engineering, hydrology, radon attenuation and site cleanup (see Appendix A).
These open issues were resolved based on NRC staff review of additional information provided in the final RAP and other documents.
i i
4 a
4 e
l 1
4 I
r TABLE OF CONTENTS
_SJction Pagg
1.0 INTRODUCTION
1.1 E PA S t a n d a rd s.............................................
1.1 1.2 Site History and Proposed Action..........................
1.2 1.3 Re v i ew P ro c e s s............................................
- 1. 6 1.4 T E R O rg a n i z at i o n..........................................
- 1. 6 1.5 S umma ry o f Op e n I s s u e s...................................
- 1. 6
'1.6 Re fe r e n c e s................................................
- 1. 7 2.0 GE0 LOGIC STABILITY 2.1 Introduction..............................................
2.1 2.2 Location..................................................
2.1 2.3 Geology...................................................
2.1 2.3.1 Stratigraphic Setting...........,..................
2.2 2.3.2 Structural Setting.................................
2.6 l
2.3.3 Geomorphic Setting.................................
2.8 2.3.4 Seismicity.........................................
2.9 l
2.3.4.1 Seismotectonic Provinces....................
2.10 2.3.4.2 Regional and Site-Speci fic Faults...........
2.12 2.3.5 N a tu r al Re s o u rc e s..................................
2.13 2.4 Geologic Stability........................................
2.14 2.4.1 Bedrock Suitability................................
2.14 2.4.2 Geomorphic Stability...............................
2.14 2.4.3 Seismotectonic Stability...........................
2.15 2.5 C o n c l u s i o n s............................................... 2.16 2.6 References................................................
2.16 3.0 GE0 TECHNICAL STABILITY 3.1 Introduction..............................................
3.1 3.2 Site and Material Characterization........................
3.1 3.2.1 Site Description...................................
3.1 3.2.2 Site Investigations................................
3.2 3.2.3 Site Stratigraphy..................................
3.3 3.2.4 Testing Program....................................
3.4 3.3 Geotechnical Engineering Evaluation.......................
- 3. 5 3.3.1 Slope Stability....................................
3.5 3.3.2 Liquefaction.......................................
3.6 3.3.3 Settlement.........................................
3.7 3.3.4 Cover 0esign.......................................
3.7 3.4 Geotechnical Construction Criteria........................
3.9 3.4.1 Contaminated Materi al Pl acement....................
3.9 3.4.2 Radon Barrier Pl acement............................
3.10 3.5 Conclusions...............................................
3.10 3.6 References................................................
3.11 1
i i
D
^
4.0 SURFACE WATER HYDROLOGY AND EROSION PROTECTION 4.1 Hydrologic Description and Site Conceptual Design...........
4.1 4.2 Fl oodi ng De te rmi nati ons................................... 4.1 4.2.1 Selection of Design Rainfall Event.................. 4.1 4.2.2 In fil trati on lo s se s.................................
- 4. 2 4.2.3 Times of Concentration..............................
4.3.
4.2.4 Rainfall Distributions..............................
4.3 4.2.5 Computation of PMF..................................
4.4 4.2.5.1 Top and Si de Sl opes..........................
- 4. 4 4.2.5.2 Apron / Toe....................................
4.4 4.2.5.3 Permanent Interceptor Ditch..................
4.4 4.3 Water Surface Profiles and Channel Velocities.............. 4.4 4.3.1 Top and Side S1 opes.................................
4.4 4.3.2 Apron / Toe...........................................
4.5 4.3.3 Permanent Interceptor Ditch.........................
4.5 4.4 Erosion Protection.........................................
4.6 4.4.1 Sizing of Erosion Protection........................
4.6 4.4.1.1 Top and S i de Sl ope s..........................
- 4. 6 4.4.1.2 Apron / Toe....................................
4.6 4.4.1.2.1 Lower Side S1 ope......................
4.6 4.4.1.2.2 Toe...................................
4.7 t
4.4.1.2.3 Collapsed Slope.......................
4.7 4.4.1.2.4 Natural Ground........................
- 4. 7 4.4.1.3 Interceptor Ditch............................
4.7 4.4.1.3.1 Ditch Side Slopes.....................
4.8 4.4.1.3.2 Ditch.................................
4.8 4.4.1.3.3 Ditch 0utlets.........................
4.8 4.4.1.3.4 Sediment Considerations...............
4.9 4.4.2 Rock-Durability.....................................
4.9 4.4.3 Testing and Inspection of Erosion Protection........
4.10 4.5 Upstream Dam Failures......................................
4.11 4.6 C o n cl u s i o n s................................................ 4.11 f
4.7 References.................................................
4.11 l
5.0 WATER RESOURCES PROTECTION 5.1 Introduction...............................................
5.1 5.2 Hydrogeologic Characterization.............................
5.1 i
i 5.2.1 Identification of Hydrogeologic Units...............
5.2 5.2.2 Hydraulic and Transport Properties..................
5.3 i
5.2.3 Geochemical Conditions and Extent of Contamination.. 5.5 5.2.4 Water Use...........................................
5.7 5.3 Conceptual Design Features to Protect Water Resources......
5.8 5.4 Disposal and Control of Residual Radioactive Materials.....
5.9 5.4.1 Water Resources Protection Standards for Disposal...
5.9 5.4.1.1 Hazardous Constituents.......................
5.9 5.4.1.2 Concentration Limits.........................
5.10 5.4.1.3 Poi nt of Compl i ance.......................... 5.12 5.4.2 Pe rfo rmance As se s sment..............................
5.12 5.4.3 Closure Performance Demonstration...................
5.13 5.4.4 Ground-water Monitoring and Corrective Action Plans.
5.13 11 i
i 1
L
\\
i 5.5 Cleanup and Control of Existing Contamination..............
5.14 5.6 Conclusions................................................
5.15 5.7 References.................................................
5.15 6.0 RADON ATTENUATION AND SITE CLEANUP 6.1 Introduction...............................................
6.1 6.2 Radon Attenuation..........................................
6.1 6.2.1 Eval uati on of Parameters............................
6.1 6.2.2 Evaluation of Radon Attenuation Model............... 6.5 6.3 Site Cleanup...............................................
6.7 6.3.1 Radiol ogical Site Characterization.................. 6.7 6.3.2 Cleanup Standards...................................
6.7 6.3.3 Verification........................................
6.8 6.4 C o n cl u s i o n s................................................ 6. 8 6.5 References.................................................
6.8 APPENDIX A - Status of Open Issues t
LIST OF FIGURES Number Pace 1.1 Gunnison Site Location Map...................................
1.3 1.2 Gunnison Processing Site.....................................
1.4 i
1.3 Locations of the Gunnison Processing, Disposal, and Borrow Sites...............................................
1.5 2.1 Site-Specific Geology of the Gunnison Disposal Site..........
2.3 2.2 Geologic Cross Section of the Gunnison Disposal Site.........
2.5 l
2.3 Structural and Seismotectonic Provinces of the Gunnison Region.....................................................
2.7 3.1 Gunni son Di sposal Cell and Cover............................. 3.8 TABLE 5.1 Proposed Concentration Limits for the Gunnison D i s p o s a l C e l l............................................
5.1 1 l
iii i
j
1.0 INTRODUCTION
The Gunnison site was designated as one of 24 abandoned uranium mill tailings piles to be remediated by the U.S. Department of Energy (DOE) under Title I of the Uranium Mill Tailings Radiation Control Act of 1978 (UMTRCA). UMTRCA requires, in part, that the U.S. Nuclear Regulatory Commission concur with DOE's selection of remedial action, such that the remedial action meets appropriate standards promulgated by the U.S. Environmental Protection Agency (EPA).
This Technical Evaluation Report (TER) documents the NRC staff's review of the DOE Final Remedial Action Plan (RAP) and associated documents (DOE, 1992a-d) (MK-F, 1993).
1.1 EPA Standards As required by UMTRCA, remedial action at the Gunnison site must comply with regulations established by EPA in 40 CFR Part 192, Subparts A-C.
These regulations may be summarized as follows:
1.
The disposal site shall be designed to control the tailings and other residual radioactive materials for 1000 years to the extent reasonably achievable and, in any case, for at least 200 years [40 CFR 192.02(a)].
2.
The disposal site design shall provide reasonable assurance that releases of radon-222 from residual radioactive materials to the atmosphere will not exceed 20 picocuries/ square meter /second, or increase the annual average concentration of radon-222 in air at any location outside of the disposal site by more than 0.5 picocurie / liter [40 CFR 192.02(b)].
3.
The remedial action shall ensure that radium-226 concentrations in land that is not part of the disposal site, averaged over any area of 100 square meters, do not exceed the background level by more than 5 picocuries/ gram averaged over the first 15 centimeters of soil below the surface and 15 picocuries/ gram averaged over 15-centimeter-thick layers of soil more than 15 centimeters below the land surface [40 CFR 192.12(a)].
On September 3,1985, the U.S. Tenth Circuit Court of Appeals remanded the groundwater standards [40 CFR Part 192.20(a)(2)-(3)] and stipulated that EPA promulgate new groundwater standards.
EPA proposed these standards in the form of revisions to Subparts A-C of 40 CFR Part 192, on September 24, 1987.
The proposed standards consist of two parts:
(1) a part governing the control of any future groundwater contamination that may occur from tailings piles after remedial action, and (2) a part that applies to the cleanup of contamination that occurred before the remedial action of the tailings.
In accordance with UMTRCA Section 108(a)(3), the remedial action shall comply with EPA proposed standards until such time as the final standards are promulgated. At that time, DOE has committed to re-evaluate its groundwater protection plan and undertake such action, as necessary, to ensure that the final EPA standards are met.
1.1 l
l
5 1.2 Ejte History and Procosed Action The Gunnison uranium mill site is located in Gunnison County, Colorado, just south of the city limits of Gunnison (see Figure 1.1).
This site was used to j
recover uranium between 1958 and 1962. During the life of the mill, approximately 540,000 tons of ore were processed by acid leaching at the i
Gunnison site.
The 61-acre processing site consists of the tailings pile, the mill yard, ore storage area, and windblown / waterborne areas. The uranium mill tailings on the designated site and the other contaminated materials total approximately
.i 718,900 cubic yards.
Figure 1.2 depicts the general features of the Gunnison processing site prior to initiation of remedial action, j
The remedial action for Gunnison, proposed by DOE under the Uranium Mill Tailings Remedial Action (UMTRA) Project, consists of the following major
]
activities:
l 1.
Elimination of existing safety hazards (interim action started September i
1991), including removal of asbestos and other hazardous materials from the buildings,' demolition of the mill buildings and associated structures, and excavation cf two underground storage tanks. The i
materials will be stored on-site until site remediation.
2.
Transport of all contaminated materials (uranium mill tailings pile and subpile, mill yard and ore storage area, windblown and waterborne l
contaminants, vicinity property material, and demolition and other debris) to the disposal site, 7 miles east of the processing site (see Figure 1.3).
l 3.
Stabilization of contaminated material in an approximately 29-acre tailings disposal embankment, partially below grade, with a maximum height of 50 feet, side slopes of 33 percent, and a top slope of 2.5 percent. A perimeter dike, 18 feet wide at the top, will be constructed i
with material excavated for the disposal cell.
4.
Coverage of the embankment with 1.5 feet of compacted sandy clay soil amended with approximately 5 percent by weight bentonite as a radon / infiltration barrier, overlain by a 6-inch-thick sand / gravel capillary break layer, a 73-inch frost protection layer, a 6-inch-thick sand / gravel bedding layer and a 6-inch-thick rock erosion protection layer. This multilayer cover is proposed to ensure long-term stability, reduce radon emissions, and protect ground and surface water.
l After completion of remedial action, the processing site will be restored with uncontaminated fill, and revegetated or mulched for erosion protection.
Following completion of remedial action, the processing site will eventually 3
be released for use consistent with existing land use controls.
i l
1.2 9-e,
,..i, n
,.:.~~..,~,.
e
,%w
- sw
d*
e 4
en A
Tw en Bwas pq n,,,
_ ^ ^ ^e%
.I I
l N
.A-n\\
j.
s l
h.%^^
s/h
[J
/
1
!e A^
(
N
/ ^ % u n.
^b'
/ Q^^^n,
A
^
$) 'e h
M otsposAL r } fr g
,\\
RSm ei;;;p, PROCESSING SITE m,. m
~
i mome 6 "g*
)
u s. %
Gwrumon Nesons Fwear l
simme age==or
[ Nag,g ; {
5= =
-- e m a=== y
--- Netwe Faum sounery l
o on l
w o
s
,o is m
= c
1 I
l
)
DENVERe COLORADO Figure 1.1 - Gunnison Site Location Map (00E,1992a) 1.3
s f
~~
- E*"
\\
./
/
M*
g I
_\\.,
I eams g
pose f
E TAlutses 1
f a
a e
a CommanaAL 8
ca weaou o!
I tc o,
s( f "e-
"cae o I ro oo.
o.
,e q oas recanes 4g(O AAEA tue0rviosose
../ in.
u.
1
- y.
d'.
s--
ri
/
e j-
,,, m g O. CAL*"
A=='
t a=-
./
(
m-e=
eo j
y
--- -.,o,,
==== sement e ets mesunw r
/
l g
j nor to scata Figure 1.2 - Gunnison Processing Site (DOE,1992e) 1.4
?
S
\\
. -a
=
an 11 ouwwison'\\.) d c
- =-
3,,
ffl%<
f 73' y%,,is
" *i"'"*
s Q~
rLs g=;gn J
A\\
[
k>$^ f
/
t
' ' j,
\\\\
\\
EEY=
,"4 m
b, u ' ) \\
. \\t
^
1 N
T
/
\\x
!\\
/
f o"* /
^
)
h
}
\\gp a
" g '"
-,; g -,
Figure 1.3 - Locations of the Gunnison Processing, Disposal, and Borrow Sites (DOE,1992e) 1.5
1.3 Review Process The NRC staff review was performed in accordance with the Final Standard Review Plan for the Review of Remedial Action of Inactive Mill Tailings Sites under Title I of the Uranium Mill Tailings Radiation Control Act, Revision 1 (NRC, 1993). The review consisted of comprehensive assessments of DOE's site l
design and remedial action plan, and is discussed in further detail in Sections 2 - 6 of this TER.
l The remedial action information assessed by NRC staff during this review was provided primarily in the following documents:
l Remedial Action Plan and Site Design for Stabilization of the Inactive Uranium Mill Tailings Site at Gunnison, Colorado, Remedial Action Selection Report, Final (with Attachments 1 - 5),
October 1992.
Uranium Mill Tailings Remedial Action Project (UMTRAP), Gunnison, Colorado, Calculations, Final Design for Review, Volumes I - VI, August i
1992.
Uranium Hill Tailings Remedial Action Project (UMTRAP), Gunnison, Colorado, Information for Bidders, Volumes I - VII, February 1992.
Uranium Mill Tailings Remedial Action Project (UMTRAP), Gunnison, Colorado, Information for Reviewers, September 1992.
Remedial Action Inspection Plan, Gunnison, Colorado, Review D,1993.
Pages changes transmitted by letters dated August 4, 1993 and September 10, 1993.
l 1.4 TER Oraanization The purpose of this TER is to document the NRC staff review of DOE's final RAP for the Gunnison site.
The sections of this report have been organized by technical discipline relative to the EPA standards in 40 CFR Part 192, Subparts A-C.
Sections 2, 3, and 4 provide the technical basis for NRC staff's conclusions with respect to the long-term stability standard in 5 192.02(a).
Section 5, Water Resources Protection, sumarizes the NRC staff's conclusions with regards to the adequacy of DOE's compliance demonstration with EPA's proposed groundwater protection requirements in Part 192.
Section 6, Radon Attenuation and Site Cleanup, provides the basis for the staff conclusions with respect to the radon control standards in 5 192.02(b) and soil cleanup in 5 192.12(a).
1.5 Summary of Open Issues On October 4, 1991, DOE submitted the preliminary final RAP and associated documents for NRC review and concurrence. NRC staff's review of the i
l preliminary final RAP and site design identified a number of open issues.
l These issues have been resolved satisfactorily (see Appendix A) in the Final l
l 1.6 L
RAP (DOE,1992a-d) and the revised Remedial Action Inspection Plan (MK-F,1993),- with the exception of groundwater cleanup at the processing site, which has been deferred until a later phase of the UMTRA Project.
1.6 References DOE (U.S. Department of Energy), Washington, D.C.,
" Remedial Action Plan and Site Design for Stabilization of the Inactive Uranium Mill Tailings at Gunnison, Colorado, Remedial Action Selection Report, Final," and Attachments 1 - 5, October 1992a.
" Uranium Mill Tailings Remedial Action Project (UMTRAP), Gunnison, Colorado - Information for Reviewers," 1992b.
" Uranium Mill Tailings Remedial Action Project (UMTRAP), Gunnison, Colorado - Information for Bidders," Volumes I - VII, 1992c.
" Uranium Mill Tailings Remedial Action Project (UMTRAP), Gunnison, Colorado - Design Calculations," Volumes I - VI,1992d.
--, " Environmental Assessment of Remedial Action at the Gunnison Uranium Mill Tailings Site Near Gunnison, Colorado, Final," February 1992e.
MK-Ferguson, Remedial Action Inspection Plan, Review D, Gunnison, Colorado, August 1993.
NRC (U.S. Nuclear Regulatory Commission), Washington, D.C., " Final Standard Review Plan for the Review of Remedial Action of Inactive Mill Tailings Sites under Title I of the Uranium Mill Tailings Radiation Control Act, Revision 1,"
Division of low-level Waste Management and Decommissioning, June 1993.
1.7
i l
2.0 GE0 LOGIC STABILITY 2.1 Introduction This section of the TER documents the staff's review of regional geologic information for the proposed remedial action at the Gunnison uranium mill tailings (processing) site and the disposal site.
The EPA standards (40 CFR Part 192) do not include generic or site-specific requirements for characterization of geologic conditions at UMTRA Project sites. Rather, 40 CFR 192.02(a) requires that control mechanisms be designed to be effective for up to 1000 years, to the extent achievable, and-in any case for at least 200 years.
NRC staff has interpreted this standard to mean that certain geologic conditions must be met in order to have reasonable assurance that the long-term performance objectives will be achieved. Guidance with regard to these conditions is specified in NRC's Standard Review Plan (SRP) (NRC,1993).
It should be noted, that since DOE has proposed to postpone cleanup of the groundwater at the Gunnison processing site to a separate phase of the UMTPA Project, the emphasis in this section is on the regional and site-specific geologic information relative to the disposal of the uranium mill tailings at the disposal site. At the time that DOE provides the information for cleanup of the Gunnison processing site, additional site-specific, and possibly regional geologic information should be provided.
Background geologic information for this TER is derived from DOE's Remedial Action Plan (DOE, 1992), supplemental information provided during the review process, staff site visits, and independent sources as cited.
2.2 Location The Gunnison processing site is located south of the city limits of Gunnison in Gunnison County, Colorado adjacent to the Gunnison County Airport. The site covers 61 acres and lies in the valley of the Gunnison River and Tomichi l
Creek. The 29-acre disposal site is located 7 miles east of the processing site and 2000 feet south of the Gunnison County landfill. The processing site and the disposal site will be connected by a dedicated haul road known as the Tenderfoot Mountain haul road (See Figure 1.3).
l 2.3 Geolooy DOE characterized regional and site-specific geology by referring to published and unpublished geologic literature and maps; reviewing subsurface geologic data, including logs of exploratory boreholes drilled on the site; and conducting field investigations. A summary of DOE's geologic characterization is presented below.
The Gunnison processing site and the disposal site are both located in south-central Colorado in the Southern Rocky Mountain Physiographic Province.
This province extends from the Laramie Range in Wyoming to the Sangre de Cristo Range in New Mexico. The principal mountain ranges in the site region are t*e Sawatch Range to the east, the Elk and West Elk Mountains to the north, and the San Juan Mountains to the south. The Southern Rocky Mountain Province contains the highest peaks of the Rocky Mountains, with many peaks extend &q 2.1
above 4300 meters (m) (14,000 ft). The region has a total relief of over 2400 m (8000 ft), with elevations of up to 4400 m (14,431 ft) at the top of Mt. Elbert in the Sawatch Range.
The major drainages in the Southern Rocky Mountain Province are the Colorado and San Juan Rivers to the west, the Arkansas and South Platte Rivers to the east, and the Rio Grande and the North Platte Rivers to the south and north, respectively. The Gunnison processing site lies between and just northeast of the confluence of the Gunnison River and Tomichi Creek. The disposal site lies 7 miles upstream from the Gunnison processing site along Tomichi Creek, which is the principal drainage at the disposal site.
The present land surface of the Southern Rocky Mountains has developed primarily since the widespread uplift of the Southern Rocky Mountain Province during the Laramide Orogeny.
Since the Laramide Orogeny, the mountains underwent at leastctwo intervals of erosion separated by a period of wide-spread Oligocene volcanic activity that was responsible for the San Juan Mountains. After the last erosional episode, the area was influenced by the broad gentle uplift of the crust that elevated the higher points to several thousand feet. At least six episodes of Pleistocene glaciation were
'i responsible for the formation of the present landforms.
Vegetation in the region of these sites is representative of semi-arid grasslands with sagebrush dominating poorly drained areas.
Precipitation averages about 11 inches per year, with most occurring from snowfall.
Frost penetration may occasionally reach 2.3 m (7.5 ft) in depth, but in snow-covered areas rarely exceeds 1.2 m (4 ft).
The disposal site is located on a south-facing pediment slope at the foot of the minor ridge where the Gunnison County landfill is located. The site-specific geology is shown in Figure 2.1.
This slope ends in a saddle that divides the highlands to the south from the landfill ridge to the north. The base level for drainage in the site area is Tomichi Creek. This creek is a small, widely meandering stream with a broad, vegetated floodplain. The site area has two additional tributaries, Chance Gulch to the west and Long Gulch to the east. The bedrock in Tomichi Creek drainage is at a depth of approximately 100 feet. The eruption of the West Elk Mountains and the San Juan Mountains volcanic centers in the Oligocene, along with contemporaneous fluvial deposits, has filled the Tomichi Creek channel with as much as 100 feet of fluvial deposits.
2.3.1 Stratigraphic Setting DOE characterized the regional and site stratigraphy by referring to published and unpublished geologic literature and maps; reviewing site-specific subsurface geologic data, including logs of exploratory boreholes drilled on the site; and conducting field investigations. A summary of DOE's characterization of the stratigraphy is provided below.
2.2
l
~
L.
JiR.3M J j#W 7
4
=
.5g w>,T'
.9 J
u-a,<
?.-
f 4
s.
m._ 9 3
Q..
n g'.-
~
,c
, -\\.:.
,;.s p/
.m
,s.
s_
.N
',' :'y,,, (**'
- 4. '
3.
u r.,
.v-1 i
l.j, Th;.,,h.]-
h '*
- ;u
.a
,s 7
~ ~
k.L..
~
[QQV3QWQ4 ?
.m,.
.t.A#. k....; op e.
s l
~
c,.
4 4
s s
.a..-===.
...u w
a
.~
=..
w
,gy,,
":'ait:"
,. T J
I
- .,,.A:,,,
/
~~
% - ~h-=D.;d??'
,;cf,~ ~g i;,..=
- )-
~~'
- .r..w.~
-2 4 wma:
.4
-- -..m. e.%e v$,gy:qrv e.w mre: m.g4
"~*
tgn
~-
p
-
s.
~=
-.~ o- -.~ ~
~. a
- u...,
e a~~ --
o-.
s Figure 2.1 - Site-Specific Geology of the Gunnison Disposal Site (D0E,1992) 2.3
____J
i Stratigraphy in the site region varies with the structural provinces. The Elk Mountain Province includes a complete stratigraphic sequence of Paleozoic, Mesozoic, and Early Tertiary deposits with an overlying Oligocene volcanic deposit. The oldest sedimentary rocks exposed in this region are a complete sequence of Paleozoic rocks from Cambrian to Permian Periods. This older i
sequence wedges out southward toward the Gunnison Uplift and is completely missing south of Tomichi Creek. The Junction Creek Sandstone is the oldest sedimentary rock south of Tomichi Creek.
The site-specific stratigraphy at the disposal site consists mainly of Precambrian basement complex overlain by interbedded Tertiary gravels and volcanics, although boreholes have encountered remnants of the Jurassic Junction Creek Sandstone and the Morrison Formation at several places beneath the site. The site stratigraphy is shown in Figure 2.2.
The Precambrian bedrock, consisting of metamorphic rocks with numerous granite intrusions, is exposed in the nearby drainages and in the highlands south of the site.
The Tertiary bedrock that underlies the site was deposited directly upon the Post-Laramide erosional surface and consists of interbedded volcaniclastic and fluvial deposits of Oligocene Age. The stratigraphy, as shown on Figure 2.2, reflects the interaction between the eruptive episodes of the volcanic centers of the West Elk Mountains and/or San Juan Mountains with the active stream environment of the Tertiary valley. The deposits are poorly indurated except for the occasional stratum of welded tuff. Figure 2.2 shows the thin layers of flat-lying welded tuff within the sequence of interbedded clastics. This is considered as bedrock in this case, because the materials are over-consolidated and are not-to be confused with underconsolidated alluvial deposits.
The fluvial deposits consist of rounded particles cf bouldery gravels, sands,'
and clays in a matrix of altered ash tuff. The lower zone of the fluvial deposits (below 2330 m) appears as low-energy clayey deposits indicative of ponding, while the upper deposits show cycles of deposition and subsequent reworking of fluvial sediments indicative of an active stream environment.
j The st;ata within these gravel deposits are expected to be discontinuous, with abrupt changes in grain size due to such conditions as erosional dissection or sudden upstream influxes of volcaniclastics from lateral tributaries.
j The volcaniclastics also include a lahar deposit of basaltic rock breccia with an ash matrix altered to clay minera,1s. This unit is at least 79 meters thick on the south side of the disposal si.te, but pinches out to the north where the fluvial deposits predominate. The lahar mud flow deposit was likely deposited in a well-defined channel that damed the drainage, while_ the axis of the rissing gap in the lahar zone suggests the trend of the drainage.
A fairly prominent clay unit in the lahar is indicated by site geophysical logs and also by the static water level of several wells. A possible gravelly or boundary zone below this clay zone appears to be a confined local aquifer within the volcanic lahar breccia. An 8-meter-thick zone underlying the lower contact of the lahar breccia may be interpreted as a possible " leaky" zone that developed hydraulically by underflow as a consequence of the 37-meter-thick mud flow damming the broad channel during the Oligocene. The impact of i
2.4
~
D D*
sourn DISPOSAt. SITE I
I stoo.
yg N-7M IN IN P g.
73g
- h..... A. f^,'. d " -.- m :e
__., %.* * * *@ - - - -4 mz
- MMf y
7,~
sw-me-am. ;.
mm
su-h\\'\\
\\\\
A
\\
\\
h r&
r
/
L y a.es en //.
mien.
1
\\
_1 es.
y=
".3, a-- /.. m. u N
/.'
/
/...
/ P/
7.'.'.
.?; ; :::.....
?~ *
- Y 3
,s
//y,f_. a,-l 7 ^,m/
..__.m I
//
//--/
/
\\
/
sssssssa m
/am.. -:-:4 -5554?%%mibeW " j/,s' /
llN N;
u -------
,_ 4 3 _
y
- AN
' ' % ;!:!%2:!O
- ggp~.
..,.. +f=*A:b -
nev.
-~ -
""'p~
n.
hummsundmusemed a rtg y1 u
ys HOfH20esTAL SCALE tFore g fg. trevuor amervt erreest:
g-sw. ventuser vetcasac esecesa 4.m eness easmerang BrDCessessepays LtuGO3 Or mee LaneameC asum #Laupt reen. esessue gna#emy tases. ceareta. DeafLMme asse alegate te ClaTS WWTe6 M
- J-- _ ~ ~ GaSGWreP7 WefM CLav ade CLav else TUuttS Sasagt OctA3ES N88; MMN ee auctstaat tenaces Costa DOCE$
E, net: weepest eget stase evessen,3 m.
_ gggggg
.t. -,.
a
'/
Sue
- FlavastWT ase Flee furf 88WD$ttuet SC Ou e usureeS StaL Classurecatsas 1
I Figure 2.2 - Geologic Cross Section of the Gunnison Disposal Site (DOE,1992) i 2.5 i
1
l l
the site-specific hydrology on the site and the disposal cell design are both evaluated in TER Section 5.0, Water Resources Protection.
The surface material at the disposal site consists of alluvium and colluvium deposits that form a pediment of unconsolidated gravelly sand overlying the poorly indurated Tertiary gravel formation. These Quaternary deposits range from 1.5 to possibly 6 meters on the south-facing slope. The surficial soils differ little from the gravelly formation from which they are derived and cannot be clearly distinguished in the geophysical logs. The large clasts within the fluvial deposits, which include numerous quartzite and granitic cobbles and boulders to 60 cm in diameter, form an armored erosion-resistant surface that protects the even steeper slopes from gullying or migrating nickpoints.
NRC staff has reviewed the details of the regional and site stratigraphy as provided in the RAP by DOE and concludes that the characterization of the Gunnison site adequately establishes the regional and site stratigraphy sufficient to support DOE's assessment of geologic stability.
2.3.2 Structural Setting DOE charccterized the region's structural setting by referring to published regional geologic maps, aerial reconnaissance, field observations, and mapping of features critical to assuring the long-term stability of the remedial action. A summary of DOE's structural characterization is presented below.
As a whole, the Southern Rocky Mountains are characterized by north-south trending, Precambrian cored, anticlinal uplifts, with. Paleozoic and younger formations turned up steeply against the flanks. The mountain ranges are comonly bounded by major north-south faults and separated by narrow inter-mountain basins. Although generally considered Laramide in origin, some of the main mountain ranges are wholly or in part the result of Paleozoic uplifts that were just rejuvenated during the Laramide Orogeny.
Similarly, many of the major faults were rejuvenated during the Laramide and at later times.
Igneous activity, rifting, and block-faulting occurred during the Oligocene and Miocene producing the present structural features.
l The Gunnison processing and disposal sites lie mostly within the Western Mountain seismotectonic Province, but includes a small portion of the Colorado i
Plateau to the east and abuts the Rio Grand rift seismotectonic Provinces to the west. As shown on Figure 2.3, the sites lie at the boundary between the West Elk Mountain, Elk Mountain, and Gunnison Uplift structural Provinces within the Western Mountain seismotectonic Province. The Elk Mountains consist of steeply folded and overturned Paleozoic and Mesozoic sediments that are intruded by several mid-Cenozoic stocks and overridden by the Elk Range thrust carrying late Paleozoic formations. The West Elk Mountains are located between the Elk Mountains and Gunnison Uplift. The West Elk Breccia was formed when numerous fissures and composite volcanoes vented in the area.
The West Elk volcanic field covers approximately 1600 square km (620 square miles). The Gunnison Uplift is an east-dipping monoclinal fold of Laraaide Age.
It is located south of the West Elk volcano field and west of the disposal site. The Gunnison River cuts the Gunnison Uplift, forming the Blact 2.6
~
s 109 108 107 106 I
i I
I L RAMlf SASIN 04 C/y UINTA - Fg g N
YA MPA' 4g['
4~
PLATEAU s
Sq f
UPLIPT I
Y g WHITE RIVER 4
g$, \\
y,
/%%s a
If*j%
sg
,s
\\
s s ge s
PICEANcE 4
E AI,'"
d 9
Sy'N \\ '} $
\\*E0I <
4, gh
~
ELK SITE g
' #+e b
k I
AN JUAN VOLCANICS g
a f
J fi il 4\\h1 s.% +9
\\
lu
~
~
%} \\
\\
NEW MEXICO 14 SAN JUAN SASIN g
l
)h g
I l
Mesified Ptems Isrkstaen sas Rogers. test.
I i
I LEGEMD SCALS sw estLis sei
.ieeien. Prwines.
i strwives Prwinese i
?
Figure 2.3 - Structural and Seismotectonic Provinces of the Gunnison Region (DOE,1992) l 2.7
)
i
. a
Canyon. The Cimarron fault forms the southern boundary of the Gunnison i
Uplift. A more detailed discussion of the seismtectonic provinces is provided in TER Section 2.3.4.
2.3.3 Geomorphic Setting DOE characterized the site geomorphology by referring to published literature and topographic maps. Site geomorphic conditions were characterized by aerial i
photographic interpretation and field observations. A summary of DOE's geomorphic characterization is provided below.
The fluvial processes of the Tomichi Creek and Gunnison River system are the most significant geomorphic processes that affect the site region. The disposal site lies on a south-facing pediment near the topographic saddle.
i between the moderately sloping highlands to the south and an isolated linear ridge to the north. The alluvial and colluvial deposits that comprise the j
soil on the pediment surface and side slopes are composed of lag gravel and j
bouldery materials derived from the underlying Tertiary gravel formation.
This desert pavement is included in, and underlatis by, approximately 4.6 cm of silty sand topsoil, as well as, a "C" horizon of a partially caliche-cemented 1
silty gravel and cobble that extends to depths ranging from 1.0 to 2.4 meters.
Surficial deposits directly overlie the Tertiary gravel formation at the site to depths ranging from one to three meters or more.
The bouldery gravel lag deposits form a natural erosion protection armament on the pediment surface and on all tributary slopes that dissect the pediment.
There are no nickpoints or headcutting erosion observed in any of these tributaries, which have slopes ranging from 0.050 to 0.11 cm/cm. There are no well-defined fluvial channels on the pediment surface, indicating that surface
(
sheet flow is the dominant geomorphic process affecting the site topography l
and the principal agent of sediment movement. The site is below elevations affected by glacial processes, but periglacial processes may have had a significant influence on late Ple~istocene topography and soil development.
The site surface appears to be undergoing net erosion at a slow rate as a result of the overall removal of surficial material almost uniformly across the land surface. Areas of increased soil erosion are present in small rills and gullies at the head of drainage channels along the east and west sides of the site. The site lies on a drainage divide between west and east drainage basins.
It is drained on the west side by an unnamed creek, and DOE i
has designated it as Creek W with four minor tributaries. The site pediment l
is drained on the east side by Creek E with three tributaries.
The small areas of soil erosion are restricted to steep gully sideslopes, which are generally devoid of stabilizing vegetation. Gravel and cobbles eroded out of the sides of the gullies tend to form a thin, discontinuous armor layer on the gully walls and floors that appears to slow the subsequent erosion rate.
Stabilization of the site surface and the existing gullies by sagebrush and thin grasses is also a major controlling factor on the rate and amount of soil i
erosion at the site. A negligible amount of overall surface erosion appears j
to be occurring at the site. Sheetwash and soil creep processes appear to be more important agents of downslope movement than gully erosion and extension.
t 2.8
t Principal potential geologic and geomorphic hazards at the proposed tailings disposal site consist of fluvial erosion and adverse soil conditions.
Slope movement processes are confined to relatively slow downslope creep of the upper soil mantle.
No evidence is present of deep-seated slope failures that would affect site stability.
Fluvial erosion processes include localized rill and gully erosion and surface sheetwash. Adverse soil conditions appear to be limited to some areas with moderate shrink-swell potential and a generally moderate to high erosion potential (Hunter and Spears, 1975).
Erosion potential appears limited to site areas with sparse to no vegetation cover or desert pavement, and areas where flow concentration will be increased as a result of the cell placement.
Fluvial processes of most concern to the long-term stability of the site area are sheetwash erosion, gully development, and lateral shifting of the j
l channel of Creek W.
Fluvial processes in Long Gulch do not directly affect l
the site. Due to the upland drainage divide position of the site, flooding from offsite streams poses no hazard to the site stability. The site is separated from Tomichi Creek by the ridge to the north, and therefore is not directly affected by fluvial processes in this channel and its floodplain.
NRC staff has reviewed the details of the regional and site geomorphology as l
provided in the RAP by DOE and concludes that the characterization of the Gunnision site adequately establishes the regional and site geomorphology sufficiently to support DOE's assessment of geologic stability. The effect of 4
the geomorphic processes on the erosion protection design is discussed in detail in TER Section 4.0.
2.3.4 Seismicity DOE characterized the regional seismicity by obtaining earthquake data bases provided by the National Oceanographic and Atmospheric Administration (NOAA),
by applying accepted techniques to determine earthquake magnitudes, and by I
employing methods for calculating peak horizontal ground acceleration generated by a design basis event.
A summary of DOE's seismic characterization is provided below.
i l
The historical earthquake record for Colorado and the surrounding states covers only approximately 120-130 years. The instrumental record is considerably shorter, generally dating to the early 1960's. This period is too short for use in developing a reliable database for the analysis of future seismic risk.
In order to predict the seismic potential of the site region, studies of the geologic and seismotectonic setting, recent geologic history, and evidence of recent fault movement, as well as regional seismicity, are needed.
A compilation of all earthquake epicenters within a 200 km radius of the site was obtained and DOE noted a total of 18 events recorded within a 65 km radius of the site.
Of these 18 events, only one was instrumentally located and 14 of the other events were recorded in 1921 during a swarm of earthquakes at a single epicentral location.
2.9
9 The largest earthquakes within the site region were several MMI IV events.
Closest to the site was a 1882 MMI IV event located 16 km northeast of the site. Another was part of the 1921 earthquake swarm located 49 km from the site.
The largest earthquake recorded within the 200 km radius of the site was a magnitude 5.5 earthquake located 72 km southwest of the site.
' Recurring seismicity is located primarily to the west along the boundary of the Rio Grande Rift seismotectonic Province and to the east along the west l
boundary of the Colorado Plateau Province on the Uncompahgre Uplift structure.
Three events have been recorded from 1960 to 1966 at the Colorado Plateau boundary, including the magnitude 5.5 earthquake discussed above.
The possibility of reservoir-induced earthquakes was also examined because of the presence of Blue Mesa Reservoir,16 to 56 km from the site.
DOE determined that due to the storage capacity (940,800 acre feet) and the maximum depth at the dam of 102 meters, that the reservoir was below the range that could induce earthquakes.
There is little probability that induced i
seismicity would occur or would significantly impact the site.
2.3.4.1 Seismotectonic Provinces I
The seismotectonic provinces that are significant to a seismic hazard evaluation of the Gunnison area are the Rio Grande Rift Province, the Columbia Plateau Province, and the Western Mountain Province.
l A.
Rio Grande Rift Province The Rio Grande Rift Province, located approximately 65 km west of the site, is a north-south trending extensional graben feature of great length and tectonic significance.
It extends from Chihuahua, Mexico, through west Texas, New Mexico, and most of central Colorado, almost to the Wyoming state line. The rift was initiated in Neogene time and has experienced continued activity through the Quaternary.
It is characterized by fault scarps in young alluvium; abrupt mountain fronts that exhibit faceted spurs; deep, narrow, linear valleys; Neogene basin-fill sedimentary rocks; and a biomodal suite of mafic and silicic igneous rocks.
A high percentage of all the potentially active faults in Colorado and New Mexico lie within this province. The rift province has been subdivided into 1
northern and southern subprovinces in Colorado on the basis of age of faulting. Well-defined evidence of repeated Late Quaternary movement is abundant on several faults in the southern subprovince, outside of the site region, whereas such evidence is obscure in the northern subprovince. Strong earthquakes within the Rio Grande Rift are generally associated with earthquake swarms, which are generally associated with active volcanoes and areas that have had volcanic activity in geologically recent times.
Kirkham and Rogers (1981) estimated a maximum credible earthquake for this province as a magnitude 6.5 to 7.5 event.
f i
2.10
B.
Colorado Plateau Province The Colorado Plateau is a major tectonic block composed of Paleozoic and Mesozoic rock that has been uplifted at a rate of approximately two mm per year since late Tertiary time.
It cover: an area of 295,000 square km (114,000 square miles) in western ColorPJo and adjoining areas of New Mexico, Arizona, and Utah.
~
j Except for the Uncompahgre Uplift, the Colorado Plateau appears to be fairly stable tectonically. A series of faults associated with collapsed salt anticlines and evaporite flowage in Paradox and Big Gypsum Valleys shows considerable Neogene movement.and some later Quaternary and Holocene activity; however, the faults' nontectonic origin and movement due to plastic deformation indicate a low potential for even moderate-sized earthquakes. The Uncompahgre Uplift, however, is a major tectonic feature that has been recurrently active at least since the late Paleozoic.
Some faults that flank the Uplift show evidence of Quaternary activity, and that implies considerable Quaternary uplift of the entire structure.
The highest earthquake magnitude recorded within the Colorado Plateau is estimated to be 5.5 to 5.75 (Dubois et al., 1982). This event occurred on July 21, 1959, near Fredonia, Arizona.
Kirkham and Rogers (1981) estimated l
the maximum event for the Colorado Plateau to be a magnitude of 5.5 to 6.5.
C.
Western Mountain Province i
The Western Mountain Province comprises the mountainous areas to the west of the Rio Grande Rift extending as far as the border of the Colorado Plateau.
i This province includes the San Juan, Elk, and West Elk Mountains, the west flank of the Sawatch Range, and the White River and Gunnison Uplifts.
Relatively few Neogene faults (Late Tertiary) are known in this province.
Neogene rocks in the San Juan Mountains are offset by faults related to caldera collapse; however, these are not considered to be active faults.
Several faults associated with evaporite flowage or solution also offset Neogene rocks.
These faults are generally not considered capable of generating earthquakes having magnitudes larger than about 4 or 5.
Minor evidence of Neogene reactivation of west-to-northwest trending Precambrian faults has been identified, but none are major tectonic faults that have experienced any known significant Quaternary activity.
The Sawatch Range is bounded on the east by the Arkansas River Valley and on the west by the Crookton fault, which has overturned Dakota Sandstone next to Precambrian rocks. A major fault in these mountains, the Elk Mountain thrust, is attributed to gravity causing the flank of the rising Sawatch Range to slide off, probably in Paleocene time. A major border fault of the Sawatch Range, the Castle Creek fault, displaces the Elk Mountain thrust fault.
The Gunnison Uplift is an east-dipping monoclinal fold of Laramide Age.
It is located south of the West Elk Volcano Field and west of the site. The Gunnison River cuts through the Gunnison Uplift, forming the famous Black Canyon of the Gunnison.
The Cimarrron and Red Rock faults are major en echelon faults that trend northwest-southeast. The Cimarron fault displacal Mesozoic strata up to 1600 m (5100 feet) and forms the southern boundary of 2.11 e
~
ua
the Gunnison Uplift. Motion has been recurrent mainly in the Precambrian, late Paleozoic, and during the Laramide Orogeny.
Its length is at least 64 km (40 miles). The Red Rocks fault has a length of approximately 32 km (20 miles) and is marked by a wide zone of shattered rock, typically forming alignments of drainages.
Seismic activity within the Western Mountain Province has mainly occurred in the San Juan Mountains region and in the vicinity of the Grand Hogback monocline between Glenwood Springs and Rifle. These features mark the boundary zone of the Colorado Plateau and Western Mountain Province.
2.3.4.2 Regional and Site-Specific Faults Within a 65-km radius of the site, five fault groups that define potential seismic source areas having faults with similar orientation or association with specific structures were identified. With the exception of Fault Group 1, DOE determined the faulting to be non-capable.
Fault Group 1, which is associated with the Southern Rio Grande Rift Province, consists of the Sawatch fault structure. The Sawatch fault is a series of narrow, high-angle, normal faults that mark the west side of a graben comprising the Southern Arkansas River Valley (Kirkham and Rogers, 1981).
This fault is located 56 km (36 miles) east-southeast of the site on the east side of the continental divide.
Relief on the west side of the Arkansas River Valley suggests that about 3000 m (9624 feet) of Neogene displacement has t
occurred on the Sawatch fault. The fault zone is marked by linear topography, weakly developed faceted spurs on the mountain front, prominent scarps in alluvial deposits, and numerous hot and cold springs. Surface scarps along the Sawatch fault provide evidence of Quaternary movement.
It is believed that a minimum of five discrete periods of faulting have occurred during the past 150,000 to 200,000 years, based on an interpretation of truncated fault zones, colluvial wedges, and soil profiles exposed in trenches. The most recent major displacement occurred during the late Pleistocene; however, a small displacement event offset a 4000-year-old colluvial unit about 0.1 m (0.32 feet). Relationships suggest a recurrence interval of 10,000 to 40,000 years for major displacement.
In addition, historic ground surface deformation has been reported on the east side of Mt. Princeton at two locations approximately 70 km (43 miles) from the site. Based on the evidence of Quaternary movement presented, the Sawatch fault is defined as capable.
Of these five fault groups, DOE indicated in RAP Attachment 2, (DOE,1992)
Table 4.1, that Fault Group 5 showed the closest proximity of faulting to the site with one normal fault at a distance of only 2.7 km (1.7 miles). There is no displacement of Tertiary deposits which indicates that this fault is non-capable.
NRC staff has reviewed the details of the regional and site seismicity as provided in the RAP by DOE and concludes that the characterization of the i
Gunnison site is adequate to establish the regional and site seismicity i
sufficient to support DOE's assessment of geologic stability.
2.12 1
4 2.3.5 Natural Resources DOE characterized the regional and site-specific natural resources by an analysis of regional and local publications, regional geologic maps, topo-graphic maps, and field observations. A summary of DOE's characterization of the natural resources is as follows.
The known resources in the Gunnison site region include uranium, coal, natural gas, oil, gold, barite, manganese, iron, silver, lead, zinc, copper, nickel, marble, and molybdenum. The majority of metallic deposits such as gold, silver, and molybdenum that occur in Colorado have been found within a belt of mid-Cenozoic intrusions known as the Colorado Mineral Belt. The belt trends northeast from Durango to approximately Boulder. The Gunnison Gold Belt lies within the Colorado Mineral Belt and is located 16 km (10 miles) south of the site. The Colorado Mineral Belt is about 48 km (30 miles) long and up to 11 km (7 miles) wide.
It consists of a Precambrian sulfide deposit that contains metavolcanics and related intrusive rocks metamorphosed to the greenschist or lower amphibolite facies. Gold and silver are mined from sulfide deposits found in the Dubois greenstone. The Gunnison Uplift (Figure 2.3), of which the Gunnison Gold Belt is a part, contains similar deposits of copper, zinc, lead, and, to a lesser degree, silver and gold. There are no known metallic deposits near the disposal site.
The nearest mines are 11 km from the site.
Uranium has been discovered in numerous locations in the area surrounding the disposal site. However, over 99 percent of the uranium produced came from the Cochetopa and Marshall Pass uranium districts located 13 km (8 miles) south and 40 km (25 miles) east of the disposal site, respectively. At the disposal site, shear zones in sedimentary rocks of Paleozoic and Mesozoic age are the most common host formations in both districts and in other occurrences in the site area.
The Morrison Formation and the Precambrian bedrock are the only potential sources of uranium at the disposal site area. An isolated occurrence of the Morrison Formation was encountered at the site during 1
drilling. Assuming the remote possibility that the formation contains an ore body, the fact that the formation is discontinuous, is at least 90 m below the surface, and is below the water table, all contribute to make it economically undesirable to mine.
Due to the small deposits in the shear zones and difficulty in mining them, few deposits are developed.
Coal deposits in western Colorado occur most commonly in the Late Cretaceous Mesaverde Group and the Dakota Formation. Thick, high ranking coals may also be a source and reservoir for natural gas. There is no potential for coal production at the Gunnison disposal site because no coal-bearing formations occur near the site. The nearest coal production is the Crested Butte coal field, approximately 48 km (30 miles) north of the site. There is also no potential for oil or natural gas production at the disposal site because the lithology lacks source or reservoir beds.
Of all the potential resources investigated in the area of the disposal site, only sand and gravel are present. An active sand and gravel operation in a river deposit is located on Temiehi Creek north of the disposal site along the site access road. Tertiary gravel deposits similar to that found below the site are also widespread throughout the Gunnison and Tomichi Creek valleys.
2.13
2.4 Geolooic Stability Geologic conditions and processes are characterized to determine the site's ability to meet standards in 40 CFR 192.02(a).
In general, site lithologic, stratigraphic, and structural conditions are considered for their suitability as a disposal foundation and their potential interaction with tailings leachate and groundwater. Geomorphic processes are considered for their potential impact upon long-term tailings stabilization and isolation.
Potential geologic hazards, including seismic shaking, liquefaction, on-site fault rupture, ground collapse, and volcanism are identified for the purpose of assuring the long-term stability of the disposal cell and success of the remedial action design.
2.4.1 Bedrock Suitability DOE's evaluation of the site region, as described in TER Section 2.3.1 and the RAP, indicated no evidence of bedrock instability, capable tectonic faulting within 60 km of the disposal site, or other structural condition affecting the stability of the site. NRC staff has reasonable assurance that DOE has adequately characterized bedrock stratigraphic and structural conditions at the site and that they should have no adverse effect on the design's ability to meet standards for long-term stability of the remedial action. TER Section 5.2 provides further discussion of the hydrogeologic characterization of the bedrock units at the disposal site.
2.4.2 Geomorphic Stability DOE has determined that the fluvial process of Tomichi Creek is the most significant geomorphic process in the site area.
Based on DOE's charact-erization of the geomorphic conditions at the disposal site discussed in TER Section 2.3.3, the site has experienced long-term geomorphic stability. The evidence indicates that stable conditions are likely to continue for the performance period of the disposal cell.
Evidence for long-term stability of the natural' slopes is indicated by the relative age of the surface, as shown by the deep, well-developed soils present on the natural land surfaces. A relative absence of fluvial channels on the surface shows that sheet flow is the dominant geomorphic process affecting the site topography.
Evidence for long-term stability of the pediment surface and its bordering slopes is indicated by:
(1) the absence of headcutting erosion in the several short tributaries with slopes ranging from 0.040 to 0.089 cm/cm, and (2) the absence of channelized flow on the pediment surface with slopes of 0.0175 to 0.030 cm/cm.
The stability of the surface is due in large part to the natural armoring by lag cobbles and boulders. Another factor that influences the slope stability is the near-surface presence of a flat-lying, welded tuff bed directly upslope from the site.
Located by test pits, the slightly cemented tuff bed appears to be just below the surface.
NRC staff, based on the information provided in the RAP, has reasonable assurance that geomorphic conditions of the site have been adequately i
characterized.
For a discussion of rock-size requirements, rock gradation, 2.14 t
quantities, durability, and other aspects of erosion protection design details, see Section 4.4 of this TER.
2.4.3 Seismotectonic Stability Studies by DOE to analyze seismic hazards included search for a design-basis fault, selection of a design earthquake, calculation of the estimated peak horizontal ground acceleration, recognition of potential on-site fault rupture, and recognition of potential earthquake-induced geologic failures at the site.
The seismic activity for the site region, as described by DOE and presented in Section 2.3.4 of this TER, identifies the significant structures and delineates the tectonic prcvinces.
This section provides analysis of the regional characteristics to determine sources that could generate ground motion that would most affect the site.
The maximum earthquakes (ME) for the adjacent seismotectonic provinces are based on recommendations by Kirkham and Rogers (1981). This study had estimated a range of 5.5 to 6.5 for the Colorado Plateau Province and a range of 6.5 to 7.5 for the Rio Grande Rift Province. The larger magnitudes are consistent with the fault lengths of the significant faults of these provinces and are selected as the MEs for these adjoining provinces.
Based on the closest approach of these province boundaries to the site, the distance attenuation relationship of Campbell (1981) indicates a resultant peak horizontal acceleration at the site of 0.09 g for the Colorado Plateau Province and 0.14 g for the Rio Grand Rift Province.
The floating earthquake (FE) for the Western Mountain Province, in which the site is located, is the largest event not associated with a specific structure and is determined on the basis of seismic history and tectonic character.
Since the largest earthquake considered possible without ground rupture is a magnitude 6.2 event, this magnitude is considered as the FE, representing the seismic potential of unknown structures in the region.
This event is considered to occur at a radial distance of 15 km (9.3 miles) from the site and would result in an acceleration of 0.21 g at the disposal site.
DOE has designated the Sawatch fault in Fault Group 1, located in the Rio Grande Province, as the closest capable fault to the site.
The maximum earthquake for this fault is estimated to be 7.4 at a distance of 56 km.
The maximum acceleration produced at the site as a result of this event on the fault is 0.14 g.
The design earthquake for the disposal site is determined as the earthquake that would produce the largest on-site acceleratton resulting from earth-quakes associated with capable faults in the site region or events associated with seismotectonic provinces.
Based on the above analysis, DOE determined that the FE for the Western Mountain seismotectonic Province would result in the largest acceleration at the site. Therefore, DOE has proposed a disposal site design acceleration of 0.21 g, resulting from the occurrence of a magnitude 6.2 earthquake 15 km from the site.
2.15
NRC staff has reviewed the data presented by DOE in the Gunnison RAP and agrees that the peak horizontal acceleration from a 6.2 magnitude earthquake at a distance of 15 km, using Campbell's (1981) 84th percentile value is 0.21 g.
Staff finds the data inputs and the cited results to be reasonable and conservative for DOE's calculation of the seismic coefficient for the site.
2.5 Cqnclusions T
Based upon review of the Gunnison Final Remedial Action Plan and Site Design for Review, the staff has reasonable assurance that regional and site geologic conditions have been characterized adequately to meet 40 CFR Part 192. When the Gunnison groundwater cleanup plan is submitted for NRC review, it should contain site-specific geologic information on the processing site.
2.6 References Campbell, K.W., "Near-Source Attenuation of Peak Horizontal Ground Acceleration," Bulletin of the Seismological Society of America, 71, 2039-2070, 1981.
DOE, (U.S. Department of Energy), Washington, D.C., " Remedial Action Plan and Site Design for Stabilization of the Inactive Uranium Mill Tailings at Gunnison, Colorade, Remedial Action Selection Report, Final," and Attachments 1 - 5, October 1992.
Dubois, S.M., et al., Arizona Earthouakes. 1976-1980, Arizona Bureau of Geology and Mineral Technology Bulletin, No. 193, 1982.
Hunter, W.R., and C.F. Spears, " Soil Survey of Gunnison Area, Colorado, Parts of Gunnison, Hinsdale, and Saguache Counties," U.S. Department of Agriculture, Soil Conservation Service, 1975.
Kirkham R.M., and W.P. Rogers, " Earthquake Potential in Coloradq: A Preliminary Evaluation," Colorado Geological Survey Bulletin Ns. 43, 1981.
NRC (U.S. Nuclear Regulatory Commission), Washington, D.C., " Final Standard Review Plan for the Review of Remedial Action of Inactive Mill Tailings Sites under Title I of the Uranium Hill Tailings Radiation' Control Act, Rev.1,"
Division of Low-Level Waste Management and Decommissioning, June 1993.
I I
2.'6
)
~
I 3.0 GE0 TECHNICAL STABILITY 3.1 Introduction This section presents the results of the NRC staff review of the geotechnical engineering aspects of the remedial action proposed at the Gunnison, Colorado, UMTRA Project site.
The remedial action consists of the removal of all contaminated materials from the processing site to the disposal cell near the Gunnison Landfill. The disposal cell will be an engineered embankment extending partially below grade. The excavation will extend to a depth of 17 feet below existing grade, and will be surrounded by a clean-fill dike 15 to 25 feet high.
Contaminated material will be placed in the excavation and will be covered by a nine-foot-thick multiple layer cover.
The geotechnical engineering aspects reviewed include:
(1) geotechnical information related to the processing site, borrow sites, and disposal site; (2) materials associated with the remedial action, including the foundation and excavation materials, tailings, and other contaminated materials; and (3) design and construction details related to the disposal site, disposal cell, and its cover.
Staff evaluation of related topics such as geology, geomorphology, and seismic characterization, are presented in Section 2.0 of this Technical Evaluation Report (TER).
Surface water and erosion control evaluations are presented in Section 4.0, and groundwater condition evaluations are presented in Section 5.0 of this report.
3.2 Site and Material Characterization 3.2.1 Site Description A.
Processing Site The processing site (Figure 1.2) is south of the city limits of Gunnison, Colorado.
The entire site is in the valley of the Gunnison River and Tomichi Creek. During the mill's operation, about 540,000 tons of ore were processed by the acid leaching process. The contaminated materials left at the processing site were shaped into a rectangular pile covering approximately 35 acres, to an average depth of 9 feet, containing about 459,000 cubic yards (cy). The mill buildings and the ore storage areas are on the south side of the site and cover about 20 acres. The natural soils at the site consist of Sand with gravel (SP) and sandy to clayey Gravel (GP-GC). Approximately three feet of subpile soil is estimated to be contaminated and will be removed as part of the remedial action at this site. The tailings pile was covered with a layer of material excavated from the adjoining gravel pit, and vegetation is well established on this cover. However, wind and rain have spread the contamination to the adjoining areas.
The total volume of contaminated materials including subpile, mill yard, ore storage, windblown, and vicinity property materials, is estimated to be 718,900 cy, with 833,000 cy for the design maximum.
B.
Disposal Site l
The proposed disposal site is approximately 2000 feet south of the Gunnison County landfill area, and 7 miles east of the processing site (Figure 1.3).
l 3.1
The disposal cell covers approximately 29 acres, located in an area of gentle slopes straddling a drainage divide. The site is bounded on the west by the Chance Gulch, and on the southeast by the West Long Gulch. The ground surface is sparsely vegetated. The disposal site surface generally consists of one to two feet of topsoil that is silty or clayey Sand with some gravel. The topsoil is underlain by a thick stratum of clayey to silty, sandy Gravel with cobbles and small boulders. The groundwater table is about 90 feet below the base of the disposal cell.
C.
Borrow Materials Sites The borrow material (silty, sandy Clay) for the radon barrier layer and the frost protection layer will be obtained from the Sixmile Lane borrow site, which is approximately 1.5 miles east of the disposal site. Sand and gravel borrow materials will be from:
(1) Schmalz Sand and Gravel pit, (2) Valco Sand and Gravel pit, and (3) Airport Runway material stored on the Valco property.
Samples for erosion protection rock materials were from:
(1) Chance Gulch, (2) Sage Hen Gulch, and (3) Flick Homestead sites.
3.2.2 Site Investigations A.
Processing Site Several subsurface investigations were conducted at the site; they included:
(1) 106 auger borings of 6.5-inch diameter by Sergeant, Hauskins and Beckwith, (2) 15 borings of 4-inch diameter by Colorado State University, and (3) 15 auger borings of 7.5-inch diameter and 28 test pits by Morrison-Knudsen Environmental Services (MKES).
Piezometers and observation wells were installed in several of the borings.
Both disturbed (Split Spoon and California Tube) and undisturbed (Shelby Tube) samples were taken from the borings.
Bulk samples were obtained from the test pits.
B.
Disposal Site Initially, 25 test pits were excavated by DOE to investigate the subsurface conditions.
Because of the presence of gravel, cobbles, and small boulders, conventional borings were not drilled during the site investigations. The test pits were logged and disturbed samples were collected. Subsequent investigations by MKES consisted of 14 test pits, 23 monitoring wells, and 25 piezometers.
Drilling for monitoring wells and piezometers was by the air rotary method.
i 1
C.
Borrow Materials Sites Numerous investigations of the borrow sites were conducted.
Initially, 40 test pits were excavated to characterize the materials at the Sixmile Lane borrow site.
In a later study, MKES opened another eight test pits by backhoe, and bulk samples were collected for laboratory testing to evaluate the characteristics of borrow material at the site.
Samples of rock, sand, and gravel were also collected for laboratory testing at the other borrow sites mentioned above.
3.2
Based on its review, the staff concludes that from a geotechnical engineering perspective, the site investigations at the processing site, disposal site, and borrow materials sites are in general conformance with the applicable provisions of Chapter 2 of the NRC Standard Review Plan (SRP) (NRC, 1993) and they are adequate to support the assessment of the materials occurring at the sites.
3.2.3 Site Stratigraphy A.
Processing Site The stratigraphy at the processing site consists of a pile of tailings with an interlayering of sands and slimes, overlying an alluvial deposit. The slime layers are encountered in distinct zones around the perimeter of the pile, except in the southwest corner where sandy materials predominate. The pile is also covered by a thin layer of soil which was placed as an interim cover to reduce scattering of contaminated materials. Materials in the tailings pile range from coarse granular to fine-grained materials of low plasticity. The coarse granular materials consist of medium to fine Sand varying from a clean gray Sand to a light brown silty Sand. The fine-grained materials are low-plasticity gray slimes, and the soil cover consists of a gray-brown sandy, silty, Clay with some gravel. An alluvial deposit comprises the subgrade upon which the tailings pile was constructed, and consists of an intermixed deposit of cobbles with sand, gravel, and a significant percentage of clay.
B.
Disposal - Site The stratigraphy at the disposal site consists of one to two feet of topsoil that is a silty or clayey Sand with some gravel. The topsoil is underlain by 40 to 75 feet of dense to medium dense clayey or silty, sandy Gravel with cobbles and small boulders (12 to 18 inches in diameter). Underlying the gravel is a lahar breccia formation, which is composed of igneous and metamorphic rocks in a tight matrix of volcanic ash that is altered into a clay. The lahar breccia formation is approximately 70 feet thick and is underlain by a silty-clayey Gravel deposit that is similar to the one occurring above the lahar formation. The piezometric surface is approximately 90 feet below the bottom of the disposal cell.
C.
Radon Barrier Borrow Site Radon barrier material is proposed to be borrowed from the Sixmile Lane borrow site. The stratigraphy at the site consists of an alluvial deposit with a surface layer of silty or clayey Sand topsoil one to two-feet-thick, underlain by a five to nine-feet-thick sandy Clay of low-plasticity, with occasional gravel and cobbles. The sandy Clay layer is underlain by a sandy Gravel with occasional cobbles, and has a fines (passing No. 200 sieve) content ranging from 30 to 80 percent. Material with higher fines content is to be used for the radon barrier layer, and material with a lower fines content will be used for the frost protection layer of the disposal cell Cover.
3.3
NRC staff has reviewed the details of the test pits and borings, as well as the overall geotechnical exploration program discussed in Section 3.2.3 above.
The staff concludes that the geotechnical investigations conducted at the processing, disposal, and borrow sites adequately establish the stratigraphy at each site, that the explorations are in general conformance with applicable provisions of Chapter 2 of the NRC SRP, and that they are adequate to support the assessment of the geotechnical stability of the stabilized tailings and contaminated material in the disposal cell.
3.2.4 Testing Program A.
Processing Site The laboratory testing of the samples from the processing site included gradation, Atterberg limits, specific gravity, moisture-density determina-tions, compaction, consolidation, saturated hydraulic conductivity, capillary moisture, direct shear, and triaxial shear. The design parameters were derived from the test data.
B.
Disposal Site The gravel, cobbly nature of the subsoil at the disposal site precluded laboratory testing. The testing program only included performing field moisture content and density tests using both sand-cone and nuclear methods.
C.
Borrow Materials Sites Testing for the radon barrier borrow material from the Sixmile Line borrow site included soil classification and material properties evaluation on unamended soil, and only limited testing on bentonite-amended soils. All of the tests were performed as per applicable ASTM test procedures or Corps of Engineers test procedures, when appropriate. Testing included gradation, Atterberg limits, specific gravity, compaction, consolidation, saturated hydraulic conductivity, capillary moisture, triaxial shear strength, and dispersive characteristics.
Rock durability tests and petrographic analyses were performed on the Flick Homestead, Chance Gulch, and Sage Hen Gulch borrow site materials and Schmalz Gravel samples. The durability tests performed included Los Angeles Abrasion, sulfate soundness, adsorption, specific gravity, Schmidt hammer, and Brazilian disk tension.
Optimum moisture content and Standard Proctor maximum dry density were determined by compaction tests on contaminated materials, borrow materials, and proposed embankment fill materials. Based on data from these tests, samples of the contaminated materials that will be placed in the disposal cell were compacted to a minimum of 90 percent of the maximum dry density at a moisture content near the optimum value in preparing the samples for permeability, consolidation, and triaxial testing. Corresponding tests for the radon barrier material were performed on both unamended soil and bentonite-amended soil compacted to 95 percent of the maximum dry density and it wet-of-optimum moisture content.
Both Unconsolidated Undrained (UU) and 3.4
?
Consolidated Undrained (CU) Triaxial tests were performed to determine the strength parameters for the clayey soils.
Pin ho%, crumb, and double hydrometer tests were conducted on radon barrier inaterial to assess the
]
erosion susceptibility of the material.
Specific test data were used to assess the acceptability of rock proposed as borrow material for erosion protection, both as the bedding layer and the outer layer of riprap. An evaluation of the rock quality against the requirements of erosion protection is addressed in Section 4.0 of this report.
J Based on the review, NRC staff finds that the type of tests conducted in the testing program were appropriate for the support of the engineering analyses performed and that the scope of the testing program and the utilization of the test results to define the material properties are in general agreement with the applicable provisions of the SRP.
Detailed comments on the adequacy of the number of tests to determine parameters and on the parameter values are addressed in the appropriate geotechnical engineering evaluation sections which follow.
3.3 Geotechnical Enoineerina Evaluation 3.3.1 Slope Stability NRC staff has reviewed the laboratory test results of materials to be utilized in the disposal cell as well as the subsurface exploration data for the disposal site. Additionally, the development of the design data has been reviewed and evaluated for compliance with accepted analysis and design procedures, and the results of the design have been reviewed and evaluated against accepted design values.
The analyzed cross-sections reflected the proposed side slopes and geometry of the cell. The slopes analyzed were:
(1) top slope of the disposal cell (2.5 percent slope), (2) side slopes of the embankment with dike, 3 Horizontal (H):1 Vertical (V), and (3) interior slopes of the dike (2H:lV). The slope of the disposal cell and its foundation were modeled, each being defined by engineering properties established during the laboratory testing program and exploratory program. The physical and strength parameters of the various materials were either established by laboratory tasting or assigned on the basis of data in the field exploration logs, puulished values in geotechnical literature, or engineering judgement.
Although some of the strength parameters are based on minimal testing, the values used in the design are conservative and therefore, are acceptable.
The geotechnical engineering parameters used in the analysis are conservative from the perspective of slope stability evaluation.
The Simplified Bishop /Janbu method of stability evaluation was performed using the PCSTABL computer code. DOE has used this code in the past on other UMTRA Projects and NRC staff has accepted its use. The stability during a design basis seismic event was evaluated by the pseudo-static method of analysis.
Seismic coefficients of 0.14 for short-term stability s 9.18 for long-term stability evaluations were used.
{
The stability of the top slope of the embankment was evaluated by the infimte slope stability method.
The factor of safety against a sliding failure of "a 3.5
critical slope was calculated to be 1.37 for the worst case of long-term seismic loading condition. This condition being the most severe from a stability perspective, other less severe conditions were not evaluated. The evaluation was for a top slope of 2.5 percent, which is the design slope.
However, in the event there is excess contaminated material, the top slope may be steepened to a maximum of 6.5 percent.
The corresponding factor of safety would be 1.2.
The acceptable minimum factor of safety is 1.1, and therefore the staff considers the critical slope to be safe against a sliding failure.
The interior slopes of the excavation (2H:1V) were evaluated by the Bishop method.
The calculated minimum factor of safety was 1.46 for short-term static condition whereas the acceptable minimum is 1.3.
For the short-term seismic loading condition, the calculated minimum factor of safety was 1.37 whereas the acceptable minimum is 1.0.
The staff has reviewed the geometry of the cross-section, parameters, and results of the analysis, and conclude that the excavation slopes will be stable in the short-term during the construction period.
The exterior slopes of the disposal cell (3H:1V) were also evaluated by the Simplified Bishop /Janbu method.
Several sections of the exterior slopes of the embankment were analyzed for both short-term and long-term stability.
1 Although DOE has presented calculations for short-term static and seismic l
condition and long-term static and seismic condition, the long-term seismic l
condition is the most severe condition and results of that analysis are discussed. The calculated minimum factor of safety for the long-term seismic condition was 1.23, whereas the acceptable minimum is 1.1.
The calculated factors of safety for the short-term static and seismic conditions and long-term static condition are higher than the corresponding acceptable minimums.
The staff has reviewed the geometry of the cross-section, parameters, and results of stability analysis, and conclude that slopes will be stable during the design life of the disposal cell.
Based on review of these analyses and the results, NRC staff concludes that the slopes of the disposal cell are designed to endure the effects of the geotechnical natural forces to which they may reasonably be subjected during the design life and that the analyses have been made in a manner consistent with Chapter 2 of the NRC SRP.
3.3.2 Liquefaction The soils in the disposal cell will be compacted to a minimum of 90 percent of maximum Standard Proctor density (ASTM D698) and will be in an unsaturated condition; therefore the disposal cell is not considered susceptible to liquefaction. The groundwater table at the site is reportedly at depths of 57 to 166 feet belove the bottom of the disposal embankment. The foundation soil beneath the disposal cell is clayey Sand to clayey Gravel in a medium dense to dense condition, unsaturated, and cohesive and thus not susceptible to liquefaction.
Because of the absence of water and liquefiable soil, there is no potential for liquefaction of material within or beneath the disposal cell and the applicable provisions of Chapter 2 of the NRC SRP have been met.
3.6 i
)
3.3.3 Settlement In order to mitigate the deleterious effects of excessive differential settlement between the tailings and the perimeter dike, the design includes preloading the cell, and constructing an interior berm or wedge (5H:1V) of tailings material placed against the interior slopes of the perimeter dike (Figure 3.1).
Such a design will both accelerate primary consolidation and
" feather" the compressible tailings such that radon barrier cracking is diminished.
Sheets 9 and 10 in calculation GUN-640-05-03, and Figure 3.1, illustrate the proposed detail.
The combined effect of preloading and feathering the compressible tailings is expected to reduce differential t
settlement such that strains in the radon barrier are within tolerable limits.
Preloading should also mitigate the deleterious compressibility effects which could otherwise result from poorly-mixed slime zones within a more granular i
matrix.
The settlement of these materials as a result of long-term secondary compression is shown to be about one percent of the compressible layer thickness. Although preloading will not improve secondary compression behavior, the magnitude of secondary compression will be within acceptable limits.
3.3.4 Cover Design The cover configuration and design have been reviewed and evaluated by the NRC staff. The disposal cell cover consists of the following layers, in descending order from the top:
(1) 0.5 foot riprap; (2) 0.5 foot sand / gravel (bedding); (3) 6 feet and 1 inch frost protection; (4) 0.5 foot sandy gravel (drainage); and (5) 1.5 feet radon barrier (soil amended with approximately five percent bentonite).
It is noted that the radon attenuation model includes layers 3 through 5.
The cover design for the perimeter dike slopes will consist of a one-foot-thick layer of riprap underlain by 0.5-foot-thick bedding layer.
This cover system provides a total of 9.1 feet of cover over the contaminated materials in the embankment. The cover will have a slope of 2.5 percent on the top and 33 percent (3H:lV) on the sides of the disposal cell (see Figure 3.1).
The cover is designed to protect the disposal cell against erosion by wind and water.
Section 5.0 cf tMs TER addresws the staff evaluation o of the cover as it applies to the riprap and its bedding layer, s't this aspect
'nd the cover configuration. The cover is also designed to retard the emanation of radon gas from the tailings embankment into the atmosphere, to reduce the infiltration of precipitation into the tailings embankment, and to be functional considering the frost depth for the region. The key elements reviewed in this section are:
(1) frost penetration depth, (2) geotechnical engineering parameters of all the layers of the cover, (3) radon barrier design to retard radon emanation (in part), and (4) radon barrier design to reduce water infiltration.
DOE has determined the frost. depth using the BERGGREN. BAS computer code developed at the U.S. Army Corps of Engineers (COE, 1968), and this code has been used on other UMTRA projects. The total frost penetration depth at the 3.7
J EY-0
_; ~,,,,J-"
c.
.,v,.f-' - *
.3 U.
'.<:lT;,::,'""
m e,,%r.. e
_y
[-i t o.as.c so.v o
. ~ f fi'L'.'s,.'O '
V iia'ta 7cr' aI"'.,
attor. art. conva marzo metaiats COVER CROSS SECTION t
\\
.oe-
.oo<
l TE' M"u'rA.f.cE ~ y""'
1"'1*f"
......u.
== c arco meaniau rence:maratso.,o
,cggecm
" f=xas
/
~:
<%E
__ e
' ' 'I in
- I a'."?!" ";",'! "' ""
- ""..- w-L - e=.
CROSS SECTION mi
~
C
'\\
'S 1/, f
\\ \\ \\ (' }
\\
(\\
')
\\g (h.>'p ; \\;;
\\
f
'N i m'h(
1,
=
=5 d f
/ / r, hrr % n. b,k )JJx
~
15 l
1 1
{
q
,) / j },' j s w is
- p'-
fV
<f S.7
- 8
/ /
7 a
_w d7
/'
t Ywd 20
'O A Figure 3.1 - Gunnison Disposal Cell and Cover (DOE,1992a) 3.8
disposal site is calculated to be 91 inches. The cover design provides for the appropriate depth by the total thickness of riprap (6 inches), bedding (6 inches), frost protection layer (73 inches), and the drainage-bedding layer (6 inches) above the radon barrier.
The staff has reviewed the input data used in determining the total frost penetration depth and these values are a reasonable representation of the extreme site conditions to be expected.
Therefore, DOE's evaluation of the frost penetration depth is acceptable to the staff. However, it should be noted that DOE's calculation was based on the 200-year and not on the 1000-year frost depth.
NRC staff accepts this approach for the Gunnison site because DOE has been conservative in using parameter values based on 82 percent soil compaction for the frost protection material in the radon attenuation model. These changes represent the estimated amount of freeze / thaw damage that could occur.
The radon barrier layer material is the sandy-silty Clay from the Sixmile Lane borrow site. The specifications require that the material shall consist of soils with 95 percent of material finer than one inch and with a minimum of 50 percent passing the No. 200 sieve. Testing has indicated that the borrow soil should generally meet the requirements, and that inspection procedures will verify gradation. However, it will be necessary to perform Atterberg limit testing during construction to verify that minimum plasticity indexes will be maintained to prevent cover cracking (see Section 3.4.2 of this TER).
The Remedial Action Selection Report (DOE,1992a) states that the radon barrier material will be amended with five percent (by weight) be The designhydraulicconductivityvalueoftheradonbarrieris1x10'ptonite.
cm/sec.
Based on laboratory testing, it should be possible to construct the barrier to the required hydraulic conductivity.
The cover design has been evaluated by NRC staff for geotechnical long-term stability and the design is acceptable with the understanding that DOE will perform materials testing during construction and revise the cover design as needed. The radon attenuation ability of the cover is discussed in Section 6 and the hydraulic conductivity aspects of the cover in Section 5 of this TER.
3.4 Geotechnical Construction Criteria The NRC staff has reviewed and evaluated the geotechnical construction criteria contained in the Remedial Action Inspection Plan (MK-Ferguson, 1993) and specifications in Attachment 1 to the final RAP (DOE,1992a).
3.4.1 Contaminated Material Placement The construction strategy requires the selective placement of a wedge of tailings next to the interior slopes of the disposal cell perimeter dike and preloading for two weeks (Figure 3.1).
Provisions for excavation of contaminated materials (page 02200-15 of Specifications) indicate that in order to minimize rehandling and stockpiling, the excavation should be generally in the order of priority of placement in the disposal cell.
DOE has provided details on the preferred excavation and placement sequence of contaminated tailings and offpile/subpile materials in the construction 3.9 l
drawings that are consistent with the design assumptions of the radon barrier 1
analysis.
j 3.4.2 Radon Barrier Placement The material for both the radon barrier and frost protection (Select Fill Type B) layers is a silty-sandy Clay (CL) from the Sixmile Lane borrow area.
The radon barrier material is specified to have a minimum of 50 percent by weight passing the No. 200 sieve, and the Select Fill Type B is specified to have a minimum of 30 percent by weight passing the No. 200 sieve. DOE has investigated the borrow area by excavating test pits and conducting laboratory tests on bulk samples thus collected. A review of the test pit logs and laboratory results indicates the presence of silty-sandy Clay material with a minimum of 50 percent fines content.
Differentiation of materials at the source will be provided by an inspector. Considering the apparent abundance of satisfactory soils and the presence of an inspector to flag objectionable material for rejection, the specified gradation testing frequency of 1 test per 1000 cubic yards is considered adequate.
The radon barrier material is.a silty-sandy Clay with an average Liquid Limit (LL) of 33 and Plasticity Index (PI) of 15.
Plasticity Index is an important physical parameter of clay materials which is used to determine the maximum horizontal strain that the radon barrier layer can tolerate as a result of differential settlement without resulting in deleterious tension strains or cracks.
Since it is possible to have a soil with a minimum of 50 percent fines like the radon barrier material, and not have any plasticity (silt material), Atterberg Limits testing will be conducted. The RAIP specifies that if any PI test result falls below 10, an evaluation of acceptance or rejection will be provided, based on placement location in the cell.
Specification Item I on page 02228-8 (Attachment 1 to the RAP) identifies requirements for the maximum permissible drying of the in-place radon barrier layer. The moisture should be maintained until the next lift, including the entire thickness of the overlying layer of select fill Type A, is placed and compacted. The specification requires verifying the in-place moisture by testing. The RAIP requires the moisture content of the previously placed radon barrier layer, with the exception of the tg iwo inches, be maintained at not less than the optimum moisture content minus one percent until the succeeding layer of select fill Type A is placed or compacted. Hoisture specifications are satisfactory, and NRC staff concurs with the RAP requirements in this regard.
DOE's proposed placement methods are acceptable based on the stated testing frequencies and since a qualified technician will visually inspect the radon attenuation borrow material during excavation to assure that gradations meet the specifications.
3.5 Conclusions i
Based on the review of the design and the geotechnical engineering aspects of the proposed remedial action, as presented in the Gunnison final RAP and 3.10
1 e
supporting documents, NRC staff has reasonable assurance that the long-term stability aspects of the EPA standards will be met.
3.6 References COE (U.S. Army Corps of Engineers), Baltimore, Maryland, " Digital Solutions of Modified Berggren Equation to Calculate Depths of Freeze or Thaw in Multilayered Systems," CFREL Special Report No.122, October 1968.
DOE (U.S. Department of Energy), Washington, D.C., " Remedial Action Plan and
[
Site Design for Stabilization of the Inactive Uranium Mill Tailings at Gunnison, Colorado, Remedial Action Selection Report, Final," and Attachments l
1 - 5, 1992a.
i
--, " Uranium Mill Tailings Remedial Action Project (UMTRAP), Gunnison, Colorado - Information for Reviewers," 1992b.
--, " Uranium Mill Tailings Remedial Action Project (UMTRAP), Gunnison, Colorado - Information for Bidders," Volumes I-V,1992c.
--, " Uranium Mill Tailings Remedial Action Project (UMTRAP), Gunnison, f
Colorado - Design Calculations," Volumes I-V,1992d.
l Lee, K.L., and Shen, A., " Horizontal Movements Related to Subsidence,"
l Journal of Soil Mechanics & Foundation Division, Vol. 95, No SM-1, ASCE, January 1969.
MK-Ferguson, " Remedial Action Inspection Plan, Review D, Gunnison, Colorado,"
l 1993.
NRC (U.S. Nuclear Regulatory Commission), Washington, D.C., " Final Standard Review Plan for the Review of Remedial Action of Inactive' Mill Tailings Sites under Title I of the Uranium Mill Tailings Radiation Control Act, Revision 1,"
Division of Low-Level Waste Management and Decommissioning, June 1993.
l l
3.11 l
1 4.0 SURFACE WATER HYDROLOGY AND EROSION PROTECTION
{
4.1 Hydrolooic Descriotion and Site Conceptual Desian The processing site is located in Gunnison, Colorado, near the Gunnison County Airport. DOE proposes to transport contaminated materials from this site to the disposal site, which is located seven miles east of the processing site.
i The average elevation of the disposal site is 8,040 feet above mean sea level (MSL). The site straddles a drainage divide; it is bounded on the west by Chance Gulch, and on the southeast by West Long Gulch.
. A 17-acre drainage area exists upland of the disposal site and will contribute some overland flow toward the disposal cell. A large gully extends along the northwestern boundary of the site and drains into Chance Gulch. A small gully i
on the southeastern edge of the site drains into West Long Gulch.
In order to comply with EPA standards, which require stability of the tailings for 1,000 years to the extent reasonably achievable and, in any case, for at least 200 years, DOE proposes to stabilize the contaminated materials in an engineered embankment to protect them from flooding and erosion. The design basis events for design of the erosion protection included the Probable Maximum Precipitation (PMP) and the Probable Maximum Flood (PMF) events, both of which are considered to have low probabilities of occurrence during the r
1000-year stabilization period.
As proposed by DOE, the tailings will be consolidated into a single pile, which will be protected by a rock cover. The rock cover will have a slope of 2.5 percent on the top slopes and 33 percent on the side slopes. The embankment will be surrounded by aprons which will safely convey flood runoff away from the tailings and prevent gully intrusion into the stabilized cell.
l
?
In addition, an interceptor ditch north of the embankment will be constructed to divert flood flows from the upland drainage area away from the toe of the disposal cell.
The ditch will be constructed in two branches, and flows will be diverted to the east and west, away from the cell.
l 4.2 Floodina Determinations The computation of peak flood discharges for various design features at the i
site was performed by DOE in several steps. These steps included:
(1) selection of a design rainfall event; (2) determination of infiltration losses; (3) determination of times of concentration; and (4) determination of appropriate rainfall distributions, corresponding to the computed times of concentration.
Input parameters were derived from each of these steps and were then used to determine the peak flood discharges to be used in water surface profile modelling, and in the final determination of rock sizes for erosion protection.
?
4.2.1 Selection of Design Rainfall Event One of the most disruptive phenomena affecting long-term stability is surface water erosion. DOE has recognized that it is very important to select an appropriately conservative rainfall event on which to base the flood 4.1 i
l
i protection designs. DOE has concluded and the NRC staff concurs (NRC, 1990) l that the selection of a design flood event should not be based on the extrapolation of limited historical flood data, due to the unknown level of accuracy associated with such extrapolations. Rather, DOE utilized the PMP, which is computed by deterministic methods (rather than statistical methods),
and is based on site-specific hydrometeorological characteristics.
The PMP I
has been defined as the most severe reasonably possible rainfall event that could occur as a result of a combination of the most severe meteorological conditions occurring over a watershed. No recurrence interval is normally assigned to the PMP; however, DOE and the NRC staff have' concluded that the probability of such an event being equalled or exceeded during the 1000-year stability period is small. Therefore, the PMP is considered by the NRC staff to provide an acceptable design basis.
Prior to determining the runoff from the drainage basin, the flooding analysis requires the determination of PMP amounts for the specific site location.
Techniques for determining the PMP have been developed for'the entire United States primarily by the National Oceanographic and Atmospheric Administration (NOAA) in the form of hydrometeorological reports for specific regions. These techniques are widely used and provide straightforward procedures with minimal variability. The staff, therefore, concludes that use of these reports to derive PMP estimates is acceptable.
A PMP rainfall depth of approximately 7.7 inches in one hour was used by DOE to compute the PMF for the small drainage areas at the Gunnison disposal site.
This rainfall estimate was developed by DOE using Hydrometeorological Report
'(HMR) 49 (Department of Commerce, 1977). The staff performed an independent check of the PMP value, based on the procedures given in HMR 49.
Based on this check of the rainfall computations, the staff concludes that the PMP was acceptably derived for this site.
4.2.2 Infiltration Losses Determination of the peak runoff rate is dependent on the amount of precipitation that infiltrates into the ground during the occurrence of the rainfall.
If the ground is saturated from previous rains, very little of the rainfall will infiltrate and most of it will become surface runoff. The loss rate is highly variable, depending on the vegetation and soil characteristics of the watershed. Typically, all runoff models incorporate a variable runoff coefficient or variable runoff rates. Commonly-used models, such as the Rational Formula (USBR,1977), incorporate a runoff coefficient (C); a C value of I represents 100 percent runnff and no infiltration. Other models such as the U.S. Army Corps of Engineers Flood Hydrograph Package HEC-1 (COE) separately compute infiltration losses within a certain period of time to arrive at a runoff amount during that time period.
In computing the peak flow rate for the design of the rock riprap erosion protection at the proposed disposal site, DOE used the Rational Formula.
In this formula, the runoff coefficient was assumed by DOE to be unity; that is, DOE assumed that no infiltration would occur.
Based on a review of the computations, the staff concludes that this is a very conservative assumption and is, therefore, acceptable.
4.2
\\
l 4.2.3 Times of Concentration The time of concentration (tc) is the amount of time required for runoff to reach the outlet of a drainage basin from the most remote point in that basin.
The peak runoff for a given drainage basin is inversely proportional to the time of concentration.
If the time of concentration is computed to be small, the peak discharge will be conservatively large.
Times of concentration and/or lag times are typically computed using empirical relationships such as those developed by Federal agencies (USBR,1977). Velocity-based approaches are also used when accurate estimates are needed. Such approaches rely on estimates of actual flow velocities to determine the time of concentration of a drainage basin.
The times of concentration for the riprap design were estimated by DOE using the Kirpich Method (USBR, 1977) and the Manning's Equation (Chow, 1959), which estimate actual flow velocities.
Such velocity-based methods are considered j
by the staff to be appropriate for estimating times of concentration. Based on the precision and conservatism associated with such methods, the staff concludes that the tc's have been acceptably derived.
The staff further concludes that the procedures used for computing tc are representative of the small steep drainage areas present at the site.
For very small drainage areas with very short times of concentration, DOE utilized tc's as low as 2.7 minutes; the staff considers such tc's to be conservative.
4.2.4 Rainfall Distributions After the PHP is determined, it is necessary to determine the rainfall intensities corresponding to shorter rainfall durations and times of concentration.
A typical PHP value is derived for periods of about one hour.
If the time of concentration is less than one hour, it is necessary to t
extrapolate the data presented in the various hydrometeorological reports to shorter time periods. DOE utilized a procedure reconnended in HMR 49 and by the NRC staff (NRC, 1990). This procedure involves the determination of rainfall amounts as a percentage of the one-hour PMP, and computes rainfall amounts and intensities for very short periods of time. DOE and the NRC staff have concluded that this procedure is conservative.
4 In the determination of peak flood flows, PHP rainfall intensities were derived by DOE as follows:
Rainfall Duration Rainfall Intensity (minutes)
(inches /hr) l 2.5 50 5
42 15 23 60 7.7 s
The staff checked the rainfall intensities for the short durations associated with small drainage basins.
Based on a review of this aspect of the flooding 4.3 d
l l
P 4
\\
determination, the staff concludes that the computed peak rainfall intensities are conservative.
4.2.5 Computation of PMF 4.2.5.1 Top and Side Slopes The PMF was estimated for the top and side slopes using the Rational Formula, which provides a standard method for estimating flood discharges for small drainage areas. For a maximum top slope length of 400 feet, and an additional side slope length of about 150 feet, DOE estimated the peak flow rate to be 0.6 cubic feet per second per foot of width (cfs/ft). This estimate is based on the conservative use of a top slope of 6 and 8.6 percent, which DOE states may be necessary to accommodate additional materials. Based on staff review of the calculations, the estimate is considered.to be conservative.
4.2.5.2 Apron / Toe The PMF flow rate for the downstream apron was computed similarly to the design flow rate for the top and side slopes. As discussed above, the flow rate is considered to be conservative.
4.2.5.3 Permanent Interceptor Ditch The ditch layout is such that upland surface runoff will be diverted through separate branches of the ditch into gullies on either side of the site.
In the PMF analysis, the Rational Formula was used to compute peak flow rates at different locations in the east and west branches. Maximum flow rates of 236 cubic feet per second (cfs) and 278 cfs were estimated for the east and west branches, respectively.
Based on a check of the calculations of drainage area, time of concentration, and rainfall intensity, the staff concludes that the PMF estimates are acceptable.
4.3 Water Surface Profiles and Channel Velocities Following the determination of the peak flood discharges, it is necessary to determine the resulting water levels, velocities, and shear stresses associated with that discharge. These parameters then provide the basis for the determination of the required riprap size and layer thickness needed to assure stability during the occurrence of the design event.
4.3.1 Top and Side Slopes In determining riprap requirements for the top and side slopes, DOE utilized the Safety Factors Method (Stevens, et al.,1976) and the Stephenson Method (Stephenson, 1979), respectively. The Safety Factors Method is used for relatively flat slopes of less than 10 percent; the Stephenson Method is used for slopes greater than 10 percent. The validity of these design approaches has been verified by the NRC staff through the use of flume tests at Colorado State University.
It was determined that the selection of an appropriate design procedure depends on the magnitude of the slope (Abt, et al.,1987).
The staff therefore concludes that the procedures and design approaches used j
4.4 1
1
i by DOE are acceptable and reflect state-of-the-art methods for designing riprap erosion protection.
4.3.2 Apron / Toe The design of the 20-foot wide apron at the toe of the disposal cell is based on the following considerations:
1.
provide riprap of adequate size to be stable against the design storm (PMP),
2.
provide uniform and/or gentle grades along the apron and the adjacent ground surface such that runoff from the cell is distributed uniformly at a relatively low velocity, such that the potential for flow concentration and erosion is minimized, and 3.
provide an adequate apron thickness to prevent undercutting of the disposal cell by (a) local scour that could result from the PMP, or (b) potential gully encroachment that could occur due to gradual headcutting over a long period of time.
The key elements which DOE considered in the design of riprap protection for the apron / toe are:
1.
the lower part (approximately the last 15 feet) of the 33 percent side slope immediately upstream of the grade break, 2.
the toe, which is the relatively flat lower slope (5 percent) immediately downstream of the grade break formed when the side slope meets the apron, 3.
the downstream portion of the apron which is assumed to have collapsed due to scour or long-term erosion, and 4.
the ground surface adjacent to the apron.
DOE used several analytical methods for designing the riprap apron / toe.
Additional detailed discussion of the riprap design of various components of the apron / toe can be found in Section 4.4.1.2, below.
4.3.3 Permanent Interceptor Ditch Manning's Equation was used to estimate normal depths and velocities under the estimated discharge conditions.
Flow depths in either branch will decrease from about two feet in the main channels to about one foot at the ditch outlets during a PMF.
Flow velocities in either branch will decrease from approximately seven feet per second (fps) in the channels to approximately five fps at the outlets. The upper portion of the gullies (where the channel outlets are located) will be filled with gravel and cobbles; thus, DOE considers that the maximum outlet velocity of five fps will not cause headcutting or further erosion of the east and west gullies. The design of 4.5
erosion protection for the outlets of the ditch is discussed in Section 4.4.1.3.3, below.
The Safety Factors Method was used to determine riprap sizes for the ditch.
Based on staff review of the calculations, the analysis is acceptable.
Additional detailed information related to the design of the ditches may be found in Section 4.4, below.
4.4 Erosion Protectica 4.4.1 Sizing of Erosion Protection Riprap layers of various size and thicknesses are proposed for use at the Gunnison site. The design of each layer is dependent on its location and purpose.
4.4.1.1 Top and Side Slopes The layer of riprap on the top slope has been sized to withstand the erosive velocities resulting from an on-cell PMP, as discussed above. DOE proposes to use a 0.5-foot-thick layer of rock with a minimum D of 1.5 inches. The 50 riprap will be placed on a 0.5-foot-thick bedding layer. The Safety Factor Method was used to determine the rock size.
The rock layer on the side slopes is also designed for an occurrence of the local PMP.
DOE proposes to use a one-foot-thick layer of rock with a minimum D
of approximately 4.2 inches. The rock layer will be placed on a 3a 0.5-foot-thick bedding layer.
Stephenson's Method was used to determine the required rock size. Conservative values were used for the specific gravity of the rock, the rock angle of internal friction, and porosity.
Based on staff review of the DOE analyses, and the acceptability of using appropriate design methods, as discussed in Section 4.3, above, the staff concludes that the proposed rock sizes are adequate.
4.4.1.2 Apron / Toe DOE evaluated the design of the apron / toe in four separate segments, as discussed in Section 4.3.2, above.
Following is the staff evaluation of each of the segments.
4.4.1.2.1 Lower Side Slope For the lower 15-foot length of the side slopes, DOE proposes to use a 1.5-foot-thick layer of rock, gradually increasing in thickness to a four-foot-thick layer of rock, with a minimum D a -size of 6.2 inches.
3 Although several methods were used to estimate the rock size required for the i
toe apron, the U.S. Army Corps of Engineers method for sizing rock in stilling basins (COE,1970) was selected as the best available method for the conditions at the toe of the disposal cell.
4.6
00E determined the velocity associated with PMP flows down the side slope and assumed that turbulence would be created on the lower portion of the slope where it meets the toe. To account for this turbulence (and energy dissipation), DOE increased the velocity in accordance with Corp of Engineer recommendations.
This is accomplished by using a turbulence coefficient; DOE selected neither the most nor least conservative coefficient. The coefficient selected increased the required rock size from 2.4 to 6.2 inches (nearly triple).
Based on staff analysis of DOE's methods and assumptions, the 6.2-inch rock proposed for this portion of the slope is acceptable.
4.4.1.2.2 Toe For the actual toe area, which will have a 20-foot length and a 5 percent slope, DOE used the USACE tractive shear stress method (COE, 1970) to determine the required rock size. The shear stress produced was conservatively doubled to account for turbulence and non-uniform flow.
The rock size calculated using this method was found to be about 4 inches, which is smaller that the proposed D size of 6.2 inches. Based on our review of g
DOE's calculations, the rock size is acceptable.
4.4.1.2.3 Collapsed Slope As part of the analysis of the toe area, DOE conservatively assumed that the natural ground downstream of the toe would be eroded due to cumulative local scour and/or erosion at its base, resulting in the collapse of the rock into the eroded area.
It was assumed that the collapsed slope of the rock would be 1 vertical (V) on 3 horizontal (H).
The required rock size for flow over this slope was calculated using the Stephenson Method, as recommended by the staff.
Using this method, the required size is calculated to be 4.6 inches. Since this computed size is less than the proposed size of 6.2 inches, the rock size is acceptable.
4.4.1.2.4 Natural Ground In order to determine the depth to which the toe must be placed, it is necessary to estimate the depth of scour which will occur to the graded natural ground slope just downstream of the toe. DOE assumed that the ground slope would be about 7 percent and assumed that a flow concentration factor of 3, corresponding to gully flows, would occur. Using methods developed by the U.S. Department of Transportation (DOT, 1975), the scour depth was estimated to be about two feet. However, DOE proposes to place the toe to a depth of four feet to provide added conservatism to account for a possible increased erosion of the topsoil layer.
Staff review of the calculations indicates that the methods are appropriate and conservative.
4.4.1.3 Interceptor Ditch The Safety Factors Method was used to determine rock sizes in the main channel and outlet sections of the interceptor ditch. The minimum D rock sizes so required are 7.2 and 11.7 inches in the ditch channels and outlets, respectively.
4.7
4 The riprap design of the interceptor ditch was further analyzed by DOE in the following areas:
1 design of the ditch side slopes from runoff directly down the side l
slopes from the embankment and from the upland drainage area, 2.
design of ditch for runoff directly through the ditch, 3.
design of ditch outlet, and 4.
sediment considerations.
4.4.1.3.1 Ditch Side Slopes i
A riprap layer with a D of 7.2 inches is proposed for a substantial length 3a of the ditch. The design of the ditch side slopes considered the effects of-PMF sheet flows directly down the proposed IV on 5H embankment side slopes and from the 17-acre upland drainage area. For the embankment side of the ditch, DOE checked the proposed rock size for a flow of 0.60 cfs/ft. Using the Stephenson Method for the IV on 5H ditch side slope, the required D was found to be 2.3 inches, which is less than the 7.2 inches proposed.3a For the other side of the ditch (which receives flows from the upland drainage area) with a flow rate of 2.64 cfs/ft (corresponding to a flow concentration factor of 3), the required D was found to be about 6 inches, also less than so proposed.
Based on staff review of the analysis, the proposed rock sire of 7.2 inches is adequate.
4.4.1.3.2 Ditch (Main Section) l For flows directly through the ditch, the Safety Factors Method was used to j
determine the rock size.
Based on a review of the calculations, the proposed rock size of 7.2 inches is considered to be adequate.
4.4.1.3.3 Ditch Outlets The ditch outlets will be flared out to 40-foot widths (from 10-foot base l
widths) in a distance of 60 feet. DOE indicates that this will reduce the outlet velocities to about five fps.
Maximum potential scour depths due to the PMF flows were computed using the i
U.S. Department of Transportation (DOT,1975) formula and lacey's formula (Davis and Sorensen, 1979). The maximum potential scour depths were calculated by DOE to be about five feet, using both of these formulae. DOE proposes that the riprap at the ditch outlets will, therefore, be extended down to five feet below grade.
The outlet section of the ditch is assumed to collapse due to either gully i
headward erosion over a long period of time, or the PMF flow in the ditch.
In order to reduw the rock size at the outlet, a pre-formed outlet slope of.
t IV on 5H will De constructed. This slope requires a stable rock size of 11.7 inches, calculated using the Stephenson Method.
Based on a review of the calculations by the staff, the design is acceptable.
4.8
i 4.4.1.3.4 Sediment Considerations DOE considers that sediment from the 17-acre taland drainage area is not expected to clog the diversion ditch, for the following reasons:
1.
The upland drainage area has an average slope of only 2.4 percent, whereas the East and the West branches of the diversion ditch are designed with relatively steep slopes of 2.7 and 2.5 percent in the upper reaches adjacent to the tailings embankment.
Concentrated flows in the ditches will increase velocities well above those occurring as sheet flow on the natural ground. Therefore, any sediments transported to the ditch by sheet flows will easily be flushed out by the higher velocities in the ditch.
Even though the slope of the ditch decreases as the two branches 1
separate from the tailings embankment, these flatter slopes (ranging from 2 to 0.7 percent) will not reduce the self-cleaning velocities.
2.
The potential for gully development (and resulting high flow velocities)
[
in the upland drainage area and subsequent transport of large bed-load material into the diversion ditch is low. This is because the overland drainage pattern in the upland area is diverging, based on review of topographic maps of the area. As the flow moves towards the diversion ditch, it tends to fan out (diverge, rather than converge), thereby reducing the flow concentration and gully incision potential. The 4
contour map does not show evidence of any existing gullies, and the staff site visit did not indicate the presence of gullies in this area.
3.
The soils in the site area, consisting of about 55 percent gravels and cobbles, will limit the depth of incision by self-armoring.
Based on a review of the design, the staff concludes that sedimentation in the ditches will not have a significant effect on the capacity of the ditches to transport flood flows.
4.4.2 Rock Durability Rock durability is considered to determine if there is reasonable assurance that the rock itself will sirvive and remain effective erosion protection for 1000 years.
Rock durability is defined as the ability of a material to withstand the forces of weathering.
Factors that affect rock weathering are:
i (1) chemical reactions with water, (2) saturation time, (3) temperature of the water, (4) scour by sediments, (5) windblown scour, (6) wetting and drying, and (7) freezing and thawing.
DOE conducted investigations to identify acceptable sources of rock in the j
site vicinity. The suitability of the rock as a protective cover was then 1
assessed by laboratory tests to determine its physical characteristics.
DOE used the results of these tests to classify the rock's quality and to assess the expected long-term performance of the rock.
In accordance with past CCE rock-testing practice, the tests included:
1.
Petrographic Examination (ASTM C295).
Petrographic examination of roca is used to determine its physical and chemical properties.
The 4.9
r l 1 examination establishes if the rock contains chemically unstable ~ minerals or volumetrically unstable materials.
2.
Bulk Specific Gravity (ASTM C127). The specific gravitj of a rock is an indicator of its strength or durability; in general, the higher the specific gravity, the better the quality of the rock.
3.
Absorption (ASTM C127). A low absorption is a desirable property and indicates slow disintegration of the rock by salt action and mineral hydration.
4.
Sulfate Soundness (ASTM C88).
In locations subject to freezing or exposure to salt water, a low percentage is desirable.
5.
Schmidt Rebound Hammer. This test measures the hardness of a rock and can be used in either the field or the laboratory.
6.
Los Angeles Abrasion (ASTM C131 or:C535). This test is a measure of a rock's resistance to abrasion.
o 7.
Tensile Strength (ASTM D3967 or ISRM Method). This test is an indirect test of a rock's tensile strength.
DOE then used a step-by-step procedure for evaluating durability of the rock 3 in accordance with procedures recommended by the NRC staff (NRC, 1990), as follows:
Step 1.
Test results from representative samples are scored on a scale of 0 to 10.
Results of 8 to 10 are considered " good"; results of 5 to 8 are cor.sidered " fair"; and results of 0 to 5 are considered " poor".
Step 2.
The score is multiplied by a weighting factor. The effect of the weighting factor is to focus the scoring on those tests that are the most applicable for the particular rock type peing tested.
Step 3.
The weighted scores are totaled, divided by the maximum possible score, and multiplied by 100 to determine the rating.
Step 4.
The rock quality scores are then compared to the criteria which determines its acceptability, as defined in the NRC scoring procedures.
DOE has determined that the rock proposed for the Gunnison disposal site scored 85-99, with an average of 94. The staff concludes that this rock will be of sufficient quality to meet EPA standards.
4.4.3 Testing and Inspection of Erosion Protection The staff has reviewed and evaluated the testing and inspection quality control requirements for the erosion protection materials. Based on a review of the information provided in the RAIP, the staff concludes that the proposed testing program is acceptable.
4.10
4.5 Vostream Dam Failures There are no impoundments near the site whose failure could potentially affect the site.
4.6 Conclusions Based on its review of the information submitted by DOE, the NRC staff concludes that the site design will meet EPA requirements as stated in 40 CFR 192 with regard to flood design measures and erosion protection.
The staff concludes that an adequate hydraulic design has been provided to reasonably assure stability of the contaminated material at the Gunnison disposal site for a period of 1000 years, or in any case, at least 200 years.
4.7 References Ab:., S.R., et al., " Development of Riprap Criteria by Riprap Testing in ~
Flumes: Phase I," NUREG/CR-4651, Vol. I, 1987.
Chow, V.T., Open Channel Hydraulics, McGraw Hill Book Co., New York,1959.
COE (U.S. Army Corps of Engineers), Office of the Chief of Engineers, " Flood Hydrograph Package HEC-1," Hydrologic Engineering Center, continuously updated and revised.
--- Hydraulic Design of Flood Control Channels," EM 1110-2-1601, 1970.
Davis, C.V., and K.E. Sorensen, eds., Handbook of Aeolied Hydraulics, Third Edition, McGraw Hill Book Co., New York, 1969.
00T (U.S. Department of Transportation), " Hydraulic Design of Energy l
Dissipators for Culverts and Channels," Hydraulic Engineering Circular No. 14, December 1975.
NRC (U.S. Nuclear Regulatory Commission), " Design of Erosion Protection Covers for Stabilization of Uranium Mill Tailings Sites," Staff Technical Position, August 1990.
3 Stephenson, D.,
Rockfill in Hydraulic Enoineerina, Elsevier Scientific Publishing Co., New York, 1979.
Stevens, M. A., et al., " Safety Factors for Riprap Protection," ASCE Journal of the Hydraulics Division, Vol. 102, No. HYS, May 1976.
U.S. Bureau of Reclamation (USBR), Desian of Small Dams,1977.
U.S. Department of Commerce, National Oceanic and Atmospheric Administration.
" Probable Maximum Precipitation Estimates, Colorado River and Great Basin Drainages," Hydrometeorological Report No. 49, 1977.
4.11
-.eL
,mJ m--s--
+
n
+
+-k p
5.0 WATER RESOURCES PROTECTION 5.1 Introduction NRC staff has reviewed the final Remedial Action Plan (RAP) 1992) and auxiliary documents for the Gunnison, Colorado, UMTRA Project site for compliance with EPA's proposed ground-water protection standards (EPA, 1987).
THe review was guided by relevant portions of NRC staff's Standard Review Plan for UMTRCA Title I mill tailings remedial action plans (NRC, 1993).
The Gunnison processing site operated from February 1958 to April 1962. The
[
Gunnison milling process extracted uranium oxide from the ore by an acid-leach method using sodium chlorate and sulfuric acid. Other chemicals used in the extraction and precipitation process included di-ethylhexyl phosphoric acid (EHPA solvent), sodium carbonate, and magnesia.
DOE has concluded that the proposed remedial action complies with the proposed f
EPA standards, because the hazardous constituent concentrations at the disposal site will not exceed the relevant concentration limits in ground water at the designated Point of Compliance (P0C) for at least 1000 years.
DOE has also concluded that the disposal cell will remain unsaturated, due to engineering design features and favorable site conditions, both of which will prevent water build-up in the cell. DOE's conclusions on compliance are based on estimates of the available attenuation capacity within the native materials beneath the disposal cell. These estimates indicate that the subsoil will adequately retard the movement of hazardous constituents from the disposal cell, and meet the EPA ground-water protection standards.
NRC staff is consistent with EPA's ground-water protection standards, by distinguishing between the disposal of residual radioactive materials at the disposal site and cleanup of existing ground-water contamination at the processing site. As with all VMTRA Project sites, NRC staff cannot accept l
DOE's proposal to defer cleanup of existing ground-water contamination, until DOE demonstrates that public health and the environment will not be impacted by the delay of remediation at the Gunnison processing site.
5.2 Hydroaeoloaic Characterization The hydrogeologic characterization provides the basis for determining which strategies will be appropriate for ground-water resources protection in the i
vicinity of the processing and disposal sites. The characterization sets the groundwork for complying with the standards of 40 CFR 192, Subparts A - C.
r.ritical elements of the hydrogeologic characterization are:
(1) identi-l fication of the hydrogeologic units, (2) hydraulic and transport properties, (3) geochemical conditions and contamination extent, and (4) present and potential water use.
Each of these elements are presented below for the i
processing and dispogal sites, as provided in DOE's final RAP.
I 5.1 I
h
--e r
-n-
+
-w
5.2.1 Identification of Hydrogeologic Units A.
Processing Site The uppermost hydrogeologic unit at the Gunnison processing site is the combined alluvial deposits of the Gunnison River and Tomichi Creek. The alluvium consists of sand and river gravel with lenses of ciayey gravel.
These deposits comprise the uppermost aquifer in the sita vicinity. The alluvial deposit's thickness varies considerably in the site area. The full alluvial thickness has been encountered in only four of DOE's monitoring well s.
These four wells exhibited an alluvial thickness ranging from about 70 to about 130 feet.
DOE cites regional information that indicates that the alluvium is underlain by the Brushy Basin member of the Morrison formation. Bedrock encountered in the four boreholes near the processing site is described as shale and igneous.
The bedrock thickness, areal extent, and hydraulic properties have not been confirmed in the site vicinity. Additionally, the hydraulic relationship with deeper aquifers in the area has not been determined.
Ground water in the alluvium is encountered from 0.2 feet to 14 feet below the ground surface.
Localized silt and clay layers may create semi-confined conditions, but the aquifer is generally unconfined. Ground-water flow is generally southerly and southwesterly, and generally follows the gentle slope of surface topography.
Ground-water elevations fluctuate seasonally through the year in response to surface-water variations. Nearby irrigation ditches in pastures, may function as recharge sources during the drier summer months.
Shallow ground water is recharged by precipitation, high surface-water events, and possibly seepage from irrigation ditches. Ground wcter eventually discharges into the river and creek through base-flow contribution.
DOE has committed to performing additional hydrogeologic characterizations at the processing site, as part of a separate DOE program for addressing ground-water restoration. NRC staff defers comment on the present hydrogeologic characterization of the processing site, because DOE proposes to defer cleanup of existing contamination until EPA promulgates final cleanup standards in 40 CFR 192.
When ground-water remediation plans are developed, DOE must include adequate characterizations of the bedrock thickness and the alluvial thickness in the processing site vicinity, along with the hydraulic relationship between the alluvium and the deeper ground-water systems. This is considered as an open issue relating to D0E's deferral of ground-water cleannp.
B.
Disposal Site
]
The Guanison disposal site is located seven miles east of the processing site.
l The disposal cell will cover approximately 29 acres in a broad, relativel_v flat upland, about 2000 feet south of the Gunnison County municipal landfill.
The hydrostratigraphic units characterized at the disposal site are:
(1) upper gravels of Tertiary age that form the unsaturated unit in the area; (2) undifferentiated volcaniclastic mudflows (lahar) that form a semi-confining unit over the southern portion of the area and ~ pinch out near the 5.2 4
t i
northern edge of the disposal cell; (3) lower gravels of Tertiary age that form the uppermost aquifer; (4) undifferentiated Tertiary gravel, consisting of the Upper and Lower gravel units where the semi-confining unit is absent; and (5) bedrock, composed of minor siltstones and sandstone of Jurassic age, j
and Precambrian metamorphic rocks. The permeable strata above the bedrock system average about 295 feet in thickness.
Ground water is encountered at depths between 34 and 69 feet in the semi-confining layer, and between 57 and 166 feet in the uppermost aquifer.
Monitoring wells completed in the upper portions of the bedrock system exhibit lower water levels than those in the uppermost aquifer. This relationship indicates a downward hydraulic gradient between the Tertiary gravel and the bedrock. The disposal site is situated on a local topographic divide which diverts the shallow ground-water flow in three directions (west, east and north).
Recharge to the shallow ground water occurs from precipitation in the uplands to the south of the disposal site.
Ground water is discharged by underflow that eventually contributes to the base flow of Tomichi Creek, approximately two miles away.
NRC staff has reviewed the hydrogeologic characterization of the disposal site and agrees with the designation of the lower Tertiary gravel as the uppermost aquifer, and the designation of the volcaniclastic lahar unit as a semi-confining layer.
5.2.2 Hydraulic and Transport Properties A.
Processing Site The hydraulic and advective transport properties of the alluvial aquifer system were evaluated by DOE with a 48-hour pumping test at the processing site and water-level measurements at the site and the surrounding area. The measured horizontal hydraulic conductivities ranged from 10 to 238 ft/ day, with a geometric mean of 44 ft/ day. The measured storage coefficients ranged from 7.0 x 10 to 1.0 x 10'3 A mean horizontal hydraulic gradient of 4
4.23 x 10'3 was determined from water-table elevations measured in February 1991. The average linear horizontal velocity, based on the mean hydraulic conductivity, gradient, and an estimated effective porosity of 25 percent, was calculated to be 270 ft/yr. Variations in flow direction or gradient resulting from transient surface-water influences were not considered in these estimates. As previously mentioned, interactions between the shallow alluvial system and deeper ground-water systems were not evaluated by D0E.
NRC staff defers comment on the calculations and measurement methods used to characterize the hydraulic properties at the processing site, because DOE proposes to defer ground-water cleanup.
B.
Disposal S;te The hydraulic ar.d a~1vective transport properties of the uppermost aquifer were evaluated by DOE rith a 72-hour pumping tat, and water-level measurements at the disposal site and the surrounding area. Slug tests were also performed in eight wells that are screened in the Tertiary lower gravel.
The measured 5.3
horizontal hydraulic conductivities from the pumping test ranged from 0.39'to 0.53 ft/ day, with a geometric mean of 0.5 ft/ day. By comparison, the hydraulic conductivities from the slug tests ranged from 0.15 to 3.0 ft/ day.
The slug test results are slightly larger than the pumping test results, which probably reflect the well construction influences in the slug test rpults.
The measured storage coefficients ranged from 2.2 x 10 to 5.2 x 10' The horizontal hydraulic gradient within the uppermost aquifer is dependent on the flow direction, because the disposal area overlies a local ground-water I
divide.
The potentiometric map presented in the RAP shows the ground-water flow originating in the upland located south of the site and branching to the east and west beneath the disposal cell. Data from two monitoring wells at the Gunnison County landfill were also incorporated into the potentiometric surface map discussed in the RAP. This map shows a prominent ground-water discharge trough associated with East Long Gulch within the County landfill property.
DOE has interpreted the ground-water trough as extending to the area past the head of East Long Gulch, north of the proposed tailings disposal cell. This interpretation depicts a long, narrow barrier north of the disposal cell that effectively redirects all ground-water flow from the north, eastward along East Long Gulch. The NRC staff does not agree with this interpretation, because of the absence cf measured data in the interpreted trough area.
Consequently, the NRC staff reevaluated the potentiometric surface in this area, using the measured unsaturated thickness of the upland areas exhibited in all of the monitoring well measurements, as a way to bound the uncertainties in areas of few measurements. A conservative reinterpretation of the potentiometric contours, performed by NRC staff, provides reasonable assurance that ground-water flow from the Gunnison County municipal landfill will be directed away from the tailings disposal cell, eastward along East Long Gulch.
Gradient measurements were made along three flow paths at the site. The eastward flow gradient averages 0.010, the northward component (originating to the south of the site) averages 0.055, and the westward component averages 0.045. These gradients were determined from water-table elevations measured in May 1991.
The average linear horizontal velocity; based on the mean hydraulic conductivity, gradient, and an estimated effective porosity of 25 percent; was calculated to be 7.3 ft/yr for the eastward flow component, 32 ft/yr for the westward component, and about 40 ft/yr for the northward component. Surface-water influences in the uppermost aquifer appear to be far removed and are not a concern at the disposal site.
The lahar unit overlies the uppermost aquifer beneath most of the disposal area.
The thickness of this unit ranges from 0 to 210 feet.
Field determinations of the hydraulic conductivity of the lahar unit were unsuccessful, due to extremely slow water-level recovery during slug tests.
The lahar unit hay a saturated hydraulic conductivity of about 7.05 x 10
ft/ day (2.5 x 10' cm/sec) or less, based on laboratory measurements.
Laboratory measurements also confirm that the unit is at or near saturation.
which indicates that the lahar is a confining or semi-confining layer over the uppermost aquifer.
5.4
Above the lahar unit are unsaturated Tertiary gravel and sand deposits. The saturated hydraulic conductivity of the upper gravel, atory testing, is approximately 1.11 ft/ day (3.9 x 10',as determined by labor-cm/sec). Unsaturated flow was evaluated as a transport mechanism in a vertical, one-dimensional seepage flux analysis.
This analysis was done as part of the geochemical attenuation assessment, which is presented in TER Section 5.2.3.
One of the assumptions for performing the vertical flow analysis is that lateral water movement will be limited because of the dry nature of the unsaturated zone.
The analysis demonstrated an accumulation of unsaturated moisture above the lahar unit, within the upper gravel. However, the lateral movement of unsaturated (potentially contaminated) moisture is a potential concern, because of the relatively close location of incised gullies to the east and west of the disposal cell. Although this flow path is a potential concern, the likelihood of contaminated water reaching the gully is small due to geochemical attenuation within the upper gravels.
The NRC staff agrees with the hydrogeologic characterization of the Gunnison disposal site.
5.2.3 Geochemical Conditions and Extent of Contamination A.
Processing Site DOE's assessment indicates that ground-water contamination from the processing site is readily distinguished from ambient ground-water conditions, due to the calcium-sulfate nature of the contaminated waters. Ambient ground-water quality is consistently of the calcium-bicarbonate type.
Sulfate concen-tration contours indicate a large downgradient area (approximately 2000 feet by 5000 feet) of the alluvial aquifer has been impacted by milling operations.
Data indicate that the full thickness of the aquifer has likely been affected.
NRC staff defers comment on the geochemical assessment and the extent of ground-water contamination at the processing site, because DOE proposes to defer ground-water cleanup. However, DOE must redefine the plume characteristics at the time of aquifer remediation, because of the dynamic nature of the contaminant plume. This is part of the open issue relating to the ground-water cleanup deferral.
B.
Disposal Site DOE's assessment of the disposal site's geochemical characteristics indicates that conditions are favorable for attenuating the hazardous constituents contained in the acidic tailings porewater.
The identified attenuation mechanism is composed of two primary components:
(1) acid neutralization, and the precipitation of insoluble metal salts; and (2) physical adsorption of the unprecipitated constituents.
j Measurements made with surrogate samples of tailings pore fluid (spiked Durango site fluid) have shown that the majority of the hazardous constituents will precipitate as insoluble salts at near neutral pH. The surrogate sample j
was initially used because of an insufficient sample quantity of the Gunnison tailings fluid.
DOE later performed batch column tests with Gunnison tailings 5.5
fluids after adequate amounts were collected. The results of this testing are discussed below.
The buffering capacity of the Tertiary upper gravel and the lahar unit were measured in composite soil samples collected from drill cuttings. The buffering capacity within the Tertiary upper gravel was measured at about 15.7 tons of calcium carbonate (CACO ) equivalent per 1000 tons of soil and 3
the capacity within the labar unit is about 38.0 tons of CACO equivalent per 3
1000 tons of soil. DOE has stated that this should provide adequate buffering capacity to neutralize the acidic tailings fluids that are expected to seep from the disposal cell.
Adsorptive capability within the Tertiary upper gravels was evaluated by lithologic analysis.
Examination of the gravel shows that the unit contains significant amounts of silt-sized and clay-sized materials within the gravel matrix. A semi-quantitative analysis of the clay mineralogy showed that the clays are dominantly of the smectite group (montmorillonite and nontronite) with additional amounts of mixed layer clays. These clay mineral types exhibit a high cation exchange capacity (CEC), and are chemically reactive.
The CEC for the upper gravel (measured in the composite soil samples) is about 22 milliequivalents (meq)/100 g, while the CEC within the lahar unit was measured about 27 meq/100 g.
DOE's Calculation No. GUN-06-92-13-05-02 summarizes the batch column tests using the Gunnison tailings fluid. Constituents that are not fully attenuated through acid neutralization are primarily cadmium, cobalt, molybdenum, uranium, and zinc. Cadmium, cobalt, and zinc were removed to levels below the method detection limits by adsorption to mineral grains. About 50 percent of the uranium concentration was removed by adsorption, while molybdenum was unaffected by contact with the Tertiary gravels. One of the basic assumptions l
behind the batch test is that the measured concentrations represent equilibrium conditions. Consequently, each incremental unit of fresh, reactive surface of the matrix material will remove the same proportion of constituents from solution as did the previous unit. With 50 percent of the uranium concentration being adsorbed within each incremental unit of gravel, there is a minimum of one order of magnitude conservatism in the natural system, given at least 15 feet of gravel between the base of tailings and the semi-confining zone. DOE's calculations indicate that uranium will be completely removed through neutralization and adsorption within the upper few l
feet of the Tertiary upper gravel. Although molybdenum is unaffected by i
contact with the gravel, the molybdenum concentration after neutralization is below the MCL of 0.10 mg/L.
l Ambient ground-water quality measurements within the uppermost aquifer indicate that arsenic, net gross alpha, and combined radium-226 and radium-228 exceed the proposed EPA maximum contaminant levels (MCLs) at the disposal site. These constituents result from naturally occurring sources which are leaching into the ground water.
NRC staff has reviewed the geochemical characterization of the disposal site and agrees that measurements indicate a significant geochemical attenuation capacity within the Tertiary upper gravel and the lahar unit.
5.6
t 5.2.4 Water Use i
A.
Processing Site j
Over 500 registered wells are situated within a two-mile radius of the i
processing site. Most of these wells are domestic, including a few livestock wells; City of Gunnison production wells; and private industrial production well s. Wells completed in the alluvial aquifer mostly penetrate only the upper 10 to 30 feet of the saturated thickness. All the city production wells are located within the city limits and are hydraulically upgradient of the site.
Five residences along Goodwin Lane and the Dos Rios subdivision, which has over 100 domestic wells, are downgradient of the site. DOE performed a risk assessment to evaluate the human health effects from uranium and other hazardous constituents in ground water from these domestic wells.
Current carcinogenic risk estimates, summed over all pathways and radionuclides at both the Goodwin Lane and Dos Rios subdivisions, indicate that an unacceptable health risk exists if residents continue to use contaminated ground water at current levels for a lifetime. The total radiological risk would be more than two orchrs of magnitude above the accepted level. DOE is currently supplying bottled water to residents of the Dos Rios subdivision, Goodwin Lane, and employees of VALC0 gravel company (located south of the processing site).
Alternative supplies of shallow ground water, other than from the alluvial aquifer, are not readily available.
Fresh water is available from the deeper Dakota and Entrada Sandstone formations.
These units are estimated to have a combined thickness of about 350 feet, with a total dissolved solids content of about 1000 mg/L. Anticipated yields from this sandstone are expected to vary from 20 to 200 gallons per minute.
Surface water from the Gunnison River and Tomichi Creek are also potential alternative water sources.
Rights for water use from these sources are based on prior appropriation.
B.
Disposal Site Six registered wells are situated within a two-mile radius of the disposal site.
Four of these wells are domestic and two are used for livestock, with the well depths ranging from 20 to 51 feet.
Five of these wells are located within the floodplain alluvium of Tomichi Creek, and the other well is located in an upland alluvial valley, approximately 2000 feet downstream of several springs.
The nearest well is more than 7500 feet from the disposal site.
Ground-water development in the disposal site vicinity is not expected to increase over the next 50 years. The regional aquifer yields only limited l
quantities of ground water; and the ambient ground-water quality exceeds the EPA MCLs for arsenic, gross alpha, net gross alpha, and combined radium-226
[
and radium-228. Currently, the only development near the disposal site is the Gunnison County municipal landfill.
Surface water is another potential source of water in the area.
The rights for use of this surface water is prioritized on a first-come, first-serve basis.
5.7
)
l
i 5.3' Conceptual Desian Features to Protect Water Resources s
00E proposes to relocate the tailings and contaminated materials from the l
Gunnison processing site to a designated disposal site approximately seven miles to the east. The disposal cell will be excavated to a depth of about 14 o
feet within the upper Tertiary gravel formation. The tailings and contaminated materials will be placed in the disposal cell in lifts and compacted at near optimum moisture levels.
The disposal cell's cover design consists of the following five components, in ascending order:
(1) 1.5-foot-thick radon barrier constructed with clay from -
borrow sources located near the disposal cell. The radon barrier will*be amended with five percent bentonite and compacted, which will result in a saturated hydraulic conductivity of approximately 1 x 10'7 cm/sec; (2) 6-inch drainage / bedding layer with a designed saturated hydraulic conductivity at least one to two orders of magnitude greater than the underlyir.g radon barrier. This bedding layer will serve as a capillary break ano hydmlically separate the radon barrier from the overlying cover component; (3) 73 inches of selected fill to serve as a frost protection layer. The designed. saturated 4
hydraulic conductivity of the frost protection layer ranges from 2.7 x 10 to i
2.7 x 10'7 cm/sec; (4) 6-inch bedding layer with a designed saturated hydraulic conductivity at least one to two orders of magnitude greater than the underlying frost protection layer, and (5) 6-inch riprap layer for erosion i
protection. All components of the cover are designed to comply with the EPA ground-water protection standards and ensure that long-term performance does i
not rely on active maintenance. The overall low permeability of the cover is designed to mitigate the long-term seepage from entering the tailings.
~
The main natural component that will serve to protect the water resources at the disposal site is the chemical attenuation capacity of the Tertiary upper i
gravel and the lahar unit. Measurements and modeling performed by DOE indicate that the majority of the hazardous constituents will be attenuated within the upper few feet of the Tertiary upper gravel. A summary of the attenuation capacity is presented in TER Section 5.2.3.
DOE has also stated that the available capacity within the upper gravel should be sufficient to i
neutralize all of the tailings porewater that is expected to seep from the I
disposal cell.
NRC staff was concerned about the potential for development of a "hard pan" i
layer below the tailings material, resulting from the rapid precipitation of insoluble salts over a relatively short distance within the unsaturated zone.
j A "hard pan" of reduced permeability.would promote moisture accumulation j
within the disposal cell, and could adversely impact the disposal cell i
stability.
NRC staff performed a calculation, using information provided in i
the RAP, which compared the potential volume of mineralized " cement" generated l
by the attenuation mechanism to a conservatively estimated volume of-effective porosity within the upper gravel over a five-foot depth. The calculation revealed that the net reduction in effective porosity from the addition of the
" cement" precipitate was on the order of less than one percent. This small reduction in effective porosity should not adversely impact the ability of the upper gravel to transmit the expected seepage from the disposal cell.
5.8
'l
t 1
DOE also evaluated the likelihood of cementation resulting from mineral precipitation, and determined that the potential for mineral cementation was very low, primarily due to gypsum replacing in situ calcite in the neutral-ization reaction.
DOE concluded that this would not be a hinderance to the ground-water compliance strategy.
NRC staff has reviewed the conceptual design features of the disposal cell and agrees that the combination of the engineered features and the natural attenuation component should protect the uppermost aquifer in accordance with 40 CFR 192.20(a)(3)(iii).
5.4 Disposal and Control of Residual Radioactive Materials (RRM)
DOE must demonstrate that the planned RRM disposal complies with the site-specific ground-water protection standards and closure performance standards in 40 CfR 192, Subparts A and C.
The four areas that must be addressed are:
(1) Water Resources Protection Standards for Disposal; (2) Performance Assessment; (3) Closure Performance Demonstration; and (4) Ground-Water Monitoring and Corrective Action Plans.
5.4.1 Water Resources Protection Standards for Disposal The three elements needed for the disposal compliance demonstration with the water resources protection standards include:
(1) a list of hazardous constituents associated with the tailings material; (2) a corresponding list of concentration limits for the listed constituents; and (3) a Point of Compliance for monitoring the hazardous constituent concentrations in the uppermost aquifer. Details of each of these three elements for the Gunnison disposal site are presented below.
5.4.1.1 Hazardous Constituents Hazardous constituents were identified from the tailings pore fluid characterization, and from knowledge of the uranium recovery process at the processing site. Hazardous constituents are those compounds listed in Table A of Part 192, or Table 1 of Part 264, or Appendix VIII of Part 261 which are identified in, or reasonably derived from the uranium mill tailings.
DOE has identified eleven inorganic hazardous constituents related to uranium processing that exceed the laboratory method detection limit from Table A of Part 192, and Table 1 of Part 264. These constituents are:
arsenic, cadmium, chromium, net gross alpha (gross alpha minus uranium), lead, molybdenum, nitrate, radium-226 and radium-228, selenium, and uranium. Additionally, the following nine hazardous constituents included in the Appendix VIII list were also identified:
antimony, beryllium, cobalt, copper, nickel, thallium, tin, vanadium, and zinc.
DOE has referenced Appendix I of the revised (post 1987) 40 CFR 192 as the j
list of hazardous constituents. Appendix I of the proposed revision to Part 192 is equivalent to Appendix VIII of Part 261. DOE also lists four elemental constituents that are components of hazardous constituents in Appendix 1; aluminum, ammonium, fluoride, and strontium. These elements are components of hazardous compounds that would not exist under the expected geochemical i
5.9
I conditions within uranium mill tailings.
Consequently, these compounds and' their respective elements are not considered as hazardous constituents at the Gunnison disposal site.
DOE proposed to delete nitrate from the list of hazardous constituents for ground-water monitoring.
DOE's rationale for this proposal was that nitrate was not identified as a component of the Gunnison milling process and that nitrate did not present a hazard to human health at the levels measured in the tailings porewater or the downgradient monitoring wells at the processing site. NRC staff did not agree with DOE's assessment, since nitrate was measured above the MCL in some sf the lysimeter samples collected in 1984 and 1990. Upgradient and downgradient measurements of ground-water quality do not fully identify nor disprove any impact of nitrate on the shallow ground-water quality.
Elevated ammonium levels identified in the tailings porewater could also affect nitrate concentrations in the tailings (ammonia conversion to nitrate through biochemical processes).
NRC staff also questioned whether tailings from other mills could have been reprocessed at the Gunnison mill, thus accounting for the identified ammonium and nitrate. The data do not provide reasonable assurance that nitrate is not contained in the Gunnison tailings, or could not reasonably be derived from ammonia conversion under certain geochemical conditions.
Nitrate was consequently included in the hazardous constituents list to address NRC staff concerns.
The processing site was screened for organic hazardous constituents listed in 4
Appendix VIII of 40 CFR 261.
DOE states that there were no organic constituents above the method detection limits at the processing site.
However, results of a special study of 12 UMTRA Project sites (letter dated November 8, 1989, Document No. JEGA/VMT/Il89-0472) indicate that diethylphthalate, bis (2-ethylhexyl) phthalate, and trichlorofluoromethane were identified in tailings and ground-water samples at the Gunnison processing site. DOE provided an evaluation of the occurrence of these compounds and stated that bis (2-ethylhexyl) phthalate occurred in three ground-water samples (both onsite and background), and also in a spiked laboratory sample, which was not spiked with any phthalates. DOE concluded that the detected phthalates were due to inadvertent laboratory contamination.
Trichloro-floromethane was identified in only one sample and was below the method quantification limit.
4 NRC staff has reviewed DOE's hazardous constituent characterization according to the following three criteria:
(1) whether the constituents are reasonably expected to be in or derived from the tailings; (2) whether they are listed in Table A of Part 192 or Appendix VIII of Part 261; and (3) whether they were detected in the tailings or ground water at the site (NRC, 1988).
Based on this review, NRC staff agrees with the listed constituents presented in the 1
RAP, in accordance with 40 CFR Part 192.02(a)(3)(1).
5.4.1.2 Concentration Limits DOE proposes to meet the HCL or background concentration, whichever is larger, for the hazardous constituents identified in the mill tailings. These limits constitute the maximum hazardous constituent concentrations in ground water that may not be exceeded in the uppermost aquifer, downgradient from the 5.10 3
i d'sposal cell.
DOE proposes to use the statistical maximum hazardous i
constituent concentrations identified in the ground water as the background concentration limit, for those hazardous compounds which do not have a designated MCL. The statistical maximum concentration accounts for potential temporal variations from seasonal influences that could produce a false-positive statistical detection. Table 5.1 provides a list of the designated constituents and the proposed concentration limit for each constituent.
TABLE 5.1 Proposed Concentration Limits for the Gunnison Disposal Cell Maximum Proposed Hazardous Contaminant Concentration Constituent level (mo/L)
Limit (mo/L)
Arsenic 0.05 0.05 Cadmium 0.01 0.01 Chromium 0.05 0.05 Net gross alpha 15.0 pC1/L 15.0 pCi/L Lead 0.05 0.05 Molybdenum 0.10 0.10 Nitrate 44.0 44.0 Radium-226 & -228 5.0 pCi/L 5.0 pCi/L Selenium 0.01 0.01 Uranium 0.044**
0.044 Statistical Maximum Background Proposed Hazardous Measured Concentration Constituent Concentration (mo/L)
Limit (mo/L)
Antimony 0.003 0.003 Beryllium 0.01 0.01 Cobalt 0.05 0.05 Copper 0.02 0.02 Nickel 0.04 0.04 Thallium 0.01 0.01 Tin 0.005 0.005 Vanadium 0.01 0.01 Zinc 0.006 0.006 1
Represents the mole equivalent for nitrogen from nitrate.
Concentration converted to mg/L from the 40 CFR 192.02, Table A uranium limit of 30 pCi/L.
5.11 l
1 NRC staff has reviewed the ambient ground-water quality data from the' disposal site and agrees with the use of the statistical maximum as a proposed concentration limit for hazardous constituents that do not have specified MCLs.
5.4.1.3 Point of Compliance The Point of Compliance (PUC) is defined as a vertical surface that extends downward into the uppermost aquifer along the hydraulically downgradient limit of the disposal area. Monitoring wells should be located as close to the disposal cell as practicable, without disturbing the engineered components intended for the long-term tailings isolation. DOE proposes to implement the P0C along the downgradient edges of the disposal cell within the lower Tertiary gravel, which has been designated as the uppermost aquifer. The downgradient edges of the disposal cell are situated along the northern, eastern, and western sides of the cell. The POC will be monitored with six monitoring wells situated along the downgradient sides of the disposal cell.
NRC staff agrees with the proposed P0C designation within the lower Tertiary gravel and the locations of the six monitoring wells along the downgradient sides of the disposal cell, as shown in the RAP.
5.4.2 Performance Assessment DOE must demonstrate that the disposal cell performance will comply with EPA's ground-water protection standards in 40 CFR 192, Subparts A and C.
The performance assessment should show that the estimated concentration of each hazardous constituent at the POC in the uppermost aquifer does not exceed the concentration limits described in TER Section 5.4.1.2 and Table 5.1.
DOE conducted a vertical, one-dimensional computer simulation of the hazardous constituents' movement through the upper unsaturated zone of the disposal i
site. The simulation was performed using the WORM code, an updated version of SUMARTA-I. The modeling efforts focused primarily on molybdenum and uranium, because these constituents were the least attenuated in the batch column tests, as discussed in TER Section 5.2.3.
The other hazardous constituents exhibited substantially larger distribution coefficients than either molybdenum or uranium. The simulation neglected the effects of the lahar unit, because the lahar is not continuous over the entire disposal site area.
The simulation assumed an upper boundary infiltration flux of 1 x 10'T cm/s, which is a conservative value that represents the designed hydraulic conductivity of the radon barrier. The simulation results indicate that molybdenum and uranium will not exceed their respective MCLs in the uppermost aquifer within a 1000-year period.
NRC staff has reviewed the input data and computer simulation results and agrees that the concentration of hazardous constituents from the disposal cell should not exceed the proposed hazardous concentrations within the 1000-year period, in accordance with 40 CFR 192.02(a)(1).
5.12 e
s 5.4.3 Closure Performance Demonstration In accordance with the closure performance standards of 40 CFR 192.02(a)(4),
DOE is required to demonstrate that the proposed disposal design will:
(1) minimize and control contaminant releases to ground water and surface water, (2) minimize the need for future maintenance, and (3) meet the initial design performance standards. DOE has stated that the geochemical attenuation capacity of the native, unsaturated soils beneath the disposal cell will prevent the hazardous constituent concentrations from exceeding the proposed concentration limits within the maximum design life of the cell. DOE's demonstration is based on bench-scale attenuation tests using soil materials collected at the disposal site and tailings fluid collected from the processing site. A summary of the testing results are presented in TER Section 5.2.3.
The potential surface-water impact has been evaluated along the vertical seepage pathway to the uppermost aquifer and then laterally to the nearest surface water (Tomichi Creek) by ground-water transport. A surface-water impact from this pathway is not likely, because the P0C in the uppermost aquifer will not be affected by hazardous constituents from the disposal cell within 1000 years. However, a potential pathway to surface water by lateral movement of unsaturated moisture within the upper Tertiary gravel was assessed.
DOE performed a one-dimensional, vertical unsaturated contaminant flow evaluation that indicates a potential buildup of unsaturated moisture beneath the disposal cell, above the semi-confining unit. The incised gullies, which expose the semi-confining unit near the disposal cell, present a potential for the near-surface exposure of tailings moisture by lateral unsaturated flow.
The tailings moisture could then potentially be carried to the surface through intermixing with infiltrated rainwater and discharged through short-lived seepage at the semi-confining layer exposures. However, DOE has indicated and NRC staff agrees that the attenuation capacity within the Tertiary upper gravel will likely prevent the lateral movement of hazardous constituents.
Long-term active maintenance of the disposal cell has been mitigated by the proposed use of natural, stable materials in sufficient quantity to achieve a design life of at least 1000 years. TER Section 5.3 presents a discussion of long-term moisture accumulation within the disposal cell, resulting from potential "hard pan" development.
The likelihood of a "hard pan" developing beneath the disposal cell appears small.
The NRC staff agrees with DOE's assessments and evaluations concerning the anticipated performance of the disposal cell in demonstrating compliance with the ground-water standards.
5.4.4 Ground-Water Monitoring and Corrective Action Plans DOE is required to develop ground-water monitoring and corrective action plans which will be carried out during the post-disposal period, in accordance with the proposed ground-water protection standards in 40 CFR 192.02(a) and (b).
DOE will monitor ground-water quality in the uppermost aquifer through the POC 5.13 l
1 I
wells along the downgradient edges of the disposal cell. Two existing upgradient monitoring wells will be retained for post-disposal monitoring, and 1
will serve as background wells for the monitoring period. DOE will provide a Long-Term Surveillance Plan (LTSP) which will address various monitoring needs of the disposal cell. Although the concern over lateral seepage reaching the gullies is small, DOE should address the need for sampling any observed water seepage along the existing gullies in the LTSP.
DOE is required by 40 CFR 192.02(c) to provide an evaluation of alternative corrective actions that could be implemented if the disposal cell monitoring program indicates that the cell is not performing adequately. DOE will provide a detailed corrective action plan in the LTSP for concurrence by NRC.
i DOE should consider reasonable failure scenarios of the disposal unit and state that corrective action could be implemented no later than 18 months after detecting an excursion.
NRC staff agrees with the ground-water monitoring plans as presented in the
[
RAP, considering that a detailed monitoring program and corrective action assessment will be provided in the LTSP.
5.5 Cleanuo and Control of Existino Contamination i
DOE must demonstrate compliance with the EPA standards listed in 40 CFR Part 192, Subparts B and C, for the cleanup and control of existing ground-water l
contamination. DOE proposes to defer the ground-water cleanup compliance i
demonstration at the Gunnison processing site. Ground-water restoration will i
be addressed under a separate DOE program as part of the National Environmental Policy Act (NEPA) process. NRC staff considers that the ground-water cleanup may be deferred, as provided by the UMTRCA amendments of 1982.
Ground-water cleanup of the Gunnison processing site may be deferred if DOE l
can demonstrate that:
(1) ground-water cleanup of the site will not be impacted by the tailings disposal, and is separable from the remedial actions, and (2) public health and the environment will be protected.
t i
DOE has demonstrated, by virtue of moving the tailings to the Gunnison i
disposal site, that the tailings disposal will not impact ground-water cleanup at the processing site. However, uranium concentrations which exceed the proposed MCL have occurred in ground-water samples from domestic wells and 00E monitoring wells at a distance of more than 2000 feet downgradient from the processing site. This represents a risk to human health, i
e DOE and the Colorado Department of Health have jointly developed a program for testing water from Gunnison homes that are potentially affected by ground-water contamination associated with the milling operation. The program is designed to determine uranium concentrations in ground water used by residences near the existing tailings pile. The testing program and a health-risk assessment, undertaken by DOE, have indicated that there is some risk to the residents who continue to use ground water from the uppermost alluvial i
aquifer.
The DOE UMTRA Project has committed to providing an alternative water source to the affected water users.
However, the action levels that trigger when the 5.14
t
. alternative supply is provided and when it'will end, are currently being evaluated by DOE and the Colorado Department of Health. DOE is presently supplying bottled water to residents in the Dos Rios subdivision, residents along Goodwin Lane, and employees of VALCO (south of the site). All residents that are potentially affected by ground-water contamination from the processing site will be provided an alternative water supply. The proposed alternate supply will consist of a county water system usir.g surface water from the Gunnison River, which will be treated and provided to residents through a water distribution system.
NRC staff concurs on the deferral of ground-water cleanup at the processing site, because of DOE's current water supply measures and the commitment to provide an alternate water supply to all residents that are potentially affected by ground-water contamination from the processing site.
5.6 Conclusions The ground-water monitoring plans are acceptable because DOE will provide specific details in the LTSP on the monitoring program and corrective action assessment. Therefore, NRC staff concludes that DOE's proposed remedial action has demonstrated compliance with all but one of the EPA ground-water standards, based on the review of the Final Remedial Action Plan and supporting documents. 00E, with NRC's concurrence, has proposed to defer the ground-water cleanup of the Gunnison processing site until a later project phase.
Consequently, the following open issue will not be resolved until the ground-water cleanup plan is approved:
Demonstrate compliance with EPA's final around-water cleanup standards in 40 CFR 192, Subparts B and C (see Issue 11 in Appendix A).
5.7 References DOE (U.S. Department of Energy), " Remedial Action and Site Design for Stabilization of the Inactive Uranium Mill Tailings Site at Gunnison, Colorado, Final," UMTRA-DOE 050508.0000, 1992.
EPA (U.S. Environmental Protection Agency), " Health and Environmental Performance Standards for Uranium Mill Tailings", Code of Federal Regulations, Title 40, Part 192, Subparts A through C, Federal Register Vol. 52, No.185, 1987.
JEG (Jacobs Engineering Group), letter transmitting results of Appendix IX Screening, November 8, 1989. Document No. JEGA/UMT/1189-0472, 1989.
NRC (U.S. Nuclear Regulatory Commission), " Final Standard Review Plan for Review of Remedial Action of Inactive Mill Tailings Sites under Title I of the UMTRCA, Rev.1," Division of Low-Level Waste Management and Decommissioning, June 1993.
NRC (U.S. Nuclear Regulatory Commission), " Draft Technical Position on Information Needs to Demonstrate Compliance with EPA's Proposed Groundwater Protection Standards, in 40 CFR Part 192, Subparts A - C," Technical Branch, LLWM/NMSS, 1988.
5.15
t 6.0 RAD 0N ATTENUATION AND SITE CLEANUP 5.1 Introduction This section of the TER documents the staff evaluation of the radoc attenuation design and processing site cleanup for the planned remedial action at the Gunnison, Colorado, UMTRA Project site. The evaluations are primarily of the material characterization, radon barrier design, proposed remedial action, and site verification plan to assure compliance with the applicable EPA standards. The review followed the NRC Standard Review Plan for UMTRCA Title I sites (NRC, 1993).
6.2 Radon Attenuation To meet the EPA standards for limiting release of radon-222 from residual radioactive materials to the atmosphere, contaminated material will be relocated to the disposal site and capped with soil from the Sixmile Lane borrow site. DOE is proposing a multi-layer cover for the disposal cell consisting of:
radon barrier layer, drainage layer (select fill Type A),
frost protection layer (select fill Type B), bedding layer (select fill Type A), and a riprap layer.
This cover, up to the top of frost protection layer, has been designed to limit the average release of radon to meet the EPA standard of 20 pCi/m /s.
The NRC staff review of the cover design for radon attenuation included evaluation of the pertinent design parameters for the contaminated materials, radon barrier soil, select fill Type A, and frost protection layer soil, and a review of the specifications for materials placement. The staff considered that the barrier layer is designed to satisfy criteria for construction, settlement, cracking, and infiltration of surface water, as well as reduction of radon gas release at the surface of the completed cell.
The parameters of the other layers of the cover were evaluated for their ability to protect the radon barrier layer from drying and disruption. The stability of the cell as a whole was also assessed because of the potential for cracking of the barrier layer due to settlement or heaving. These aspects of the cell design are discussed in detail in Sections 3 and 4 of this TER.
NRC staff evaluated DOE's input values for the RAECOM computer code that were used to calculate the radon barrier thickness required to meet the radon flux limit. The RAECOM code input was analyzed as discussed below.
The staff then performed an independent analysis of the design using the RAD 0N code (NRC, 1989), which is a modification of the RAECOM code.
6.2.1 Evaluation of Parameters The required thickness of the radon barrier depends on the characteristics of the radon barrier soil (s) and the underlying contaminated materials.
NRC staff evaluated the physical and radiological data for the contaminated materials and the radon barrier soils used for input into the RAECOM and RADCN computer codes.
In some cases, conservative estimates instead of measured values were used for input, and in other cases measurements wi.re made, 6.1
-,,--,w
-e
--,,w
, + -
v' v" ' ' ' -
l although not under design (.onditions. NRC staff evaluated the justification and assumptions made, to confirm, that each value was representative of the material, consistent with anticipated construction specifications, and conservative, or based on long-term conditions. The sampling and testing methods for the materials were also reviewed to determine their appropriateness and to insure that the data was sufficient.
The design parameters of the contaminated and cover materials that were evaluated include:
long-term moisture content, bulk density, specific gravity, porosity, material layer thickness, and radon diffusion coefficient.
In addition, the radium content and radon emanation coefficients of the contaminated materials were evaluated.
A. Contaminated Materials Physical characteristics of the contaminated materials were determined from laboratory testing of representative samples. The results from a series of maximum density tests (ASTM D-698) on tailings materials were type-weight-averaged to determine a representative maximum dry density. As the specifications require relocated contaminated materials to be compacted to 90 percent of the maximum Standard Proctor dry density, DOE selected an average density, at 90 percent compaction, of 1.67 gm/cc (104 pcf). Results from a series of specific gravity tests on tailings materials were used to determine an average of 2.70.
A porosity of 0.38 was then calculated for the tailings materials.
The corresponding properties for subpile/off-pile materials were assumed to be: in-place compacted density of 1.84 gm/cc, specific gravity of 2.75, and porosity of 0.33.
Since the tailings, subpile, and off-pile material will be excavated and placed concurrently in the disposal cell, an average volume-weighted density of 1.73 gm/cc, specific gravity of 2.72, and porosity of 0.36 were calculated for the contaminated layer.
DOE selected a representative long-term moisture content value for the tailings materials after examining the results of 80 natural (in-situ) moisture measurements that gave a weighted average of 17.1 percent, ten 15-bar capillary moisture tests (6.5 percent for sand / slimes,14.6 percent for slime samples), and assuming 4.0 percent for the sand fraction. DOE conservatively selected a long-term moisture content of 6.2 percent by weight (calculated saturation fraction of 0.27) based on the tailings type-weighted average results of the capillary moisture tests. The cobbly soils (subpile and off-pile material) had an average long-term moisture content of 4.0 percent (saturation fraction of 0.22) based on three 15-bar capillary moisture tests.
For the tailings /subpile/off-pile material layer, a representative long-term moisture content of 5.4 percent was estimated by volume-weighting the moisture contents of these materials.
For the tailings, DOE selected a radon diffusion coefficient of 0.032 cm /s I
2 based on tests on four samples at 6.2 percent moisture by weight at 90 percent compaction. DOE presented recent data on the radon diffusion coefficient of cobbly soil measured in special equipment (Attachment 5, DOE, 1992a) and concluded that the value can be reasonably estimated from measurements made on 6.2 i
_ t only the finer soil fraction of cobbly material. Basedontpismethod, DOE estimated an average radon diffusion coefficient of 0.022 cm /s for the cobbly subpile and off-pile materials at 4.0 percent moisture.
For the combined tailings /subpile/off-pile mixture, a conservative value of 0.032 cm /s was 2
used in the RAECOM analysis.
Testing indicated that the radon emanation property of the material at the Gunnison processing site is independent of its moisture content and Ra-226 concentration.
Eight measurements were made on each of the three different fractions of tailings. The type-weighted average emanation coefficient for the tailings was 0.18.
Three measurements of radon emanation were performed on the finer fraction of the composite subpile/offpile soils. The results ranged from 0.11 to 0.58 and the average was 0.31.
Due to the high variability, limited sampling, and lack of verification of the representiveness of the tailings samples, a conservative radon emanation coefficient of 0.35 (RAD 0N default value) was used for the contaminated material Ic.yer in the computer model.
The radium content of the contaminated materials was determined primarily by gamma spectroscopy. The various contaminated materials identified at the Gunnison site are:
tailings pile, subpile, mill yard and ore storage area, windblown contaminated area, perimeter roads, demolition debris, and vicinity property materials.
DOE did not consider the latter three categories in the radon barrier calculation due to inadequate data to provide reasonable volume and average Ra-226 characterizatior. However, DOE assumed that the average radium content of these materials will be low. The tailings pile average Ra-226 concentration was 327 pCi/g based on 379 samples.
The average upper limit Ra-226 concentration was determined to be 56.5 pCi/g, based on 179 samples, for the combined windblown, mill yard, and ore storage areas (off-pile materials). The subpile average of 9 samples was 26.5 pCi/g, including the amount of Ra-226 that would result in 1000 years from decay of the Th-230 that is present.
The volume-weighted average Ra-226 concentration i
for the single contaminated layer in the disposal cell model is 222 pCi/g.
The total contaminated material thickness of 12 meters (40 feet) is based on estimated volumes of the various contaminated materials.
B.
Radon Barrier, Frost Protection, and Drainage Layer Materials The physical properties of the radon barrier and frost protection layer materials were selected by DOE based primarily on the results of laboratory testing of samples from the Sixmile Lane borrow site from where both will be obtained. The specifications require the radon barrier soil to contain a minimum of 50 percent fines (material passing the No. 200 sieve), and be compacted to 100 percent Standard Proctor density at, or on the wet side of, optimum moisture.
The frost protection layer material is specified to contain a minimum of 30 percent fines and be compacted to 95 percent Standard Proctor 1
density (Specifications Section 02200, pages 11 and 26; Attachment 1, DOE, 1992a). DOE has conservatively tested samples compacted to densities less than those required for placement, and used most of the resulting values in the radon attenuation model.
6.3
o l
The average specific gravity for the radon barrier soil was 2.71 (seven tests). The average density, based on 16 tests compacted to 95 percent of maximum Standard Proctor, was 1.66 g/cc (103.7 pcf) resulting in a calculated porosity of 0.39.
Based on these results, the 100 percent compaction density (required at placement) for the radon barrier soil was estimated to be 1.75 gm/cc (109.2 pcf), resulting in a calculatd ponsity of 0.35.
These latter values were used in the model.
The average specific gravity for the frost protection soil was also 2.71 (five tests). At a compaction of 82 percent, the frost protection soil had an average (eight tests) density of 1.45 gm/cc (90.0 pcf), and a calculated porosity of 0.46.
At the 95 percent compaction required at placement, the values are estimated to be 1.69 gm/cc for density, 2.71 for specific gravity, and 0.38 for porosity.
Both sets of data were used in RAECOM analyses. The values obtained at 82 percent compaction represented the soil parameters after freeze-thaw damage.
For the radon barrier soil at 95 percent compaction, seven capillary moisture saturation tests yielded an average of 13.4 percent, and 11 in-situ moisture measurements averaged 9 percent. DOE chose 9 percent by weight for the long-term moisture content (saturation fraction 0.39). The 9 percent long-term moisture content was assumed even for the bentonite amended soil, although DOE stated that a value of 13 percent would be more realistic for the amended soil.
Frost protection soil at 82 percent compaction had an average 15-bar moisture content (three tests) of 8.9 percent.
The one in-situ moisture content measurement for this material was 6.5 percent. The lower value was selected to represent the long-term moisture content for the frost protection soil.
The radon diffusion coefficient for the radon barrier material, based on tests of two samples compacted to 95 percent and at 9 percent moisture content, was 2
0.015 cm /s.
Since this material will be amended with 5 percent bentonite and compacted at 100 percent, DOE stated that a conservative estimate of the 2
amended Gunnison barrier material diffusion coefficient would be 0.005 cm /s.
However, calculation 643-01-02, Sheet 03, indicates that three samples of barrier g/s. oil amended with 5 percent bentonite had a diffusion coefficient of 0.012 cm DOE presented data from the Rifle UMTRA Project site that indicatedthatadding10percentbyweightbeptonitetothesoildecreasedthe diffusion coefficient from 0.020 to 0.0002 cm /s. DOE has stated that the final value used in the RAECOM analysis will depend on further testing and that the justification for the constructed radon barrier thickness will be based on radon diffusion measurements of the bentonite-amended Sixmile Lane soil (RAS, page 83).
NRC staff concludes that by adding 5-10 percent bentonite, the radon barrier layer will achieve the design value. This issue is discussed further under 6.2.2 below.
Diffusion coefficient tests were performed on three samples of frost haw protection soil at 82 percent compaction, to simulate cumulative freeze-}/s degradation.
Based on these data, the diffusion coefficient is 0.022 cm for a moisture. content of 6.5 percent (saturation fraction of 0.20).
For 95 percent compaction, a value of 0.018 was assumed. This estimate is not 6.4
important since NRC staff concludes that the model should incorporate the test values obtained at 82 percent compaction to represent the long-term condition of the frost protection layer.
The drainage layer material, designated as select fill Type A (sand-gravel),
between the radon barrier layer and frost protection layer, is also part of the material that would limit the release of radon through the cover. Because the gravel and sand material is to be processed from the granular soils at the borrow site, the physical properties and design parameters of this layer are assigned. The layer is assumed to have a long-term moisture of 4.0 percent, specific gravity of 2.75, and in-place compacted bulk density of 1.83 gm/cc (114 pcf). These values result in a calculated porosity of 0.34 and a saturation fraction of 0.22.
The diffusion coefficient of cobbly soil was not used because this layer will be coarser p/s, for use in the model.
rained select fill Type A.
DOE selected a more conservative value, 0.032 cm The staff considers these assigned values to be reasonable for sand-gravel granular soil.
The thickness of the various layers of the cell cover is set by the design and the estimated long-term frost penetration.
This is acceptable to NRC staff, assuming that the amount of bentonite would be increased from 5 percent to as high as 10 percent, if necessary, in order to meet the design criteria.
C.
Ambient Radon The ambient air radon concentration is another required parameter value for the RAECOM model, and has been measured in the Gunnison area.
The average value is 1.0 pCi/1. The technique used to measure the radon concentration, and the result of the measurements is acceptable to NRC staff.
D.
Conclusions NRC staff has reviewed the physical and radiological parameter values assigned i
by DOE to the contaminated materials and to the various materials of the disposal cell cover considered in the radon barrier design. Staff concludes that, given the conservative assumptions employed and verification to be done by further testing, they are acceptable for the radon barrier design.
For the radon barrier layer, DOE plans to amend the soil with a minimum of 5 percent by weight ben onite, with the intention of achieving a diffusion coefficient of 0.005 cm}/s.
DOE has stated (RAS, page 89) that the final cover design will be based, in part, on final measurements of the radon diffusion coefficient and minus 15-bar moisture tests for the bentonite-amended soil. NRC staff accepts this situation with the understanding that the final data, and the revised RAECOM analysis incorporating this data, will be provided for review as part of the Completion Report, and will incorporate an ample margin of safety.
]
6.2.2 Evaluation of Radon Attenuation Model DOE used the RAECOM computer code to evaluate the radon attenuation of the 2
cover for compliance with the EPA radon flux standard of 20 pCi/m /s.
DOE 6.5 i
~
'?,
proposes a composite radon barrier consisting of an Ib-inch-thick, i
fine-grained earthen radon barrier layer amended with 5 percent by weight bentonite, overlain by a 6-inch (sand-gravel) drainage layer, and a 73-inch frost protection layer.
DOE's RAECOM analyses (calculation 643-01-02, i
Appendix G) used the parameters discussed above, varying the p/s to represen adon diffusion coefficient of the radon barrier layer from 0.015 to 0.005 cm the addition of increasing amounts of bentonite up to 5 percent by weight.
To be conservative, the expected changes in density, porosity, and moisture values, as the bentonite increased, were not used for the computer input.
The frost protection layer was evaluated at both 82 percp/s,respectively),
nt and 95 percent compaction (diffusion coefficients of 0.022 and 0.018 cm j
with corresponding changes in the bulk density cod porosity. NRC staff agrees that the frost protection soil should be modeled at 22 percent compaction to i
simulate cumulative freeze / thaw events, for the final estimation of radon fl ux. DOE has presented adequate justification for selecting 82 percent relative compaction to represent the frost-damaged soil.
With the frost protection layer modeled for frost damage, RAECOM results 2
indicated that in order to achieve a radon flux of 20 pCi/m /s, an 18-inch-i thick bentonjte-amended radon barrier layer would need a diffusion coefficient 2
of 0.0072 cm /s. The design value of 0.005 cm /s provides a small margin of i
safety.
NRC staff used the RAD 0N computer code to model the radon flux using the conservative combination of parameters proposed by DOE. As DOE reported, the analyses resulted in a radon flux of less than 20 pCi/m /s at the tcp of the frost protection layer, provided that a low enough diffusion coefficient is used for the radon barrier layer.
There are no site-specific data to demonstrate what percent bentonite is needed in the radon bg/s).
rrier layer to achieve DOE's design radon diffusion coefficient (0.005 cm However, DOE has committed to testing the bentonite-amended barrier layer material and assuring that the required degree of radon attenuation is achieved.
Based on the data presented, NRC staff is confident adequate radon attenuation can be achieved with less that 10 percent by weight bentonite in the radon barrier layer.
NRC staff has recommended (NRC,1993) using 0 for the ambient radon value to be conservative. Although DOE reported the ambient radon as 1.0 pCi/1, a value of 2.0 was used for RAECOM input. However, use of 2.0 instead of 0 in the computpr model for this parameter resulted in a radon flux difference of 0.02 pCi/m /s which is insignificant.
j DOE has not performed a parametric or sensitivity study to assess the relative impact of those parameters that have been assumed, or that have been based on minimum testing. DOE states (RAS, page 87) that the RAECOM input values are highly conservative and provide an ample margin of safety. NRC staff concludes that the current design of the radon barrier does have an acceptable degree of conservatism for the preliminary radon flux estirates.
Based on review of the design and analyses presented in the RAP and associated documents, NRC staff concludes that the radon attenuation model supports the radon barrier design, but that model input values must be substantiated by 6.6
further testing and analysis during material placement. This will be accomplished since DOE has committed to base the final cover design on measurements of the as-built bulk Ra-226 concentrations and associated emanation fractions, and the radon diffusion coefficient and minus 15-bar moisture tests for the bentonite-amended soil (RAS, page 93).
6.3 Site Cleanuo 6.3.1 Radiological Site Characterization Field sampling and radiological surveys at the Gunnison site identified approximately 718,900 cubic yards of contaminated materials covering 68 acres at the processing site and adjacent areas. Subpile contamination exceeds 15 pCi/g Ra-226 above background to an average depth of 2.5 feet. This cut-off concentration includes the radium that would result in 1000 years from the decay of existing Th-230.
DOE had identified, on a construction drawing, that contaminated areas along Gold Basin Road were wetlands and were not to be excavated. DOE subsequently indicated that these areas were not offically designated wetlands, and that a Project Interface Document (change order) would be issued so that these areas would be excavated.
Background levels of Ra-226 were measured in the Gunnison area and the average value is 1.7 pCi/g. The methods and results for radionuclide levels are acceptable to NRC staff.
6.3.2 Cleanup Standards DOE has committed to excavate contaminated areas to meet the EPA standard of 5 pCi/g (surface) and 15 pCi/g (subsurface) plus background, for Ra-226 in soil, and to place the contaminated materials in an engineered disposal cell.
Excavation will be monitored to ensure that cleanup efforts are complete. The surface will be restored to a grade that controls surface drainage.
There are no buildings or equipment remaining to be decontaminated at the processing site. Demolition debris from Phase I of construction is stored on the site and will be buried in the disposal cell.
Since most of the processing ::ite has a high percentage of cobbles and gravels greater that a number 4 sieve, DOE proposed use of the generic procedure for determining the bulk radium and thorium content of cobbly soil for excavation control and verification which the NRC staff has concurred in by letter of April 4, 1992. A report on the detailed site-specific procedures used will be provided in the Completion Report.
00E states that supplemental standards for Th-230 contamination will be based on the NRC-approved generic thorium policy. Uranium concentrations, after the Ra-226 has been removed to meet standards, will be assessed by a pathway analysis of potential environmental and health impacts.
If remedial action is indicated, a supplemental standard will be proposed.
6.7
\\
6.3.3 Verification The final radiological verification survey for land cleanup will be based on 100-square-meter areas. DOE may use a variety of measurement techniques, depending on particular circumstances. The standard method for Ra-226 verification is analysis of composite soil samples by gamma spectrometry.
Verification for Th-230 will follow the generic thorium policy. Areas known, or suspected of containing elevated levels of Th-230 below the depth of Ra-226 1
remediation, will have 100 percent of the grids analyzed. Windblown areas will not be analyzed for Th-230, and other areas verified after the generic policy is final, will have 10 percent of the grids verified for Th-230.
No on-site structures at the processing site will require radiological verification, since all structures have been demolished and the debris will be buried at the disposal cell.
6.4 Conclusions Based on review of the radon attenuation design of the Gunnison remedial action plan as presented in the final RAP and supporting documents, NRC staff concludes that the radon barrier, as designed, should meet the applicable EPA standards contained in 40 CFR Part 192.02. DOE has committed to achieve the required radon barrier diffusion coefficient to meet the radon flux standard, and has indicated that the final cover design will be based, in part, on measurements of the radon diffusion coefficient and minus 15-bar moisture tests for the bentonite-amended soil. NRC staff accepts this situation with the understanding that the final data and the revised RAECOM analysis incorporating this data, will be provided for review in the Completion Report.
In general, the staff finds the radiological characterization program, the proposed processing site cleanup, and the verification plan acceptable as supporting the conclusion that the remedial action will meet the EPA standards. As stated above, a report on the detailed site-specific cobbles-to-fines procedures will be provided in the Completion Report.
6.5 References DOE (U.S. Department of Energy), Washington, D.C.,
" Remedial Action Plan and Site Design for Stabilization of the Inactive Uranium Mill Tailings at Gunnison, Colorado, Remedial Action Selection Report, Final," and Attachments 1 - 5, 1992a.
" Uranium Mill Tailings Remedial Action Project (UMTRAP), Gunnison, Colorado - Information for Bidders," Volumes I - VII, 1992b.
--, " Uranium Mill Tailings Remedial Action Project (UMTRAP), Gunnison, Colorado - Design Calculations," Volumes I - VI, 1992c.
MK-Ferguson, Remedial Action Inspection Plan, Review D, UMTRA Project l
Gunnison, Colorado, August 1993.
6.8
t NRC (U.S. Nuclear Regulatory Commission), Washington, D.C.,
Regulatory Guide 3.64, " Calculation of Radon Flux Attenuation by Earthen Uranium Hill Tailings Covers," June 1989.
--, Division of Low-level Waste Management and Decommissioning, " Final Standard Review Plan for the Review of Remedial Action of Inactive Mill Tailings Sites under Title I of the Uranium Mill Tailings Radiation Control -
Act, Revision 1," June 1993.
)
i 6.9
4 e
I APPENDIX A STATUS OF THE DRAFT TECHNICAL EVALUATION REPORT (TER) OPEN ISSUES dTER RAP SECTION l
OPEN ISSUE SECTION WHERE RESOLVED 1.
DOE should clarify which specific over-consolidated Tertiary deposit (s) is to be 2.3.1 considered as bedrock, the welded tuff or page 42 all Tertiary deposits.
I 2.
DOE must adequately characterize faulting 2.3.4.2, in the disposal site area.
2.4.3 Sec. 2.2.4, p.16 l
1 3.
DOE must adequately justify their Calc. 640-05-03; will conclusion on the settlement analysis and 3.3.3 preload area most j
potential for cracking of the cover.
susceptible to differential settlement 1
l 4.
Specifications of radon barrier material need to be revised to reflect the fines Calculation content of the materials tested to 3.3.4 680-01-01; and determine the design parameters. Also, additional testing consideration should be given to planned specifying intermediate sieve sizes.
5.
DOE needs to clearly state the maximum value of the coefficient of hydraulic conductivity that satisfies the infiltration barrier criteria.
If the i
i required value is 1 x 10'7 cm/s, DOE
'i should include discussion on the 3.3.4 conservatism achieved by amendment with pages 15-16
^
bentonite.
If the required value is 1 x 10'8 cm/s, DOE should provide data to i
support the assumption that laboratory
[
l determined values are representative of field values, given the specificatien of the material.
6.
Drawings and specifications indicating Revised contaminated
}
the placement of contaminated materials material placement.
need to show the sequence of placing the 3.4.1 Drawing D5-10-0310 various contaminated materials, and Spec. Section l
consistent with design assumptions in the 02200 Part 3.3.B 1 radon barrier analysis.
3.4.A 7.
Specifi'ations should include a Spec. Section 02200 requitement for the placement moisture 3.4.1 Part 3.5.B.3 implies content for contaminated materials.
moisture limits for a
design compaction A-1 1
~-
w
- = - -
dTER RAP SECTION OPEN ISSUE SECTION WHERE RESOLVED 8.
Both radon barrier and frost protection New borrow subarea (Type B fill) materials are borrowed from avg. 69% fines and the same source with different percent minimum 50% fines fine requirements.
Frequency of 3.4.2 (Calc. 680-01-01);
gradation testing should be increased for propose more testing the barrier material. The staff plus visual considers one test per 200 cubic yards or inspection during 20 truck loads to be appropriate.
construction 9.
The radon barrier borrow material specifications and the RAIP should be revised to include a Plasticity Index 3.4.2 Revised Section 6.4.4 requirement on the material, and the of the RAIP frequency of Atterberg limits (PI) testing should be specified.
t
- 10. Specifications and the RAIP should state the frequency of testing to verify that Spec. Section 02228 the moisture of the in-place radon 3.4.2 Part 3.3 barrier will not dry beyond the specified range.
RAIP adequate
- 11. 00E has deferred ground-water cleanup to a re9arate phase of the program. NRC staff agrees with this deferral, provided that human health and the environment are not adversely impacted (see Issue 15).
When plans for ground-water remediation Section 4.0 are developed, DOE must include. adequate 5.2.1, characterization of the thickness of the 5.2.3 bedrock and alluvial deposits in the To be addressed in vicinity of the processing site, along the Ground-water with the hydraulic relationship between Remediation Plan the alluvial aquifer system and the deeper ground-water systems in the area, and redefine the contaminant plume characteristics.
- 12. DOE must determine the influence of ground-water flow originating north of the disposal site, in the vicinity of the 5.2.2, Gunnison County landfill. Appropriate 5.4.1.3 Section 3.2.1 revisions to the ground-water monitoring plans should be made.
A-2
<., S dTER RAP SECTION OPEN ISSUE SECTION WHERE RESOLVED
- 13. DOE should conduct a detailed evaluation of the potential impact of hazardous constituents along the unsaturated lateral flow path in the upper gravel.
If the detailed evaluation indicates a 5.2.2, potential impact along the aforementioned 5.4.3 Section 3.2.1 pathway is possible, then DOE should modify the cell design to minimize the impact and develop a monitoring program to demonstrate that the design modifications are effective.
- 14. DOE must perform the batch column tests with the representative tailings fluids Section 3.2.6; from the Gunnison processing site to 5.2.3 Calculation verify the test results that utilized the GUN-06-92-13-05-02; spiked surrogate samples.
RAS Report Sec. 5.6
- 15. DOE must demonstrate that future public health and safety will not be adversely impacted by deferring ground-water c.leanup at the Gunnison processing site.
The status of residences in the affected area other than the Dos Rios subdivision, 5.2.4, should be detailed in the RAP.
Also, DOE 5.5 Section 4.0 should provide information on measures (alternative water sources) taken to mitigate the risk to other affected residences and state the action levels for supplying alternative water.
- 16. 00E should include a detailed evaluation RAS Report Section of mineral cementation potential of all 5.1.4.2; precipitated salts over the expected 5.3 attenuation distance in the upper gravel.
Section 3.2.6; Section 3.2.2
- 17. DOE must address the occurrence of RAS Report p. 53-54; organic compounds detected at Gunnison in 5.4.1.1 Attach. 3 Sec. 3.1.5; the special sampling study conducted in Attach. 4 Sec. 3.1.1 1989.
- 18. DOE must reevaluate the POC monitoring well locations and provide assurance that all downgradient flow paths will be 5.4.1.3 adequately monitored. Appropriate Section 3.4 revisions to the ground-water monitoring plans should be made.
A-3
4,.,
dTER RAP SECTION OPEN ISSUE SECTION PHERE RESOLVED
- 19. The radon diffusion measurements for processing site subpile soil and large samples of cobbly soil, as mentioned in RAS Report the RAS Report (pages 78 and 79), should 6.2.1 Section 6.3.2; be provided in the final RAP. The Calc. 643-01-02 subpile material data should be used in the RAECOM code.
- 20. A detailed justification for selecting the density value of 82 percent relative compaction, to model the frost-damaged 6.2.1 RAS Report soil, must be provided to demonstrate Section 6.4 that a conservative radon diffusion coefficient was used in the RAECOM code.
- 21. DOE needs to present an analysis of the radon attenuation model that includes a diffusion coefficient derived from tests RAS Report on bentonite-amended Sixmile Lane borrow 6.2.1, Section 6.4, material that just meets the 6.2.2 and later testing to specifications for percent fines, models be done on bentonite-the contaminated material placement based amended soil on RAP specifications or a worst-case scenario, and discusses the conservatism.
- 22. DOE needs to provide details on the design and location of the trench for debris and building components. Also, RAS Report explain why the DOE release criteria is 6.3 Section 6.2, not to be used and why the material is trench deleted being buried if it meets release criteria.
- 23. DOE needs to provide a detailed site-, generic specific procedure for determining procedures radium / thorium content of cobbly soil 6.?
(bulk nerages) for cleanup or site site-specific verification.
procedures to be in a,
report l
l A-4 l
..t o;
- STATUS OF OPEN ISSUES RESULTING FROM REVIEW OF TINAL RAP (based on letters to DOE dated 12-17-92, 2-25-93, and 4-19-93)
OPEN ISSUE RAP Section HOW/WHERE RESOLVED 1.
The cleanup standard for bulk Th-230 RAS Report Section 6.5.2 should not be 35 pCi/g since the suppl-page 91 revised mental standard is to ensure that the Section bulk Ra-226 levels will conform to the 6.5.2 EPA standards for 1000 years.
DOE must consider the residual Ra-226 in each grid and not assume it is < 6.2 pCi/9 2.
The statement on the Th-230 cleanup RAS Report Statement deleted standard at the Durango site was bottom of misleading.
DOE must clarify that page 91 supplemental standards for Th-230 in Section excess of the projected 1000-year Ra-226 6.5.2 standard, will be based on site-specific conditions.
3.
DOE needs to elaborate on the RAS Report Reference NRC-verification of Th-230 that reflects the top page 93 approved generic proposed policy with which NRC has Section policy agreed.
6.5.3 4.
DOE should provide a frost protection Calc.
Justification in design that meets the 1000-year 640-02-02 Calc. 9-420-03-00 criterion, or include justification that its use is impracticable and a 200-year criterion is adequate.
5.
There is conflicting information of Note 13 on Revised drawing to whether or not the wetlands areas will drawing GUN-indicate removal of be excavated. DOE should indicate what PS-10-0212 areas adjacent to areas are official wetlands and provide unpaved runway, for justification for not excavating PID-S-11 Rev.1 contamination from unofficial wetlands.
6.
DOE should revise Specification 02228 to Spec. 02228 Testing and QC will require a minimum of six percent Sec.
ensure that a bentonite to ensure the design (five 3.2.B.2, minimum of five percent) is met.
Calc.
percent bentonite 640-05-03 is incorporated.
7.
DOE should revise the list of hazardous RAS Report NRC agreed to constituents with HCLs to include p.70; delete silver; nitrate and silver.
Attach. 4 nitrate listed 8.
00E should revise text that references RAS Report Text revised the designed radon barrier hydraulic
- p. 30; conductivity as being 1x10'8 cm/st.r.
Attach. 4 Previous Comment: Clarify if stored Executive Materials hazardous materials will be placed in cell.
Summary-RAS designated as RRM A-5