ML20217N392

Informs of Completion of Review of Remedial Action Plan for Upper Burbank Disposal Cell & Processing Site at Naturita Title I Site.Concluded Addl Info Needed Re Erosion Protection,Radon Barrier Design & Geotechnical Engineering

ML20217N392
Person / Time
Issue date:	12/22/1995
From:	Bell M NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS)
To:	Gillen D NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS)
Shared Package
ML20217N400	List: IA-99-336, Informs of Completion of Groundwater Review of Naturita, Colorado Final Completion Rept Dtd Nov 1998.Determined That All Monitor Well Abandonment Repts & Maps Have Been Correctly Signed & Approved by Licensed Engineer ML20217N392 ML20217N670 ML20217N695 ML20217N740 ML20217N774 ML20217N831 ML20217N879 ML20217N946 ML20217N965 ML20217P011 ML20217P041 ML20217P045 ML20217P055 ML20217P139 ML20217P306
References
FOIA-99-336 NUDOCS 9910290013
Download: ML20217N392 (53)

Text

{{#Wiki_filter:_ (! 2 ,i December '.,,L'1195 71 .t MEMDRANDUM TO: Daniel M. Gillen, kling trauch Chief MLUR/IRM/NMS$ j FRON: Michnel J. ' Bell, Bruch U.hf EFE8/UW/ mss

SUBJECT:

MATURITA TE,R IMptli-EECTIOC 2. 3l, A, AND 6 I In accordance with Charlotte Abrams' rtem.t we have completed our review of the Remedial Action plan (rap) for the Ltter Burbank disposal cell and the procesting site at the Naturtta Title I site. Based on our/ review, we conclude that additional information regsrding erosion protection, radon barrierdesign,geotechnicalengineering,andsel'saicdesjgabasiswillbe needed before the RAP can be approved. At Ps. Abrans' request and In order to help expedite the process,lts for the we previously provided preliminary draft versions ofeour review resu

4. 'nd 6',

These preliminary Technical Evaluation Report of Sections 2, 3 a draftshadnotundergonemanagementreviewwIthin.ENGB. This memorandum transaits the approved ENG8 review of the Naturita RAP. Reference lists as however, ye have not specified a also provided at the end of each sectionE sections formet for references in the various ENG This review was performed'by Steve McDuffle, Section !; Dan Roe, Section 3; Ted Johnson, section 4 and Elaine Brummett Sedton 6. If you have any question, please contact the appropriate revieder.

Attachment:

As stated ( / 19115r/f ENG8r/f JHolonIch* MFederlins* CAbrams Tasktog 95-7 POCIMDif NME: 8:\\0let\\EN05\\t!M\\ERTRfA.M1 we Er u ainai. / nnet Ence C non n>%i.no % CmGlt~/ (WbitW sneWfHe'* Ja'Ps..31$11ss. m M' meldd&m 4?/E!u nem .a. we anse .1 men 1 nur Ncennen METf'TW DATI 1213 45 I IB/IIll OfF GIAL RM) 00tl ACDRH Yf.3 , M l re i vt _m i.t. rn..rt., di iris.tioni res __ no.e, LS$ s fli _,_ ND 9910290013 991027 PDR FOIA MIDDLET99-336 PDR

wwm '$'N ,o, 4 .e g 2.0 GEOLOGIC STABILITY " 2.1 Introduction -j I This section of the draft ITER documents the staff's review of geologic and seismologic information for the Upper Barbad disposal cell, slated for disposal of the remaining 4 Title I material from the Naturtte site. Appendix A of 10 CFR Part 40 require the disposal cell to be closed in accordance with a s design which provides reasonable assurance of ctMrol of radiological hazards h for 000 years, to the extent reasonably achtenbl% and, in any case, for at least 200 years. HRC interprets from this standani that certain geologic conditions must be set taterovide reasonable asskr uct that the long-term performance objectives will be achieved.,.g } This review follows ther Final Standard Reyfew Plan (5RP) (NRC,1993) for the Review and Remedial Action of inactive Miti Taf!!ngs Sites under Title I of IMTRCA. The review is based on informatten f,rovided in the Naturita Remedial Action Plan (RAF)t,(DOE,1995), references cited 'in the RAP, additional references avalla e in the geologic literatured and a telephone conference call between NRC staff and DDE and its contractors. 2.2 Location 4 The Upper Surbank cell is located on Club Mesa, an crosional remnant bounded by the San Miguel and Dolores River valleys, as well as Hieroglyphic Canyon. The mesa Ifes immediately west of Uravan in wester:i Colorado. TER Section 1.2 contains additional location information. q 2.3 Gaglagg o The MP contains a description of the alte geolo which is cosplied from a i variety of maps, books, and journal articles ayy able in the open Itterature. 2.1.1 Physlographic Setting The site Iles in the eastern portion of the Colorado Plateau physlographic province, a roughly circular region cornring mJch of eastern Utah, northern Artrona, northwestern New Mexico, and western Colorado. The Plateau is characterized by scel arid ellmate elevatioe" m itly over 1500 meters, and, with the notable exceptlen of the haradow Basin, oedrock of generally flat-lying sedimenttry snits. The site is in the Canyonlands section of the Plateau, en area characterfred by deeply incised drainages and isolated remnant sesas. The major landforms ner.r the site are the Uncompahgre Plateau to the northeast, and the l'aradox Sasin to the southwest. Club Mesa is one of several sesas between the: San Mievel and Delores Alvers. Tributaries of these streams have incised the upland between them to define these mesas, which are characterited by bedrock dipping at IWW engles to the northeast. The Uncespahgre Plateas ie a northwest tretding vpland about 160 km long and 60 ka wide. The southwest border f a marked geomorphically by the San M'guel and Delores Alver valleys, and the Mrtheast border by the Gunnison and Decespahgre River valleys. The Pietetu was spilfted during the Late Crotoceous farly Cenerels Laraelde orogeny, presumably due to reactivation of high angle faults in the Precogbriah basement rocks (Ely, et al.,1986). As

1; +j.g'ff', y

• e u

y,- g 14 a,, p 19 r 9 9 the Precambrian basesint uplifted, Mesozoic sedimaets faulted and folded into monoclinal structures which presently drape the arr/hs of the Plateau. The site lies approximately 10 km southwest of the Wge 1tne marking the base of the southwest monocline. The Paradox Basin is a northwest-trending, elliptially shaped region of relatively low relief within the Colorado Platsau. Its long axis extends over 200 km free near Green River, Utah, to Cortel, Colorado. The Paradox Basin is a paleostratigraphic' basin which accumulated;evaportte deposits 1-2 km thick in the Pennsylvantan* Period. An overburden of clastic sediments shed in a southwesterly direction from the ancestral Uncompahgre uplift, in combination with basement fault blocks trending toward the northwest, caused the salt beds to deform into a series of northwest-trending antic 1tnes End synclines from the Pennsylvanian through Jurassic Periods (Baars, 1989). Overlying Mesozoic sediments were folded"during this episode of. salt diapirism. Subsequent dissolution of salt has caused some subsidence of the Mesozoic units, and deformation during the Laraaide Orogeny may have further folded Paradox Basiti sediments. The northeastern-most anticitne of the Paradox Basin underlies the Paradox Valley, approximately 5 km southwest of the site. Club Mesa lies i between the Uncompahgre Plateau and Paradox Basin. Based on a review of the RAP and other references, the staff finds the physlographic setting to be adequately characterized. 2 1.2 Stratigraphic Setting The Upper Burbank site is underlain by a sequence of Mesozote and Paleozoic sarine and continental sedimentary rocks. A> thin sequence of pre-Pennsylvanian ilmestones and shales may overlie Precambrian basement at considerable depth, but these rocks are not exposed in the site area. The Pennsylvanian Hermosa Group comprises the evaporttes which have experienced ductile deformation througiout the Paradox Basin. This unit is interpreted to although it thickens rapidly toward its bethinanddeepbeneaththesitel cline 6kmsouthwestintheParadoxValley. outcropping at the core of an ant The Perulan Cutler Formation is a sequence of continental clastics derived from the ancestral Uncompahgre highlands, and the unit is several V.lometers thick beneath the site. The Triassic Moenkopi and Chinle FcanaW ons, both are separated from the Cutler and shallow water clastics with minor Ilmestone,ko I and overlying Chinle are from each other by unconformities. The Moen relatively thin in the site area (approximatel 30 and 60 m, respectively), and only the Chinle has minor exposure. The i lassic Wingate Sandstone and are conformable over the Chinle, Kayenta formation (sandstone and slitstone)Itkewise exposed only in deeply are each about 60 m thick near Uravan, and incised channels. The water table is near the Win ate-Kayenta boundary beneath the Upper Burbank cell. The Jurassic Nava o Sandstone conformably overlies the Kayenta Fermation. Table 2,1 of the P states that this n,rsation is not found at the site aret but is found in the western part of the site region. However, the geologic map of the site area, figure 3.2, shows Navajo landstone exposed in the bottom of Hieroglyphic Canyon near the confluence with the San Miguel River, less than 2 km northeast of the Upper Burbank disposal cell, t

1:- Q $ g y M. j g 7 .e a, w 4 to ,( 6 9 u the Precambrian basesint spilfted Mesozoic sediwaets faulted and folded into monoclinal structures which presently dr. ape the wryhs of the Plateau. The site lies approximately 10 km southwest of the Moge line marking the base of 4 the southwest monocitne, m.; The Paradox Basin is a northwest-trend {ng, alliptically shaped region of relatively low relief within the Colorado Platsau. Its long axis extends over 200 km from near Green River, Utah, to Cortei, Colorado. The Paradox Basin is a paleostratigraphic' basin which accumulatedjevaportte deposits 1-2 km thick in the Pennsylvanian' Period. An overburden of clastic sediments shed in a southwesterly direction from the ancestral Uncompahgre uplift, in combination with basement fault blocks trending toward the northwest, caused the salt beds to defone into a seri's of northwest-trendt anticlines and synclines from e the Pennsylvantan through Jurassic Periods aars,1989). Overlying Mesozoic sediments were folded"during this episode c. salt diapirism. Subsequent dissolution of salt has caused some subsidence of the Mesozoic units, and deformation during the Laramide Orogeny may have further folded Paradox Basin sediments. The northeastern-most antic 1tne of the Paradox Basin underlies the Paradox Valley, approximately 5 km southwest of the site. Club Mesa lies between the Uncompahgre Plateau and Paradox Basin. I Based on a review of the RAP and other references, the staff finds the Physiographic setting to be adequataly characterized. 2 1.2 Stratigraphic setting 1 The Upper Burbank site is underlain by a sequence of Mesozoic and Paleozoic marine and continental sedimentary rocks. A? thin sequence of pre- ) Pennsylvantan limestones and shales may overlie Precambrian basement at i considerable depth, but these rocks are not exposed in the site area. The Pennsylvanian Hermosa Group comprises the evaporttes which have experienced ductile defonsation throughout the Paradox Basin. This unit is interpreted to be thin and deep beneath the site, although it thickens rapidly toward its outcropping at the core of an anticilne 6 km southwest in the Paradox Valley. j The Peralan Cutler Formation is a sequence of continental clastics derived from the ancestral Uncompahgre highlands, and the unit is several kilometers i thick beneath the site, The Triassic Moonkopi and Chinle Formations, both j are sepsrated from the Cutler and shallow water clastics with minor limestone,kopi and overlying Chinle are j from each other by unconformittes. The Moen relatlyely thin in the site tres (approximately 30 and 60 m, respectively), I and only the Chinle has minor exposure. The Triassic Wingate Sandstone and are conformable over the Chinle, Kayents Fonsation (sandstone and siltstone)Ilkewise exposed only in deeply I are each about 60 m thick near Uravan, and incised channels. The water table is near the Win ste-Kayenta boundary beneath the Upper Burbank cell. The Jurassic Nava o Sandstone conforsahly overlies the Kayenta Fersation, Table 2,1 of the P states that this formation is not found at the site area but is found in the western part of the site region. However, the geologic map of the site area, figure 3.2, shows Navajo landstone exposed in the bottom of Hieroglyphic Canyon near the confluence with the San Miguel River, less than 2 km northeast of the Upper i Burbank disposal cell. t i

9 I w$ { { i t x .f- } M. y Approximately 40 m of Jurassi~c EntrW Sandstm ue::ahrmably overlies the Navajo. Entrada forms steep slopes in the lower Mrt.ioes cf Club Mesa. The Summerville Formation, a marine shale with stitstent., h conformable over the Entrada. This unit is about 30 m thick near 'the rite, urd it serves as an aquitard inhibiting vertical fluid transport. Avrtch the stratigraphic column is the Jurassic Morri,s'on Formation, which seim as -the cap rock of Club Mesa. The lower Salt Wash Member is the host fact for the tailings cell. This fluvial sandstone and siiltstone unit is more the 100 m thick at Club Mesa. Select horizons of th'e' unit are highly compelerst; the basin which will hold the Naturita materials was excavated to recovey Salt Wash bedrock for use as riprap for erosion protect' ion on other tallings 611s nearby. Immediately up-slope of the cell (southwe'st) is the base of the} Brushy Basin Member of the Morrison Formation. Over loot s of variegated shale', mudstone and sandstone of theBrushyBasinarepresentponClubMesa. A remnant of the Cretaceous Burro Canyon Formation remains at the highest elevations of the mesa, approximately 1.5 km west of the Upper Burb'ank cell. This remnant consists of approximately 50 m of fluvial sandstone ovFrlain by shale-mudstoni. Any Tertiary units which may have been present in the site area have tsen removed by erosion. NRC finds that DOE has sufficiently described the s ratigraphy of the site area. e 2.3.3 Structural setting The site lies within the eastern portinn of the Colorado Plateau, a relatively stable, intracontinental subplate with greater crustal thickness than adjacent provinces. Two major shear zones were established in Precambrian time, and these features appear to exert some control on the Plateau features present today. The northwest-trending Olynsic-Wichita Lineament extends from ~ Washington to Oklahoma; the Uncompaigre uplift and Paradox Basin lie within this trend. The northeast-trending Colorado Lineament extends from Arizona to Minnesota and may have some relation to localized northeast-trending areas of seismic activity in the Colorado Plateau (Wong, 1981; Bernreuter, et al., 1995). These broad lineaments intersect within the-site region (Baars and Stevenson,1981). The Colorado Lineament apparently served as a preferred pathway for Tertiary magmas as they intruded Mesozoic sediments. The Miocene laccolithic intrusions of the La sal Hountains are in the junction zone of the Colorado and Olympic-Wichita linehments. In addition, Abajo Mountain, a stallar intrusion, is within the Colorado Lineament farther to the southwt. Club Mesa exhibits two orthogonal fracture sets, northeast and northwest trending, but these are regional sets which are probably not related to t,- lineaments. The site Iles several kilometers northaast of the Paradox Valley antic 1tne, which contains strata faulted'and folded from salt dispirism and perhaps Laramide deformation. Some northwest trending faults are associated with this antic 11nes these are discussed further in TER Section 2.4.3 on seismotectonic stability. Several klionsters northeast of the site is the Uncompahgre Plateau,whichwasup11fledalong) basement-rooted,steeplydipping, curved reverse faults (Ely, et al.,1986 during the Laramide Orogeny (Late CretaceoustoEocenetime). Ther; may be associated with shallow-rooted normal faults which compensate for curvature in the: reverse faults as they a v' i - 3 f a', i L

t a . m.c; t y ,i,

t..-@

j o 1 plunge beneath the Precambriah core of the" uplift. :Some structures bounding the Plateau remain seismicall) active today, though possibb from reactivation as extensional structures. The bounding fault labeled huit 81 by Kirkham and Rogers (1981) has no historical seismicity but war,dtt n a b d to be capable. This fault is responsible forl the design earthquake tf the Upper Burbank dte (seeTERSection2.4.3). n! .., y%. w -j a Club Mesa lies on the southwest limb of the norttwshtrending Nucia syncline. This broad, shallow synclinelles between the salt-cored Paradox Valley anticline, the northeastern-siost anticline of the PAfadox Basin, and the faults and monoclinal folds Mrking the southwest m.irgin of the Uncompahgre Plateau. The strata of Clubpsa dip homoclinally to the northeast t.t I to 3 degrees. Thestafffindsthatthestrukturalsettinghasbeenadequatelydescribed. P ,c 2.3.4 Geomorphic Setting { There is considerable topogra'phic relfef near the Up' par Burbank site due to both tectonic uplift and erosional denudation processes. Between Uravan in and the crest of the the San Miguel valley (about~1525 m elevation)here is nearly 1500 m of relief. Uncompahgre Plateau 20 km northeast (3000 m) t The Upper Burbank cell is approximately 200 m above Uravan on the eastern margin of Club Mesa, in a quarry into the Salt Wash Member of the Morrison Formation. Club Mesa is bounded on the north by the San Miguel River valley, the east by tributary Hieroglyphic Canyon, and the west by the Dolores River valley. The southwest margin of the mesa is marked by a tributary of the Dolores River. The mesa is approximately.4 km east to west and 5 km north to south. The mesa edges are generally steep,' supported by the resistant Salt Wash Member and Entrada Sandstones. The site is drained by side drainages of Hieroglyphic Canyon. ,b ( c Both the San Miguel and Dolores Rivers are undersized for their valleys, indicative of greater runoff.during Pleistocene deglaciation. The braided channel patterns and wide valleys of these rivers contrast with the steep, narrow valleys of their tributaries and the dendriti.c drainages of tributary headwaters. Since Miocene time, the major geomorphic processes in the area have been incision and widening of major stream valleys. Valley widening occurs largely through scarp retreat. Mesa cliffs consist of resistant sandstones, and retreat occurs primarily by mass wasting of more erodible underlying units. t i NRC considers that the RAP adequately describes the geomorphic setting. j 2.3.5 Seismicity ThesitelieswithinanareathatissubjecttoseismiceventsintheColorado Plateau, the Plateau's more se.ismically active bordet zones, and the surrounding tectonic provinces. The border zones are defined by Kirkham and Rogers (1981) based on structural boundaries and trends in seismicity. The ColoradoPlateauisboundedonthesouthandwestby;theBasinandRangoProvince, w 2 C i j 4 ~

b., h[. gp W e. b o Mf f l Uinta-Elkhead and Wyoming B'asin Provinces. There is some debate about the precise geographic boundari'es of the provinces an'd Plateau border zones, but there is little question th'at seismicity overall is greater around the margins than within in the Plateaut(Kirkham and Rogers,1981). However, any characteristic seismic eve'nt in the Plateau boundaries or adjoining provinces would be too far away (>50!;km) to adversely impac't the site. Historical seismic records show less than a dozen earthquaket of magnitude 5 or greater within the Colorado Platenti and all of these wer proximal to the margins. e An 1882 earthquake with ma'gh,itude estimated at 6.5 or greater may have been centered on the northeast margin of the Plateau (McGuire, et al.,1982), although more recent evideh~ce places the epicenterj in the Colorado Front Range (Kirkham and Rogers, 1986) ! There are several regionali' structural features within the Colorado Plateau which contribute to the site's seismic hazard: Th'e Colorado Lineament, the Paradox Valley fault system, and the Uncompahgre liplift. Many small seismic events have occurred along' specific segments of th' Colorado Lineament in the e past. However, these spatlh11y limited earthquake' swarms have differing characteristics (Wong, 1981), and, overall, the lineament does not appear to be a major seismogenic structure (Brill and Nutt11, 1983). Bernreuter, et al. (1995) believe that a 50 ka'long basement fault beneath the Colorado River south of Moab, Utah, may be" capable of an event as large as magnitude 7 but with very low probability.A This section of the Colorado Lineament is more c than 100 km west of the site. The staff notes that Sections 2.10.2 and 4.2.3 of the RAP misinterpret Behirsuter, et al. (1995)iregarding the name of this seismic zone. Bernreuter,9et al. (1995) do not label t; tis northeast-trending zone the Moab fault; they ' ecognize the Moab fault'to oe a northwest-trending r structure most likely due to salt tectonics. U Ten-fifteen kilometers southwest of the site, tk ortheastern limb of the Paradox Valley anticline hosts several struct:.res one of which is 62 km long. Thit, fault was labeled as Fault 90 by Kirkham and: Rogers (1981). The. RAP states that this and parallel faults of Paradox Va'11ey are almost certainly underlain by structures related to evaporite flow or dissolution; Hunt (1969) and Cater (1970) are cited as support for this interpretation. The RAP states that because of this relationship, the normal length-magnitude relationships do not apply. Therefore, Fault 90 is interpreted to be capable of no more than a magnitude 3 event. NRC requires additional information to demonstrate that Fault 90 is rooted in the salt stratigraphy, thus capable of no more than a magnitude 3 event and exempt from standard length-magnitude relationships. If the standard relationship holds for Fault 90, then it would exceed the capabilities of Fault 81 in terms of ground motion' at the site. The staff is aware that DOE is currently compiling information on similar long, northwest-trending faults near the $1tek lock UNTRCA Title ! site. The staff believes that analyses demonstratingthe salt tectoalc character of the Big Gypsum Valley fault group near $1 ck Rock may also serve as evidence that the Paradox Valley faults are salt tectonic structures. The tectonic character of Fault 90 is an M. t t Steeply dipping faults associated with the Uncompahgre uplift approach the site as close as 15 km, and'at least one of theseifaults appears to be historically seismogenic 1!n 1985, an event of magnitude 2.9 centered 55 km 5 i 5

T I ', i' ' f (fQQ*:7:E:} .;'.C f, .yd p-north-northwestofthehtemayhaveoccurredslongthesteeplynortheast- ,,.,c dipping Granite Creek fault (Ely, et al.,1986 There is some question whether the Uncompahgre uplift is continuing or)('whether this seismicity results from reactivation of reverse faults astextenstenal features. Cater (1966) states that the iiplift continues today,'however, Ely, et al. (1986) believe that the 1985 event was extensional. The Uncompahgre-related structure which passes 15 km from-the site is called Fault 81 by Kirkham and Rogers (1981). Cater (1970),found this fault to have Quaternary offset, so it is considered to be capable.9With'a length of 34 km and possibly capable of a magnitude 6.9 event, this is currently the desihn fault for the site. p - ; op..m n 7 The staff has some concern about the'11st of historical earthquakes in the northernColoradoPlate4uandsurroundingreg16h, table 2.3intheRAP. The table supposedly lists the same events p otted:jn figure 2.8, but there appear to be far more earthquak,es listed in the table ((particularly those of unknown magnitude) than plottedj'f the data aside from codes such as such as SRA, PD in the figure. Furthermore, no explanation is provided on the source o or USHIS, which have no! explanation. i The table ls difficult to use even for purposes such as identifying the December 6,1985, magnitude 2.9 Gateway earthquake. The table does not enable a reviewer to easily identify the location and magnitude of this event; in fact no 1985 event of magnitude 2.9 is listed in the table. hDespite this weakness jn the presentation of seismic data in the RAP, NRC believes that DOE presentsjadequate inforsation on the historical seismic eventi of the Colorado Plateau and surrounding areas of interest. n y a 2.3.6 Natural Resources The site region contains natural resources of okl, natural gas, uranium, vanadium, coal,andpotash;theimmediateareafeaturesabundanturaniumand vanadium. Deposits of these minerals are concentrated in the Salt Wash Member of the Morrison Formation, the disposal cell host rock. However, boreholes surrounding the site indicate no economic deposits beneath the tallings cell (Umstco and Fael,1994).L Economic deposits of oil or gas are very unlikely, as the site lies near the axis of a low, broad ' syncline. No faults which may serve as hydrocarbon traps are close to the site. Large oli and gas deposits in the area are generally associated with the salt anticlines of the Paradox Basin to the southwest. Potash is mined from the Paradox Salt Member of the Pennsylvanian Hermosa Formation,its near the site are likely too deep to bealtho 100 km from the site. Any depos economically viable. The Dakota Sandstone contains coal deposits in the region, but this unit has been eroded from Club Mesa. Hard rock minerals are mined from the San Juan. Mountains to the southe'ast, but these minerals are found in igneous units which are not present in the site area. NRC finds the RAPadequatelydescribes'naturalresourcesint;hearea. GeoloateStability(; 2.4 in an effort to provide Measonable assurance tnt radiological barriers will remain intact for at le(st 200 years and up to'1000 years to the extent achievable, the RAP provides information on the! bedrock, geomorphic, and seismic stability of the3 site. / h h 6 y i 1i q I { 4 6 4 g: g x n nn

y y.y. w 6 A&b. Q.&-.., k. Af ' 2f j%ec y b Y f" 2.4.1 Bedrock Stability a i The Upper Burbank cell is an excavation entirely within the Salt Wash Member of the Morrison Formati'on, a competent sandstoine and siltstone unit. Thirty-five meters of Salt Wash lies between the basetof the cell and the underlying Suinnerville Formation; po soil lies beneath.the cell. Rock durability studies (Umetco,1995) suggest that portions of.the Salt Wash Member are the most erosionally resistant ho'rizons in.the local stVatigraphic column, one indication of stabilitN The Salt Wash is moderately fractured in vertical, orthogonal northeast-a^n~d northwest-trending sits, with average spacing around I meter. The Summerv111'e Formation is an effei!tive aquitard beneath the Salt Wash. Water originallyTwithin the tallings cells northeast of the Upper h Burbank cell has caused a zone of perched raffinate to form on top of the Summerville, but this lit down-dip and down-gra'dient of the Up,,er Burbank cell. The low dip of the conti.ct, the thickness of o'verlying Sali Wash, and the coherent elastic strength of layers and their ' interfaces indicate that slip along the Salt Wash-Susserville contact plane is virtually impossible, even with the addition of a phrched water table. The closest mapped faults, related to the salt antic 11ne and collapse structures of the Paradox Valley, are 4 km from the site.", Therefore, the cell i's not at risk of displacement due to fault offset. ,I t t P The staff considers that! the RAP adequately addresses the topic of bedrock stability. ) y q 2.4.2 Geomorphic Stability Given the location of the Upper Burbank cell on' Club Mesa, the cell has potential for destabilization by mesa scarp retreat. Scarps at the edge of Hieroglyphic Canyon and: San Miguel River valley will continue to retreat, but long-term erosion ratespare low. The San Miguel River appears to be presently aggrading or migrating across its flood plain u{'pstream of Uravan. incising its channel from Uravan several alles downstream, while it is either The channel incision is not expected to exceed the maximum l average rate of 0.4 m per thousand years (Hunt, 1956; Yeend, 1969; l. arson, et al., 1975) for the Colorado River system in the Colorado Plateau.; The San Miguel River tributaries, such as Hieroglyphic Canyon, are also unlikely to exceed this rate. Can may occur at a rate 3 times that of incision, yon widening (scarp retreat) Mass wasting of large fracture-bounded or up to 1.2 m/1000 years. blocks can lead to high rates of scarp retreat, but the absence of large block accumulations or conical colluvium mounds at scarp bases in the area, as well as the smoothness of canyon walls, indicates that this is not a concern. Moreover, scarp retreat rates on the order of 1 m/1000 years in the site area have been documented byrdating packrat middens.and applying other geochronologic methods (Smith, 1980). NRC concludes that scarp retreat of San Miguel River valley and. Hieroglyphic Canyon does not pose a threat to the site during cell life. (n Headward erosion by Hieroglyphic Canyon tribut' artes which drain the site will not affect the cell, because it will sit 230 m.from the canyon rim. Runoff from the mesa surface up' slope of the cell will: be redirected to avoid the cell,soonlyrainfalltngdirectlyonthecelJcancauseerosion. Use of i [ 7 a% .k e o f ,a

jp.p, fy$ ~ .w riprap will minimize such e'rosion'; this is discussed further in TER Section 4.4. y , e. y y .y N; y.y q e The site'is not at risk from debris flows,' soil crieep, rock falls, or eolian. processes, although the undisturbed surface of Club Mesa does have a very thin mantle of eolian sand and silt. No large solian features, such as dunes or sand ramps are apparent inithe area, although such; features are not specifically addressed in the RAP. Subsidence from salt dissolution is not a hazard in the inusediate vicinity..as any evaporttei which may underlie the site are interpreted to be52-3 ka' deep and relatively thin. The area has extensive uranium and vanad'um deposits, but boreh' oles indicate no economic 1 deposits beneath the Upper 48urbank site. None of Mesapassbeneaththecell.?Thesiteissubjectt{themanymineadditsonClub 4 o ash fall from very large volcanic eruptions in the We~ stern United States, tibt the probability of such an event occurring over the'jlifetime of the cell l's negligible. S t% j The RAP presents sufficient 1l evidence that Club Mesa and the tailings cell will be geomorphically stable far beyond the 1000 year performance period. 1 x. 2.4.3 Seismotectonic St* ability. ', b .2 6 ~ The Club Mesa site is subje'c.t to ground motion frclin seismicity on faults associated with the Paradox Basin fold and fault belt and the Uncompahgre uplift, as well as floating earthquakes and those located within the Colorado 1.ineament. Seismicity along the southwest margin 'of the Uncompahgre uplift appearstopresentthegrea;testhazardtothesiteT ,o o The RAP identifies fault 81kof Kirkham and Rogers (1981) along the Uncompahgre margin as the untrolling fspit for seismic hazardjat the Upper Burbank site. j Although this fault shows no! historical seismicityg it has Quaternary offset based on geomorphic relationships observed in Unaweep Canyon (The RAP states Cater,1970). The fault is 34 km long,le of a magnitude 6.g earttiquake based on the fault paksing within 14.9 km ofTthe cell. that this fault is capab length-magnitude relationship of Bonilla, et al. (1984). The precise dip is unknown, but the fault is suspected to be steeply dipping. Section 4.2.6 of the RAP states that a 6.9 avant from Fault 81 at 14.9 km would produce a peak horizontal ground acceleration (NIA) of 0.306 g atithe site using the attenuation relationship of Campbell and Sozorgnia19,1995(,1994). In a conference call between NRC and DOE staff on December DOE stated that 0.406 g is the mean value from thisi1994 attenuation model.i However, the definition of ' acceleration" on page Ale 4 of the RAP states that the 84th percentile valuefromtheCampbellandSozorgnia(1994)model'isadoptedasthedesign value. An 84th percentile value agrees with the Final SRP (NRC, 1993), which calls for use of this more conservative number. Th's SRP lists Campbell (1981) as an attenuation relationship which can be used to' calculate PHA, and this older model yields a PHA value lower than Campbel1Fand Sozorgnia (1994). DOE stated during the December 19 conference call thatJthe 0.306 g mean value of Campbell and Sozorgnia (1994) is nearly identical t'o the 84th percentile value so the mean value from the 1994'model should be adequate of Campbell (1981),lerationf DOE has provided infomation to NRC to justify for the design acce thisapproach,andthestaff[iscurrentlyconsideringtheadequacyofthis response. Pending a decision by the. staff, the PHA for th's 6.9 design basis h h 8 6 \\ p c e v hn

y 4 q q. dm p ? -.ggy 1 w , p y d g,W 2* c( { earthquake at 14.9 km remal'n;s an M. M f k

p.

J RAP figures 2.g and 3.2 indi'cate that Fault 81 is n'ot the closest fault to the site, nor is it the longestlin the site area.. Several short salt collapse features are present about 4' km west of.the site, and Fault 90 of the East Paradoxanticlinegroupis)2kmlongandonly10kmfrom<thesite. The staff requires additional inforza{1on to determine that Fault 90 is a salt tectonic 1 Section 2.3.5.pable of no more than a magnitude 3 h; vent, as feature and ca ph ' j.,2 g g;, j i vem i l The RAP uses 6.2 as the maximum magnitude'for a fl ting earthquake in the L site area. PHA calculation' tfor a floating earthqiiake assume an epicentral s distance of 15 km, so the floating earthquake PHA is below the design ground l motion from Fault 81. Seismic zones within the Colorado Lineament may be capable of generating events: as large as magnitude 17 with low probability i (Bernreuter, et al.~,1995),(but-the distance of such zones from the site exceed 100 km. Therefore, site ground motions restilting from such events are also well below the design ijround motion for the site. The staff notes that presentation and discussion of)information on seismic sources and hazards in the RAP is adequate but requires significant effort by l the reviewer to evaluate. Haps, tables. and discus'sions are lacking desirable detail and consistency. Examples of omissions and: contradictory statements in the RAP include: i i it 11 Page A4-4 of the RAP fitates that the 1985 Gathay earthquake may be associated with fault's1within the Ute Creek graben. The discussion of this event on pages A2-44 and A2-45 mentionsithe possibility that the event may have occurred on the Granite Creek 4 fault. The relationship between the fault, thetgraben, and faults numbered 76-80 on figure 2.9 l 1s not clearly specified. In addition, Section 4.2.6 on page A4-8 . mentions multiple Gatenay earthquakes, whereas all previous references indicate a single event. s.e - q,.. $ .m. e,.m Table 4.1 lists the Ute Creek fault group, assigning a single length and distance from the site. -Individual fault lengths and distances, such as ~ provided for the Norwood fault group (fault numbers 84-89) are not provided. ,j The 1985 Gateway event"is said to be 78 km from the site at the bottom of page A4-5, but 55 km from the site at the top of page A4-4. On page A4-7, the des iption of the Nomood hleinit faults states that the nearest approach to the site of any of these faufts is 23 km. l However, the Norwood faults listed in table 4.1 list no fault closer than'40 km. Perhaps Fault 83 shown on figure figure 4.1asamemberjoftheNorwoodgroup,r]2.9shouldbeincludedin ir i L PageA4-7statesthatF]ault73is62kmfromthesite,whereasta lists it as 55 km fron the site. .J The site area geologl(( sap, Figure 3.2, showqseveral small salt

5 f7 h

k .h, ? I l

{ collapsefaultsap&'oximatd)4ka"w;:s,t'odtho'pperBurbanksite.

t.,o. -

4 pr e 7-Figure 2.9, the onif comprehensive map of faults: in the site region. shows some small faults 6-7 km southwest of the site but no faults 4 km it to the west. These'itwo figures', as well as'the site region geologic sh'ould all.be consistent.- A map with the importance map, (Figure 2.3)ld$' deally 'show topography and major geologic units of Figure 2.9 wou .t along with structures. Moreover, a map such as this should label the various fault groups' discussed in Section 4'.2.5. Notwithstanding the above:aEd a.on-!dWbb;?linformation relating to Fault ..b. v:) Wir:. ssuming the'. provided, the staff believes that.the m,aximum earthquake detemined for the site, magnitude 6.9 at 14'.9 km,'is'4' conservative. conclusion. 2.5 Conclusions '/; a.: <.. .., :.. m Based on a review of the RAP a. itional' material,'NRC finds that the geology and geologic stability of the Upper Burbank disposal call have been adequately characterized,'aside from questions regarding the tectonic nature of Fault 90 and the peak horizontal. acceleration from the maximum earthquake capable of affecting the site. Resolution'of the' Fault 90 open issue could possibly be aided by infomation on the tectonic character of long faults near the Sitek Rock site; that; analysis may be applicable to Fault 90. Resolution of the PHA open issue is dependent on the staff receiving and accepting the i additional information onlthe calculation of the PHA, as requested during a conference call on December 5, 1995 9 s.. '. u 6 ih

4. '

e , = y - e

d ci o

7, s r 3'] s, n y 10 ') 7 e1 9~or ( ".}.,,'

• b e

e ...a

e 4 t[i,.yjs g,y.7 c 2.6 Referencel M: j y, r" Baars, D., Canyonlands Country, Canon Publisherb.,Ltd., Lawrence, Kansas, 1989. Mi Baars, D.L., and G.M. Stevenson, " Tectonic Evolu, tion of Western Colorado Eastern Utah," in Western Slooe Colorado,JNew Mexico Geological Society Guidebook, 33rd Field Conference, pp. 105 ll2, 1981.

, m;.

Bernreuter, D., E. McDermott,L and'J.= Wagoner, "Siismic Huard Analysis of Title II Reclamation Plans," Lawrence Live'rmore National Laboratory, 1995. g y. j,w c. _ h 9 . : a. Bonilla, M.G., R.K. Mark, and J.J.'t.fenkausper, 1 Statistical Relations Among Earthquake Magnitude, Surface Rupture Length, and Surface Fault Displacement " Bulletin of the Seismologic'al Society of America, v. 74, no. 6, pp. 2379-2411, 1984.

({V;q.

,q Brill, K.G., Jr., and 0.9. Mutt 11, " Seismicity of the Colorado Lineament," Geology, v. II, pp/ 20-24,1983.p., Campbell, K.W., 'Near-Source AttenuaIton of Peak" Horizontal Acceleration," Bulletin of the Seismological Society of America, v. 71, pp. 2039-2070, 1981.

1. p M-4 Campbell, K.W., and Bozorgnia Y., "Near Source Attenuation of Peak Horizontal Acceleration From Worldwide Accelerograms Recorded From 1975 to 1993,"

Fifth U.S. National Conference on Earthquake Engineering, Chicago, IL, July 10-14, 1994. Cater, F.W., Jr., " Age ofl he Uncompahgre Uplift $and Unaweep Canyon, West-t Central Colorado," U.S. Geological Survey Professional Paper 550-C, pp. C86-C92, 1966. t. L y Cater, F.W., Jr., " Geology of the Salt Anticlines Region in Southwestern Colorado," U.S. Geological Survey Professional Paper 637, 1970. Ely, R.W., I.G. Wong, and P.-S. Chang, "Neotectonics of the Uncompahgre Uplift, Eastern Utah and Western Colorado," in Contributions to Colorado SaisaicityandTectonics,W.P.Rogersed.,JoloradoGeologicalSurvey, Denver, Colorado, pp. 75-92, 1986. Hunt, C.B., " Cenozoic Geology of the Colorado Plateau," U.S. Geological Survey Professional Paper 279, 1956. Hunt, C.B., " Geological History of the Colorado Plateau," U.S. Geological Survey Professional Paper 669-C,1969. Kirkham,R.M.,andRbers,!.P.,'EarthquakePote'ntialinColorado,a W Preliminary Ev ustion," in Colorado Geolooical Survey Bulletin. no. 43, Denver, Colorado, 1981. Kirkham, R.M., and Rogersh.!W.P., " Interpretation {of November 7,1882 Colorado L 11 { f.l f'l. r l' \\

• +^ h l}

V' s .s

[>" ?tyWr'.'4l a ' d/$/[ airs $3 ew, y.[ M ' c.u o h S.h'ffl!,, _ ' W$j '.' Earthquake," Colorado Geological Survey Open File Report 86-8, 1986. ? n ..c vu s.x1 F Larson,E.E.,M.Ozina,ahdW.'CJBradley',')Lde'CenozoicBasinVolcanismin T; Northwestern Colorado'and Its Implications Concerning Tectonism and the Origin of the Colorado' River. System," in Cenozoic History of the Rocky hountains, Geological Society of America Memoir 144, Boulder, 00, pp. 155-178, 1975.- T/; q. ' g%Jl.%.n.t.pt e, .., v..h. ..... ~. McGuire, R.K., A. Krust, 'a'Ad's.D'.foaks, "Th'e'Colo#rado earthquake of November 7, 1882: Size, epicehtral l'ocation? intensities, and pessible causative fault," The Mountain). Geologist,.v.19, pp.(11-23,1986. e..r. Smith, R.S.U., "Long Tem' Stability of Union' Carbide's Tailings Piles at ^ Uravan, Colorado." University lof Houston Giology Department, Houston, a TX, 1980. Y ' p gpW.a V u. af v, ...,9 \\;. Umstco Minerals Corporation, ' Calendar Year 1994' Environmental Monitoring / Data, Performance Evaluation Report,' ALARA Report," prepared by Umetco Minerals Corporation for Radioactive Materials License C0-660-02S,1995. ,. d: 4 ~ . Umetco Minerals Corporation and Peel. Environedtal Services "Hydrogeology of Club Mesa, Uravan, Colorado,i unpublished report, 1994. g w.,g_ _.. m ~ U.S. Department of Energy, "Remedia1' Action Plan and Site Design for Stabilization of the Naturita Title ! Residual Radioactive Materials at the Upper Burbank Repository Uravan,, Colorado,". November 1995. 7' '.M., -: 4.. = U.S. Nuclear Regulatory Comeission " Final'Standafd Review Plan for the Review and Remedial Action of Inactive Mill Tallings Sites under Title I of the Uranium Mill Tallings Radiation Control.ActV Revision 1," June 1993. .nsgi,w.Y r 1 c Wong, I.G., " Seismological evaluation of the'. Colorado Lineament in the Intenmountain Region," (abs.)@:.' Earthquake Notes, v. 53, pp. 33-34,1981. w y, i a Yeend, W.E., " Quaternary Geology of the' Grand and Battlement Mesas Area, Colorado," U.S. Geological Survey Professional Paper 617, 1969. 1.. m p t .t s 12 t ? h n i

~ ] 3.0 GEOTECHNICALSTABILITN dM ' b. ; O w d' 3.1 Introduction W . ",. 9. f 43 h 9 .c n E This section presents the rFsults of the NRC' staff review of the geotechnical engineering aspects of the broposed remedial actichs at the Naturita, Colorado, UMTRA Project site, as detailed in DOE's Final Draft Remedial Action i Plan (DOE, 1995a-f). A current Remedial. Action Inspection Plan (RAIP) was not l submitted for NRC review. The remedial action consists of the removal of all remaining contaminated mate'rjals from.the processing site to the Upper Burbank disposal cell 13 road miles: northwest of.,t1e Naturita processing site. 4. -,:.a..gt cg,%.a The disposal cell will bs below-grade and'will' provide for the segregation of the Title I material from other. radioactive material at Uravan. Contaminated material will be consolidated and encapsulated,in the cell and will be covered I by a 3-foot-thick compacted earth radon / infiltration barrier, a 5.5-foot-thick compacted earth frost barrieia, a 0.5-foot-thick bedding layer, and a 1-fo6t-thick rock riprap layer. The geotechnical engineering aspects reviewed include: (1) information related to the processing, disposal, and borrow sites; (2) materials associated with the remedial action, including the foundation and excavation materials, building debris, and other contaminated materials; and (3) design and construction details:related to the disposal site, disposal cell, and itstcover. The staff review of related geologic aspects such as stratigraphic, structural, geomorphic, and seismic I characterizations of the site is presented in Section 2.0 of this report. 3.2 Site and Material Characterization it 3.2.1 Site Descriptions 1) Q l 4 i A. Processing Site Q { 1( ) The processing site (Figure 1.X)f the town of Naturita which is off of State is located in Montrose County, Colorado, i about 2 elles to the northeast o Highway 141. The Naturita Processing site includes the following features:

1) the abandoned Naturita al11 yard; 2) the former tallings pile area which is located on the floodplain of the San Miguel River between Highway 141 to the west and the San Miguel River to the east; and 3) the former ore storage area l

located to the west of Highway 141. j During 1976-77, the tailings at the Naturita Processing site were transported for further processing to the Durita Facility heap leach plant. Although the l Naturita Processing site no longer has a tailings plie, the site has 400,000+ cubic yards (cy) of residual contamination which istdistributed approximately as follows: e Contaminated Soll (389,000 cy)8,300 cy), I. i Stockpiled demolitton debris ( Stockplied VP materials (3,000 cy) j .o n +3 1 i 1 0 ,l

.e. h,, t, r,. h V &hik,hi,&Y.f.. Stockpiled drums (55 gal) containing(processing waste petroleum products (72)y;s(approx 18cuftea)[g h y.' / Stockpiled ba with asbestos-containing materials (65)pgg l DOE reports that ' actual qiaantities'could vary from those tabulated above. M.

w %, ' m., 3 Olsposal Site at UravYn k 7i M,*.8! h jd % N B.

{g d m..Q n -.1 The Upper Burbank disposa1Esite a't'Uravan is^abc$+t 13 road miles northwest of the Naturita Processing S'ite (Figure 1.?). c The ' embankment will be located slightly to the south ofi' drainage divide, and :1t will be necessary to a accosanodate only minor anli>unts of offsite runon p.lus the runoff from the surface of the dis >osal celleThe disposal celigwill cover approximately 10 acres and will >e destined to contain approximately 500,000 cy of contaminated materials; hWever,' an alternate de' sign for an 800,000 cy cell hasalsobeenevaluatedby' DOE.7Theactualcel1%izewilldependonthe extent of contaminated mat'erials excavated during construction. The top slopes of the embankment will range from 2 to 4 percent, and the sideslopes will be SH:1V. The disposal cell configuration is shown in Figures XX and YY. The Upper Burbank Repository is locsted in the n% >hwestern sart of the Upper O 'E L ort Burbank Quarry, and is ent'irely underlain by sandstone and siale of the Salt Wash Member of the Morrison Formation. E 9 y i Borrow Materials Siteh) A C. h The borrow material for the radon barrier and frost protection layer will be obtained from the Upper C:1ub Mesa borrow site, which is approximately 1500 feet northwest of the Upper Burbank disposal site. N f. 3.2.2 Site Investigations: '4 d \\ i Geotechnicalinvestigation.andsitecharacterizationprogramswereperformed at the mill site, the disposal site, and the borrow site. Characterization of the borrow site was incomplete at the time this TER was written. Data obtained during the characterization programs were reported in Appendix 0 to the RAP. The scope of the geotechnical inve tigations included excavating test pits and drilling boreholes. Information from several monitor well installations was also utulized. Borings and test pits were logged by a fleid engineer or geologist. The locations of test pits, boreholes, and monitor wells were given in Appendix 0 of the RAP. Subsurface investigations for material properties of the underlying soll at the processing site were carried out in conjunction with the investigation to define the limits of contamination. The resulting samples of site l materials were tested and analyzed in the laboratory to evaluate the engineering characteristics of the materials. 4 Test pits were excavated $i th a backhoe. Bulk soil samples were collected t the drilling methods usedh Generally,gs provide detailed information ab from the pits. Individual, borehole lo hollow-st'n augers (6.5-inch) were used e 2 0 1' a ?. .) 2 ['l{ t' ,j i. =1

L f 3 -[M.. fKpgf [' &. y., ; N:.h,( y p V p, g until refusal; thereafter, La' rotary bif (d:4.0inche$)andcasingwereusedto bedrock. Three sampling methods were use

1) th'a Standard Penetration test (ASTM D-1586); 2) a 2.42-in'ch inside diameter, ring lined, split-barrel sampler; and 3) a 3.0-inch {iameter, thin-walled Shelby tube.

, a:m Y The available data obtainedTfrom the field inve' tiNations and iaboratory tests s were used to construct strat'igraphic sections,' and(to define the engineering parameters of the soils to tie incorporated into the cell. At the time of our review, DOE was in the proc'e'ss of further characteqizing soil materials proposedforuseasborrowgtheengineeredcover .s~ n 3.2.3 Upper Burbank Disposal Site Stratigraphy M. A e .,o ilis'nabedrockbeNchapproximately600 feet The Upper Burbank disposal 'cpThe bedrock beneath th's disposal cell consists of o above the San Miguel River.h sandstone of the Salt Wash Member. wThe bedrock atIthe site is reported to be stable, and the sandstones ht the contact between the Salt Wash Member and the Summerville Formation are r'eported to be dry; Additional information and a stratigraphic column are fo'und in the Section.2.3.2i \\ asygg y The staff has reviewed the bete 1s of the test pitijand borings as well as the scope of the overall geotechnical exploration program discussed in Section 3.2.2 above. The staff concludes that the geotechnical investigations conducted at the Naturita processing and Upper Burbank disposal sites adequately establish the stratigraphy and the soillconditions, that the explorations are in general;conformance with applic' ble provisions of a Chapter 2 of the NRC Standard Review Plan (SRP), and that they are adequate to ] support the assessment of geotechnical stability of the stabilized contaminated material in therdisposal cell. Investigation at the Upper Club Hesa borrow site was incomplete. 4 3 w 3.2.4 Testing Program Q N, M lhe materials at the three sites were classified according to the Unified Soil Classification system (ASTM D-2487). Atterberg limits (ASTM D-4318) and gradation tests (ASTM D-422) were performed on selected samples to classify the soils. In addition, the following tests were conducted: specific gravity (ASTM D-854), compaction (ASTM D-698), saturated and unsaturated hydraulic conductivity, consolidation (ASTM D-2435), shear strength (EM 1110-2-1906), radon barrier erodability (Crumb test, STP 623; dispersion, ASTM D-4221; and pinhole, STP 623), and erosion barrier durability. :The results of the individual tests completed to date are included in. Appendix 0 of the RAP. The testing program for the processing and disposal' sites was consistent with the needs of the proposed remedial action; representative samples of construction materials and samples of geotechnical materials that may affect or be affected by the remedial action were tested. The number of samples tested is considered sufficient to support the necessary geotechnical analyses described in subsequent sections. In particular, the testing approach is consistent with the NRC SRP/and the Technical Approach Document. Samples were tested in accordance with standard procedures. Quality assurance and quality control were performed in accordance with standard (UNTRA Project procedures. E q 3 2 { b p a .N . 3. 6

[

i. o Because testing of mater (als'from the Upper C1d$ Mesa is ongoing, the completeness of such testing is undetemined.E M 3.3 Geotechnical Enaine rina Eh15tfN h' W., ..,y. p d ii ! - 3.3.1 Slope Stability Eyaluation: W,3 %[ , fy, &. The staff has reviewed the exploratio~y n.An data ; characteristics, and methods of. analyses per teit' results,- critical slope tin'e'nt to the slope stability aspects of the remedial!attion plan for the Nat'urita UMTRA Project disposal / embankment. The analyzed cross-section with th vertical sideslope has theen compared with the e'y longest S horizontal to 1 xploration records and the design details.- The staff finds that the most' critical slope section has been considered for the stabil:ity analysis. p, Soil Jarameters for the ariousnihialsinth have men adequately est'abitshed by appropriate"lesting of representatives t material. Values of par'a' eters for other earthn materials have been assigned m i on the basis of data obt'a'ined from datapublishedinthelit^erature..geotechnical:.;explorationsatthesiteand 4 13 <., -.. 9 DOE has proposed applyiny a peik he'rizontafghoilnd acceleration (PHA) of 0.306g as discussed in Section 2.4.3. For a PHA!of approximately 0.3g, staff finds that the DOE evaluation has employed the n'ppropriate methods of i stability analysis (Bishop's Simplified Method,! Ordinary Fellenius Method. Janbu's Simplified Method, and Spencer's Nethod) and has addressed the likely i adverse conditions to wht'ch the slope may be subjected. However, as discussed further in Section 2.4.3 4a PHA significantly iniexcess of 0.3g may apply. If the PHA is detemined to)I xceed 0.3g, then morefsophisticated treatment (dynamic or deformation analysis) would be appropriate in order to confirm stability. The stabilit analysis remains an OPEN Is5UE until the PHA value is nsolved. .g 4; N, Factors of safety againstifailure of the slope for seismic loading conditions and static loading conditions were evaluated for both the short-term (end-of-construction) and long-term state on the basis of a PHA equal to 0.39 The values of the seismic coefficients used in the pseudo-static analysis are 0.20g for the long-tern condition and 0.15g for the short-term condition. The staff finds that the use of the pseudo-static method of analysis for seismic stability of the slopes is acceptable considering the flatness of the slopes and the conservatism in the soll parameter values, if the PHA is less than or equal to 0.39 The minimum factors of safety against failure of the slope were 2.2 and 1.2 for the'short-term static and pseudo-static conditions, respectively, compared to' The minimum factors of safety! required minimums of U3 and 1.0, respectively. against failure of the' slope were 3.6 and 1.7 for the long-term static and pseudo-static conditions, respectively, compared to accepted'ainimuss of 1.5 and 1.0, respectively. MA dynamic or deformation analysis based on a PHA J.n excess of 0.3g was not provided, so cell stability remains an OPEN ISSUE. In addition to the questb[n of acceptable PHA vMue, N dependent on the resultsjf ongoing characterization of 'the Club Mesa soils. h 4 y N hi IN (r 0 l 4 'e v i,

s; ( 5 f The analyses appear to h' ave been mad $ in a mansr consistent with Chapter 2 of the NRC SRP. Although the pseudo-static slopeIstability analyses are likely to be only marginally affected by using new soil parameters, the analyses remain an 0PDI ISSUE unt'll testing of. Upper Club Mesa soils is complete. If, t.wnw%. : Settlement and Coiler CrackingY.h. U :Mp5& Q 3.3.2 41. 9 The staff has reviewed the analysis of total a differential settlement of the disposal cell and foilndation materials and the resulting potential for cracking of the radon baffrier.. Calculations indicate that all settlement due to placement of the relo6ated contaminated mateElais, radon barrier, frost i prctection layer, and erosion protection will include immediate (elastic) and secondary (creep) components.c,The foundation is assumed to be incompressible, t because it will consist 76f competent, bedrock. Y ;.-. .Q : R.%::Q..< jp.f y Fifteen point locations tn section"A-A'. and 13 point locations on section B-B' were selected for settle' ent analysis for both 500,000 and 800,000 cu yd cover e options. The staff agrees that appropriate sections have been chosen to assess the most critica10 conditions for differential settlement. Calculated settlements along the pr' files varied from 3 inches to 9.5 inches o (section A-A') and 7.5 inches to 10.5 inches (s' ction B-B') for the 500.000 cy e cover, with resulting maximum local slopes of 0'.005 and.0075, respectively. Calculated settlements along the profiles variedifrom 4 inches to 21 inches (section A-A') and 3 inch's to 21 inches-800,000 cy cover, with resulting maximum local slopes (section S-B') for the e of of007 and.014, respectively. 1 A h The maximum tensile strain was determined to bei0.00054 (section A-A')'and 0 920 (section B-g')(set: tion B-B') for the for the 500,000 cy cover option, and 0.000068 (section A ) and 2.3 x 10' 800,000 cy cover option. The calculated tensile failure strain for the proposed radon barrier material (PI-30)was0.14percenty 1, jy; g. m.. DOE has concluded that total and differential settlement of the materials cceprising the proposed disposal cell will not have an adverse effect on the ability of the cell to meet the stability standards. The staff agrees that settlement generally will be small due to the compaction of the materials in the cell and the granular nature of much of the material, in addition, differential settlement should not cause pending concerns due to the sloping configuration of the cell, and cracking of the cover due to settlement should not occur because the resulting maximum strain is well below the calculated tensile failure strain. W - @n.l, N @ c. 7 um. h' t Despite the otherwise satisfactory nature of DOE's findings regarding the potential for settlementiand cover cracking,imunidry density) versusthe\\ discrep radon barrier compaction,(100 percent of max calculated values (95 percent of maximum dry density) discussed in Appendix G Construction Documents)d jan OPDI ISSUE.(Calculations) must be addr .and Appendix D The iscrepancy is considere bc ) 3.3.3 Liquefaction The staff has reviewed thi information presented!on the potential for s t 5 a ...g;N 'lj q, ., 0,f.

4 o

E If.

• s

[h... h.. k liquefaction at the site based on the'results.of geotechnical investigations, including boring and test'ptt logs,'. test data, soil profiles, and other information. The soils n the disposa1' cell wil',1 be compacted to a minimum of a9 90 percent of maximum St dard Proctor density (ASTM D-698) and will be in an unsaturated condition; t erefore the disposal calicis not considered susceptible to liquefactipn.;:.The groundwater ta)le.at the site is substantially below the bhse of the disposal embanknant. The foundation s L heneath the disposal celUis' stable bedrock undtthus not susceptible to liquefaction. Because ofithe absence of. water aind liquefiable soll, there is P no potential for liquefa,ction of material.withi or beneath the disposal cell and, therefore, applicabJe provisi.ons of Chapte 2 of the NRC SRP have been s - Q '. f y g g h y 'y Q >p. met. % *AbM ;O pf. 3.3.4 Cover Design The cover system will provide *a total.of 10 feetiof protection over the contaminated material andicollectively istdesigfed.to limit inflitration of precipitation, protect the pile froa erosion, aWd control the release of radon from the cell. Details of the staff review of the cover's performance related to erosion protection fehtures is presented in S' ction 4.0 of this TER; the e review of the cover's performance related to limiting infiltration are i addressed in Section 5.0; and the review'of the radon attenuation aspects of the cover is presented iM;Section 6.0.x However tthere are certain other aspects of the cover thatLare addressed herein.h'd n .3 TheRemedialActionPlanh(DOE,1995a)'indicatesjthattheradon/inflitration barrier will consist of compacted silty to clayey soil that will limit infiltration and inhibittradon emanation.$ The gradation requirements call for a minimum of 501ercent by weight passing the Noj 200 sieve. Testing has indicated t!4t t1e borrownsoll should generally: meet the requirements, and that inspection procedures will verify gradation? Test results indicate that radon / infiltration barrier material, when compacted to at least 95 percent of maximumdrydensity(ASTMD-698)T.willproduceajlaboratorysaturated permeability on the order of 10 cm/sec..% ' {

1 %.,d:;9. 1 The frost protection layer will corisist of materials excavated from the Club Mesa borrow area. DOE has evaluated the frost depth using the BERGGREN. BAS computer code developed at the U.S.. Army Corps of Engineers (COE, 1968). This code has been used for other UNTRA' sites. The total worst-case 200-year frost penetration depth at the disposal site is calculated to be 44.1 inches. The cover design provides for, the appropriate depth by the total thickness of 3

riprap (12 inches), bedding (6 inches), and frost protection layer (60 inches) above the radon barrier.i:The staff has reviewed'the input data used in i determining the total frost penetration depth, and concluded that these values are a reasonable representation of the extreme site conditions to be expected in a period of 200 yearsh Because DOE's calculstion was based on the 200-year I rather than the 1000-yearifrost depth, actual frost penetration is likely to be somewhat in excess of1the stated values. NRCistaff accepts this approach for the Naturita site because the additional frost penetration, if it were to f the cell. occur,wouldnotadverse(yaffectthestabilityj/ W l V ( 6 l n g i a F p r,,'y y 5 e l Nlkh Y %, l ,a

W H + h ".; te,,h,mepsm n) N} ; g[;g,N.s,, ybNl[ 'a h=- 3 < p; The RAP Indicates that the' layer'immediately abov/'the frost protection layer is to be a 6-inch-thick bedding / drain layer, intehded to drain water laterally + off the call and protect th's radon barrier from th'e riprap. Details of the review of the erosion prot'eption design,are,found in Section 4.0 of this 7 h report. A pg f y.. ; y The cover design has been & f..L. p r. [ m. : q pivaluated by'NRC staff for geotechnical long-term stability and for these aspets.the designfis accptable. e Geotechnical Construchon&i %p,,,Q D '.. ang '.W[ pk

c 3.4

' Details gjj W.. [ 3.4.1 ConstructionMethodsjandFeatures 4..' ) 3 r4 s w,;. g.j p .4 y The staff has reviewed an# evaluated the geotechni al construction criteria y provided in Appendices D and G to the RAP.1Desig calculations were typically y based on achieving 95 perceht of, standard compaction for the radon barrier 9 soils; however, specifications require 100 percent With this noted exception, the staff. concludes.that the. plan} compaction. s and drawings clearly convey the proposed remedial action design features. In addition, the excavation and placement methods,and specifications represent accepted standard practice. The discrepancy between'desigrilcalculations and construction requirements for compaction of. the radon barrier is an OPEN ISSUE. 13 d m ;' ~ 'I r sp.. Testing and InspectiSn w; M r %AI'I c ~.,g, ~ 3.4.2

. n',i y At the time of our review,$.,.the current Remedial Ad(tion Inspection Plan (RA had not been made available? For this' reason,'we'are unable to comment on the RAIP at this time. The lack of a RAIP is an OPEN! ISSUE.

[h . Q. k.. Mi The RAP and specificationsido not address the' problems associated with compacting a Fat CLAY (CH)isoll at 100 percent of tStandard Proctor maximum dry density. Because the CH spil must presumably be compacted wet of optimum, and desiccation cannot be tolerated,' additional discussion will be required. The satisfactory placement and'p'erformance of the radori barrier is an CPEN ISSUE. 3.5 conclusions 'jf f, ( Based on tha review of the ' design and he 'eotechriical' engineering aspects of the proposed remedial action as presented n the Naturita preliminary final RAP and sup)orting documents, NRC staff has not been provided with reasonable assurance t1at the long-ters stability. aspects of:the EPA standards will be met. Open issues which musti be satisfactorily addressed include the following: ll The acceptable PHA vbue will impact the stability analysis. For a PHA higher than 0.3g, mod sophisticated analyses will be required. For this reason, determination of a satisfactorRPHA and its relationship te slope stability remai,n, OPDI !$50ES. y 'q Although the pseudo-static slope stability snalyses are likely to be only marginally affecjed by using new soil girameters, the analyses $.1 7 h L o u c ' N_f & M

9 4 j,. t remain an OPEN !$$UE~until1testin i er Club Mesa soils is complete. Despite the satisfactory,.Mfp.8g,g of.'l'.b 's;.. w g%yf s.4h potential for settlement'and.' cover. DOE' nature of findings regarding the y cracking, the discrepancy between specified radon barrier compaction.(100 percent of maximum dry density) .versus calculated 'valuesl(95 percent of maximum dry density) must be addressed. The discrepancy:1s' considered an 0 PEN ISSUE. @. l. vwgf&sq;qwhyy.. The discrepancy betwe'en'desigif calculations:and construction fj-requirements for compaction of.the radon: barrier is an OPEN ISSUE. At the time or ou!?,;. m af.%MRPM{'l~r.:reviewv.EbRAIP hid riot been m For M;- this reason, we are unable to comment oni he~RAIP at this time. The [ lack of an RAIP iiQan OPEN ISSUE.jgf.l.c k sk - t The satisfactory niacement and pe;rNn.: e s -rmance of.the radon barrier is an .( OPEN ISSUE. ' }i3'. ~.4 4ggr,.c. ; 70 ~ 3.6 References 4 > &, y.y.g,.n COE (U.S. Anuy Corps of Engineers) Baltimore,',Haryland, " Digital Solutions of Modified Berggren Equation to Calculate Depths 'of Freeze or Thaw in Multilayered Systems," CRREL Special Report NoN122,.0ctober 1968. N. y 9 a,cd9 9,:, :, : dcA,.- DOE (U.S. Department of Energy),' Washington, D.C.'t " Remedial Action Plan and Site Design for Stabilization of the Inactive Uranium Mill Tallings at Naturita, Colorado, Remedial. Action Selection Report, Final," and Appendices A - G, 1 W a. jgj y " Uranium Mill Tallings Remedial. Action Pro' ject (UMTRAP), Naturita, Colorado - Information for Reviewers,". 1995b. e.tv ~ 9 ";myq/ A 2ji.. 6 ~ " Uranium Mill Ta111'ngs Remedial 'Actiori Project '(UMTRAP), Naturita, Colorado - Information for Bidder,s.)n>A Volumes (!-W.1993c. e'; R:.?;;.m Mp vfi --, " Uranium Mill Tallings Remedial ctionProject(UMTRAP),Naturita, Colorado - Design Calculations,&. Volumes I-V,1993.. 1 3m am -t. T MK-Ferguson, " Remedial Action'!nspect'io'n Plan',' h.aturita Uranium Mill Tailings Site," 1993. pgg'7 ll Lee, X.L., and Shen, A.',' " Horizontal ~ MovementslRelated to Subsidence," Journal of Soil Mechanics & Foundation Division','; Vol. 95, No SM-1, ASCE, January 1969. y, jj NRC (U.S. Nuclear Regula' tory Commission), Washlhton, D.C., " Final Standard Review Plan for the Reviey of Remedial Action o'f inactive Mill Tailings Sites under Title I of the Uranium Mill Tallings Radiation Control Act, Revision 1," Division of Low-Level Waste Management and Decomissioning, June 1993. b ls-

j q

p Li. - a

6. ' S ',

s u e si N h id I. i 4

t p jf'.' m.'df. .jc 4.0 SURFACE WATER HYDROLOGY AND EROSION'PROTEC. TION'* .. W:gky bwp:sjfgg,.l 4.1 Hydroloale Descrintion and Site Concentunt.r Destan

• DOE proposes to move b 'eYshi Nf1iW he ' town of Naturita, Colorado, to the UM":TC0 Uravan siYe.7,This site:is locat'e'd northeast of Naturita.

Small localized drainagebareas?r%WGb)d of.It 2.* Shy .n W -existNplan site and will contribute flood flows which must be diverted around the 21sposal cell. Several gullies L exist in the immediate ' site. area upstream'and downstream of the sito, t ts 4 MrMW%.Yf d In orde to comply with1 EPA standards which require stability of the tailings for 10W years to the ektent reasonably achiev'able and, in any case, for at least 200 years, DOE p ' poses to stabilize the ' contaminated materials in an engineered disposal ce e to protect them from. flooding.and erosion. The { rotection included the Probable l design basis events fo 3esign of the erosion p'ximus Flood (PMF) events, bot Maximum Precipitation (PMP)have low probabilities of occurrence during the

and the Probable Ma of which are considered 1to l

s I 1000-year stabilization period .g As proposed by DOE, the;tallings'will be consol'idated into a below-grade disposalcellwhichwi1UbeprotectedbyarockIcover. The rock cover will have a maximum slope ofi2-4% on the top slopesf'and 20% on the side slopes. The disposal cell will be surrounded by channel's which will safely convey flood runoff away free the cell.o In addition,t'an interceptor channel, north of the cell, will be con'structed to divert flood flows from the upland 3 drainage area away frc.m the disposal cell.rt o' I The upland drainage areas around the'ce11'C" J. p

/mh s haveIshort steep slopes and scattered steep gulliesh Some of these slopes!have competent rock exposed on the surface. These slopes and gullies will didharge flows directly into the diversion channels surrounding the cell and will; require the use of extensive rock aprons to protect against erosion.,. ;

d .. :n : n. 4.2 Floodina Determinations (. ']. ' The computation of peak flood discharges for various design featurer at the 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) Ctermination of appropriate rainfall distributions,icorresponding to the erLpted times of concentration. Input parameters were derived from each of u ese steps and were then used to deters!ne the peak flood discharges to be used in water surface profile modelling and in the final detemination of rock sizes for crosion protection. i'7,., ,,. y.... j fg y < 9.q- ~. 4.2.1. Selection of Design Rainfall EventEL y f er f. Oneofthemostdisrupt{3ve phenomena affecting;1ong-tem stability is surface water erosion. DOE hasdecognized that it is very important h select an appropriately conservative rainfall event on which to base the flood protection designs. DOEhasconcludedandthe.pRCstaffconcurs(NRC,1990) g@ . e. m y 1 a 7 .p 1 l n ih.n.k... f 9

i( 3.; .. g. :. y y,q. ! r that the selection ~ of a d ign flood e' vent should not be based on the

b extrapolation of listited hl'storical flood data, due to the unknown level of accuracy associated with sVeh extrapolations.a nn)er, DOE utilized the PMP, U,

which is computed by deterisinistic methods rathe has been defined as the suist severe reasonab y po[i th ~y and is based on site-specific hydrometeorol icaI characteristics. The PMP ssible raiafall event that u could occur as a result ofira combination of the sio'st severe meteorological . g. conditions occurring over alwatershed.9No'recurrynce interval is normally assigned to the PMP; however, DOE and RC (t@ hve concluded that the probability of such an even' t being equalW er odMed during the 1000-year stability period is small.N.Therefore, a M L usidered by NRC staff to ir provide an acceptable design basis.4 ( ylti%mb! a tw y e t Prior to determining the riino'ff from"the ' drainage asin, the flooding analysis requires the detere nationtof PMP amounts for theIspecific site location. Techniques for determining *.the PMP have been developed for the entire United States primarily by NOAA inithe form of hydrometebological reports for specific regions. These t' chniques are widely us(J and provide e s straightforward procedurestwith minimal variabil1G. The staff, therefore, 6 concludes that use of these reports to derive PMPlestimates is acceptable. A PMP rainfall depth of app]roximately 8.2 inches in 1 hour was used by DOE

• n.

O< . ~. 6 compute the PMF dischargestfor the small drainagefireas at the disposal site. This rainfall estimate wasideveloped by DOE usingtHydrometeorological Report (HMR) 49 (NOAA, 1977. value based on the pr)ocedures given in mR 49.The staff performed an independen Ba' sed on this check of the rainfall computations, the staff concludes that the PMP was acceptably derived for this site. yt q+;. - 6 3 y 4.2.2 Inflitration1.ossesp 1l g r. L 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 1sisaturated from previous rains, very little of the rainfall will inflitrate and most of it will become surface runoff. The loss rate is highly variable, depending on the vegetation and soll characteristics v the watershed. Typically, all runoff models incorporate a variable runoff - ficient or variable runoff rates. t Commonly-used models such as the U.S. e eau of Reclamation Rattaal Formula cufficient (C ; a C value of I represen(USBR,1977) incorporate a runoff Other models su)ch as the U.S. Army Corps of Engineers (ACE) Flood Hydr ts 100% runoff and no inflitration. Package HEC-1 separately compute infiltration losh s w1 thin a certain period of time to arrive at a runoff amount during that time period. n In computing the peak flow rate for the design of'.the rock riprap erosion protection at the proposedidtsposal site, DOE used!the Rational Fomula (USBR, 1977). In this formula, the runoff coefficient was assumed by DOE to be unity; that is, DOE assumed that no infiltration w9uld occur. Based on a review of the computations,Yty,e staff concludes that this is a very conservative assumption and'.is, therefore, acceptable. p e a h 2 h x ~ g n Y.

.xm a ;

x

~' p. ,b, [ hhU 41: m.c r.p:.0 * .n Times of Concentration'W.;j,%.F :)P.a Jsu7 i.E. h ? 4.2.3 0 St The time of concentration is' the amount of time re'qu' ired for runoff to reach the outlet of a drainage basin from the~most remote point in that basin. The peak runoff for a of concentration. given drainage basin is inversal ' proportional to the time If the t,ime of conce.ntration is peak discharge will be conservatively large."c Ties. computed to be small, the of concentration and/or lag times are typically comp'uted using empirical rplationships such as those developedbyFederalagencias(USSR.:1977).*..Velectty-basedapproachesare also used when accurate estTsates are needed..Such approaches rely on estimates of actual flow veTpcities to determine th's time of concentration of a drainage basin. /, g. 4 Various times of concentration foia thetirapde'sinnwereestimatedbyDOE considered by the staff to )e appropr)iate for estiusing the Kirpich Method (USBR y-based method is 3; p; ting times of concentration. Based on the.' precision and conservh 1sm associated with this method,bly derived.the staff concludes!that the times of concentration have been accepta The staff further concludes that the procedures used for computing the times of concentration are represent &tive of the small steep drainage areas present at the siteQ%hg$g.up M.e :;g t X i:A 4.2.4 Rainfall Distributions. i,c$ MM. y..;, g'y .s. a After the pMp is detemined2it is 'necessary to determine the rainfall intensities corresponding to shorter rainfall durations and times of concentration. A typica, PMP value is derived for periods of about I hour. If the time of concentratiehlis less than I hour, $t is necessary to extrapolate the data presented in the various'hydrometecrological reports to shorter time periods. DOE utilized a procedure recoemended in WR 49 and by the NRC staff (NRC,1990). :This procedure involveslthe determination of rainfall amounts as a percentage of the 1-hour PMP'hnd computes rainfall amounts and intensities for very short periods of time. DOE and NRC staff have concluded that this procedure,is conservative.n

v. p.,M Vq >< g.& Q in the determination of peak flood flows,' approximate PMP rainfall intensities were derived by DOE as follows:

g. $ y i. %e j mh,..,p9 Rainfall Intensity a. Rainfall Duration ./ . g; 3x(1, riches /R) (minutes) 2.5 ' f.i.. b54.0 5.0 9 744.0 &w.D: U d j [.pt4.0 15.0 60.0 a. *3;. l 8.2 [p

4. s. %

9 The staff checked the rainfall intensities for the:short durations associated with small drainage basins.VBased on a review of this aspect of the flooding determination, the staff concludes that the computed peak rainfall intensities are conservative. .M.,.y ? 7 [

k t

6' 3 V y lj fj.' d t. O ~, y ; L:. v a [$i.','.$Mo M lp /t. u

'~ ,y. w.., _ ... g. N. kd, 4.2.5 Computation of PC ;d,6. Q h)j.E ~ .( g 4.2.5.1 Top and $1de 51 opes 4 M y . Wi$ , w w, ,.or.4 7he PMF was estimated for thef top'and s' lopes us"ing the Rational Formula (USSR,1977), which provides"a stk.edard method for s'st sating flood discharges for small drainage areas. FWr a maximum top slope length of 600 feet, DOE estimated the peak flow rate'$o be about 0.64 cubletfeet per second per foot of width (cfs/ft). For a side storhlength of to fset and a flow concentration factor of 3. DOE rstistted the peak flott rate to be 1.1 cfs/ft. Even though other slope lengths and confleurations p'roduce lower peak discharges, DOE adopted the sort conservative flow ines (or deston purposes. Based on staff review of the' calculations, the esti tes are cons'dared to be acceptable.- ly. ' g r,. 4{ q, l 5 4.2.5.2 Aprons The side slopes of the cell discharge directly intoici.urston channels. Therefore, no aprons are reqstred downstream of the side slopes. However, aprons are required upstream of the cell and channels to prevent erosion. q DOE computed peak pW flow rates for the upstrees aprons using assumptions of sheet flow. Based on the topography immediately upstress of the cell, the staff concludes that several existing natural gullies may further concentrate flood flows and produce greater peak pMF flow rates than sheet flow. Because the shear stresses produced in a gully are Ilkely to~ exceed the flow depths produced by untform flow spreading across a plane surface, DOE should consider natural gullies in the analys'is. The design of.the; apron should be stallar to the design of the upstream apron at the Leuman site wtere various factors were addressed, including the externt of energy dissipation h velocities, and flow spreading, gggggg, ydraulle jumps, flow The staff does not agree thatSthe currently-propose ( prellsinary de:Ign is edequate. Several meetings and discussions have been held with 00E and its representatives, and it was determined that a revised design incorporatini the concepts discussed above would be subaltted. ; lased on this commitment 1:1e staff concludes that a satisfactory final design can1be provided. UntIla satisfactory design is provided and approved by the staff, this is considered Mr[ k to be an erDI IS8UE. y

s n 4.2.5.3 Diversion Channels

.y Diversten channels are provided to intercept and divert runoff from the upland drainage areas en all sides of the cell.1 Channels l',1 and 3 will be constructed in a horseshoe shape around the celle An Interceptor channel will be constructed north of the cell to intercept runoffsbefore its reaches the other diversten channels er the tell area,y@n. y% y,,.. :n.. fn the pW analysis, NCC-1 was used to esapete peak flew rates at different locations in the channels. Based on a check of thycalculatiens of drainsge the b me of concentrationIs,and rainfall intenlltyeithe stiff concludes that area ti k, estlestes are accep ble. W. y:.h o mn n( m...n t~ n m s

iy j-4.3 Water carface Proffhe and char:1 ValacfifeW Following the determinatfon of"thN ak"fNdthcharp. It is necessary to determine the resulting water. levels.. velocities',' and shear stresses ,e associated with that discharge.$ These parameters then provide the basis for the deterutnation of the required riprap size and layer thickness needed to ensure stability during the occurrence of.the de 1p event. 4.3.1 Top and $1de slope f } f ,g in determining riprep requirements for'the' top and side s1ons. 00E utilized the Safety Factors Methos? Stevens and others 1976) and tae Stephenson Method (Stephenson, Ig79)V'(resoectIvely.' The Infety Facters Method is for relatively flat slopee ef less than 10 eercevit the ste*enson Method is used for slopes greater th"an 10 percent. -The validity of tasse design approaches has been verified by the let staff thr at Celerade State University.m It was determinedi,ough the use of flismo tests that the selection of an sopropriate design precedure depends a the segnitude of the slope (Abt and stars,Igt7). The staff,' therefore,' concludes that the procedures and design approaches used by 00E areLacceptable and reflectistate-of-the-art methods for designing riprap ereston ph tection. g c % .p 4.3.2 Upstream Aprons I h DOC has not adeguately addressed the design of the upstream apron. 00E needs to consider the fellowing in the designs Q .r s.. ? em

1. Provide riprap of adeq,uate site to be stable'6 gainst concentrated flows

. associated with the design sters (PMp): w ' n p . r. o. mrt r 2. Provide unifers and/erleentle trades along the apron and the adjacent ground surface such that runoff late the elversion channels is distributed uniformly at a relatively les velocity, sintatting the potential for flow concentration and eros' lent and (fo ,, O L s n :m

3. Provideanadequateapronthicknesstopreventundercutting.

.,..... s u ~~ Staff has discussed these issues with 00t in meetings and telephone conversattens. As stated in Sectlen 4.t.l.t 00t has committed to providing a dest e chich incorporates these design concep.ts. However,IEIUE at this time the design of the upstress aprons is considered to be an OpDi deta led discussinn of the riprap design of the upstrees apron can be found in Sectlen4.4.1.1,below, p g;yq 4.3.3 Olversten Channels ,lgggy ( thcamputattenswere$sedtoestimatede The Act MEC t and norse; telecities ender the esti ed discharge conditlebs in the channels.pths and The Safety Facters Method was oped to determine riprap sites for the ditch. Based on staff review of the calculattenst the analyststis acceptable. Additional detailed infernstlen relatM to the design of eresten protection for the ditches may be feend in Sedtfen 4.4,_beleuep p 4 Q q, q Q 9;", p{] g 4 k.dhdbd, e

W 9 g A L ' nJ3 W W e '

p ry s

.g "L. -[" 4.4 tres1na Protactfog t-g( 4.4.1 Siting of Eresten ' tectlen [ i' N gww Riprap layers of verless stres and thicknesses are reposed for use at the site. The design of each layer is de

)

Riprap of the following types,.stres, pendent on its locatten and purpose.and la use at the site s p ne,pt ) 1 in (laches)ge5fte /. Type Avera r Thickness j,p r,gf;. (I s) p; g[ Type A Id Y lt Type B 5.t' It It.eji to { Type C 4 4.ce fj 4.4.1.1 Top and tide Slopes. -j L ;w: ,; r, m The rfprap on the top slope has ewith6andtheerosivevelocities resulting from an on-cell pfr as discussed above.900E proposes to use a I.0-feet-thick layer of Typef d rock with a minimum 1050 of 1.5 inches. The riprep will be placed on a O'.5-feet-thick bedding ayer. The safety facters Method was used to deterstne'the rock size. y.rf p gg-9,,,, n.2,.,., y,. The rock layer en the side slopes'is;also designed for an occurrence of the 'ecal pfr. 00E proposes to use a 1.0 feet thick layer of free I rock with a alnfmum Die of approxlestely 5.6 Inches.t The rock layer will be placed on a 0.5-feet-thick beddine layer. Stephenson's Method was used to deterstne the regelred rock size. Conservative values were usedifer the specific gravity of the rock, the rock angle ofi.laternalgp, tlen perestty. S una tr-Based on staff revlew of the 00t analyses and the acceptability of using l~ design methods recommended by the fulC staff, as discussed in settlen 4.3 of this report, the staff concludes that the proposedereck sites are adequate. 4.4.1.2 $ stress Aprent j 5f .[, Alprep stres for the aprons wee tempsted using in6dequate assumptlens of concentrated sheet flow and gully flows.' 00t aid not correct 1F evaluate the potential for high velocity flows free opstress drainage areas to tapact the apron. Staff has discussed this lesse ulth DOE intseveral meetings and telepheno conversations. As noted in Sectlen 4.3.t DOC has indicated that staff centerns are understood and that a redesign wl'11 be subaltted. Gr y s n'!a ,R jf. Mi pa.s t t L. -;,' 3 ff 4 b t;, as r W ljf

• Nkh, j ' [.

f e. $ W, Q[. ..,q n. ii X q .(~ , t s1 4.4.1.3DiverstenChannelg The deste of the diversten#chann'ls"was analyzed by DOE in the following e h I. desty of the side slopps'kMhh3M h ' entering the cha for concentrated flows f$hk deste for runoff directly three&&fkh.{ 2.

3. desty of channel evtlets: and a, j y, mghthechanne F

s9uw

.3 uj' sedimentconsiderattens'.%h M

4.

1. !hh-u: 4 4.4.1.3.1 Channel Side 51 s ;//mi

L e t,

/s ATypeCrfpraplayerwithhm.gatinisma 050 of !! Inchis is proposed for a 3 substantial length of the ditch.t.< The desten of tho' ditch si:le slopes considered the effects of pMF sheet flows directly;down the proposed side slopes free the spland drainage areas. Usteg the Stephenson Method for the IV on Nf ditch side slope, the required 050 was found.to be less than the size b y, M k i 9 I 4.4.1.3.2 Channels (Main 5setten) gf([ g,'{ [, the SafetyI acters Method was used to for flows directly threegh the channel F determine the rock sizes. Based on a review of the1 calculations the proposed Type C riprap layers are considered to be adequate.ti ?r +mn w% 4.4.1.3.3 Channel Outlets M[,. g.9 4.@ gl. r4 'a % D The channel outlets renen11y will be constructed 15 competent rock. Therefore, no tresten 3r headcutting is expected to occur. However the cwrrently proposed diversion channe outlet locatten near the Title,!! co'1 is not considered to be acceptable. Based on a site visit,ffect the toe of theit appears dischareing free the diversten channel could adversely a Title Il cell. DOC should either relocate the outlet of the channel er provide a revised desty to account for flows potentially impacting the Title !! cell. At this time, this is considered to be an OPEX ISSUE. 1.4.1.3.4 Sedleent Considerattens kTh. 'my[@,p ,, b y 4 s1sp of tbsediment depelliten can be e problue Ifdiversten ditches when the In M eeral diversten ditch is less than the slope'of th natural ground ehere flows enter the ditch.. It is esse 11y necessary to provide sufficient slope and capacity in the diversten ditch to flush or store any sedleents shlch will enter the ditch. ?In particular, slptftcant desten features may be necessary in areas where natoral gullies are Intercepted by the diversion quantitles of sedlaent, and the s' te of the particl(s transported by the ditch. Concentrated flows and high telecities cou1 transport large e nateral gully say be larger than the man made diversion ditch can effectively flesh set. q g 't y i. d ~ kV y 4.a.

G* e, ' ' p, y ?'} ' ~ l 5 (r,, j f : ;, e n. ((, s.*I h e For this site, a considerable amount'of s'edIsent Erom the upland drainage area can be expected to enter the diversion ditch, for'the following reasons: 4 c, i. r. q <.; P r 1. The apland drafnage arsas have st'eep slope channels have bun designed with relativel fwhereas the diversion flat slopes. Flow velocities In the ditches may notebe as high as those occurring on the natural ground. Therefore, sediment ~ cobbles, and b6ulders may be transported to the ditch and may not)be easily flushed out by the lower velocities in the ditth. q fgg 2. The_ potential for gully development (and restilting high flow velocitie in the spland drainayarea and subsequent transport of bed-load materia thte the diversion channels is hl h.. Based on review of topographic maps of the area and a staff site vist to the area pullies and areas of flow concentration are evl36at upstream of the divler,s on channels. Flows moving towards the dive'rsion channels will tend to concentrate in these stilles, increasing the potential for gully incision and transport of sedleent. p 3. because steep areas of; concentrated flow are resent a potenttal exists for large rocks and coobles to enter a channe. AddIttonal inforsation and ana yses are needed to document the adeqdacy of the channels, if 1srge amounts of rocks 7and so11 can be transported into the channels. np ., nv., u, <f. o In order to document the acceptability of the' channel desten Dot should demonstrate that: 1 the channels will have sediment car particularly if larg(er) rocks and cobbles can enter.a'channe apacitylal potent sediment sitten in a channel ul11 not significantly affe t e flow. cap *citys()hecentaalnatedtallings; erd (4)theriprapinthechannelsany blocka stability a t prevlees adeguate protection. j j,, g ^ DOE shes14 provide analyses which indicate'that the channels will be First able la flush out auch af the sediment, including larger eravel and cobblu. U6 top stors events rang ng.in segnitude free the annual flood to the pHF, DOE shou d calculate the er t tal shear stresses and velocities needed to transport materials of earlous stres, DOE should' determine that the slopes of the ctannels are sufficient to transport such of the deposited materials during most fioed events. o .e w-a second, Dot should estimate the esovnt of sedfeent' that will be depostted. DDE should deterstne that the channets will have adequate flow capacity, even if a significant amount of blochge occurs,' Third. 00t should estfaste.the asevnt of sedleent:which could build up th the channels ever a long period.ef time. Taking no credit for sediment reenval, DOE sheeld perform analyset:esteg MtC t end determine the affects of sediment builde, ions of Isrge flev blockages, Dot should determine that M flows will flow velocities and water surface profiles. Wder conurvative assimpt be safn1;/ conveyed. g ,q s o t H n 3 f sjkl w'., l fit c

pi&Qfl$$ h These issues have been discussed with'D0E on severalfoccasions.DOE has indicated that staff concerns were understood and would be addressed in revised submittals. tintil those analyses are provide'd, this is considered to be an OpDb !$3UE. g ,p.ga.g fn k# 4.4.2 Rock Durability b. . ). j EPA standards require that control of residual radioactive materials be effective for up to 1000 years, to the extent reasonably achievable, and. In any case, for at least 200 ydrs..The prevfous sections of this report examined the ability of the esMslon protection to withstand flooding events reasonably expected to occur in 1000 years. 'In this 'section, rock durability is considered to deterstne tfithere is reasonable assurance that the rock itself will survive and resal(effective for 1000 years. Rock durability la defined as(y. y nthe ability,of a matert$1 to withstand the m:a y forces of weatherterJ. Factors that affect rock durat111ty are:

1) cheefcal reactions with water 2)indblo,wn scour, 6) wetting and drying, and 7) freezin scourbysediments,h)w saturation time, 3) temperature of the water, 4) ard thawing.

U

.gf,,c...

00E fdentified saiveral swrces of rock in the fusediate site vicinity. The suitability of two rocks as 'a' protectlye cover was 'then assessed by laboratory tests to determine the physical characteristics. DOE conducted the tents ated esed tk results of,these tests to classify, the rock's quality and to Arsess the spected long-ters performance of the rbck. In accordance with past 00E rock-testing practice,! t se tests included j 1. Petrographic Exastnation (ASTM Ct95)d chemical properties. 2 Petrographic examination of rock is used to deteristne its physical an The examinatfon establishes if the rock contains cheefcally unstable minerals or volvestrically unstable!satorials... H n 3.y w. y.,- 9 2. fulk Speciffe travity (AIM Cit?)bility. peelfle' gravity of a rock is an ." The s Indicater of its strength or dura In eeneral the higher the s;4cificgravityis,thebetterthequalityofthero,ck, i 2. Absorptfen JA5 3 Cl!1). A low absorption is a dellrable property and Indicates s'ou disintegratten of the rock by salt action and mineral hydration. ggi 3 '.., 4. Sulfate Seerdness (Alm Cat. exposuretosaltwater,a'le)wpercentageisdestrable.In leestlens subject to freeting or 5. Schmidt Aebeend Hauser. This test measures the hardness of a rock and canbeusedineithertheyle1derthelaboratory. 4. Les Angeles Abrasten (Alm Cllt er Clll). This }j est is a measure of rock'sresistancetoabrafen. 1. Tensile Strength JAlm 03N7 er IllWI Method). Th.ls test is an Indirect test of a rock's tensile grength. I 9 j 4 Y p a> u i$ J 6 i

- (( 2.., p .a 1po .,l0,kq. 00E then used a step-by-st'ep procedure or evaluat,ing durability of the rock, in accordance with procedurps recommended by the NRC, staff (NRC,1990), as follow.t: y ' ... q:. 4.c ,v. g. vi 9.6 a. qa y s. h . p Step 1. Test results from representative" simple's are scored on a scale of 0 to 10. Results.cf 8 to 10 are considered 'eood" results of 5 to 8 are considered ' fair"; and results of 0'to $ are; considered " poor." n m,,,. w,qct.,u _ n w Step 2. The score is multiplied by a weightinifactor. The effect of the weighting factor is to focus the scoring on those tests that are j the most app 11'ible for the particular, rock type being tested. l c.y,. .1..,.. .. ;- - n Step 3. The weighted scores are totaled, determine the rating.' divided by score, and multiplied by 100 to y; ...xyA ~ y Step 4. The rock quality scores are then cosopred to the criteria which determines itsiacceptability, as deffped in the NRC scoring procedures. H . j In accordance with the proEedures suggested by staff DOE determined from preliminary testing that the rock proposed for the(disposal site scored above to. In accordance with staff recommendations,be 6f sufficient quality to mee DOEfwlli need to oversize this rock. The staff concludes)that the rock will EPA standards. ' 's

• N.5 a N i -

. %Bni'j':' 3 y a 4.4.3 Testing and Inspection of Erosten protection i DOE did not provide a M!p for staff review. Therefore, staff has not reviewed or evaluated the testing and inspection quality control requirements for the erosion protection' mater' als.? This information will need to be provided by DOE for the staff to conclude that thesproposed testing pr ram is Igtable. Untti the M!p.is provided, this is considered to be an op ac 0 ff. j ' D

1. $):v.7.h' 4.5 unstream can rallures O

m w u There are ne lepoundsents riear the site whose failure could potentially affect the site, y, wy ,g 4.6 Cagglgalang { iMy > t Jto. S... a Based on revlev of the infersstlen sabeltted by DOE RC staff concludes that thesitedesignwillnotmeetEPArequirementsasslatedin40CFR192with regard to flood design sensures and eresten protection. The staff concludes that an adequate hydraulle design has not been provided to reasonably assure stabl11ty of the contestnated material at the disposal site for a period of 1000 years, er in any case,Lat least 100 years. .F' m, ..a 3 e u At the present time, open is' sees include (1) deston of upstream aprons- (2) sedimentation in channels: 1(3) design of channel dulletst and (4) providing a MIP for staff review, h /; yx 0 si u L t0 a I ,? 4g. ; 4 n i.' y 3 !'b

45 9 'g 9.. ., *)*, o* fs. Ag'g . :&}' ' *..-{. ~

• l.a

'j y aj.' J t, y DOE has' agreed to redesign;a.y - v-. 4,., s .,.. 3.. -. m e. 4 ~ i' 2 ith prap aprons,:to"pfavide' addittonal sedimentation analyses-forithe, channels,'lto redesign or. relocate-the channel outlets, and to provide the'.RA!P;'! These'. revision will'be reviewed when z's. .substtted. .. 2.. e.;- 4 "z.r < W, ,j ' l l,i. 1 5 T.: ,N ot** ' F N' ',h i.l} ' ' I4r J ' 4 .. g,.;jf i .1, w $xa3 ,3 >g x g,. ': ~...W@.M*g g;* "y w..y-k.,... <9 se V. e. y Q': *y &;f..iq

• • T:i,, y :'&x?..

\\ ~ ?;. v ,.. :s.:., ; %JS;% g.9[ .T

• , '=. 'f Q_ t*Q

.A i p*.. .\\ ' z* L. u. 3B l* lf }* e

• i.

a q&yf,f 4~ i

1. (.. '{h[f:.

Y ,o c... s.._.,,.. ' ' {,i iN.,* ls .n&??.,l5 1 a ,l,... <..% ?c' y - , t,.f,, 'y., y. [..Y.Vg.} //' v[ $ 'i.t A,s. ),f '* ' f ?f Q/q,M., {l" ' '* "{$';.., 'd,Mk. I," * <. .3 ., :p u,',,i Nr [J,.Q ' c!J..,. .s hr ...f. ' B.,g[,[.!;.y* j7' l.,.< a...i > * %.y 4 ';Y... y)f/, ['s.- "f g

y..s.y m.-

., ' /y. - ' ! ' 'I, # ". (t-h,f,.*, .. ~) h ', i) ol:.

• . n.

'~

• * - j, y

,. ;g .(a 9 t.. a 'f ,nn, s.. ~;;Ofjp.. r 4 p:v. . --N.e . $Qp,jNQ a; m~.n,. p A:q u x q $)fi. S -. ~ 2,:,,,t,.Y,'.^*b. .A '.. -I '.., f < p",[h',%.'Es,*t ,I;4 ' ') t3 g.pI*,;o Y t.p. m..,

1. g

', 4 s'. U. Q' M M fh ',l 4" , . ',,,)R&, 'y','. f, i . n,. eq y,t e j1 4 4

c. e m6..,.

.y. ,i j g ?,;,. h y, -.a*.

]

v.2 y 'll f,,,, ' ' (f, 3f..,. 1 6, ti = L' n .t. l, '6

y n

t. / '... ;J !,i,l - c 1,.,e) s@? h.,3 b{ e y. ! kk. 7 M v['i. j.; ..k;6,"k.Q. .d.. ,y. .u ,i 0 "1

kk",, kh. 4,b.h,i. N -ky. <e,-' Nkl#'k ...q$4: .s. .n. z ~- :, ,A ..A w .,.;,;,.s., a:.sx .g ~ .y:. .p.., v.g y. Q.,. . (g .: >..;~;g y>y. prap aprons,lto p,rovide.i.,;;,.

v... t

,g n,. ,3 u.,.,, 3 4 DOE has' agreed to redesign.the, hannels,9te redesign cr. relocate the channe additional t : sedimentation analyses-for!.the c ~" g outlets, and to provide the;RAIP;g.hThese. revisions. will'be reviewed whe subaltted. ..,j.. g 4f "'V ci Jf r ':9: ivy J - [7,,, %l' ,, V ' hb l l;d" i

j

+ ..p,C, %.'t,, $. ' 2 't ((, Y 1.. g d" e, j,h. w e. .c ~ . ~ ;.'Q;1R'iM.y')CP ',t.&8 3 T W *t ?g s.y %@? '..

i

p

't.. . !,. 4.} Q

a$li

+ ?;. $ 3 .s y;h

~3;,,,;p f t.y 4

-~ pp.,4 u. l }l. iv

s. Q' [;f}', 5:%'4

-) N } f ('4 ?,. ?; 4 e 9 p. 7 -Q N, J,

• b t.g' :

i ,,,t 3,... g ( t "I g.g.,,'8."l g*;!d ' ',.j [ ' i, g./ $} ,J., 4a 5 1 $' n: a *'k

.). f, 'h

'5h-lk..h..W . < Mg.m s ., ?,e g,*,. ":[a.,.!.), f * *.. ' s,, g, d J k :. ' .k.'...#.-. k/h ~ 8 .2. ( '.' (d*'g.j',9f 9,,s,&l/.f.',,$ ff.. ' p,., i' f ..i ri,. '. .. < : Y';..*i'N.;A

• U

'i;r,,.

3. j.

...f. .g;.

y.

' '8es 't,,,'* ',ja i., f.y,4h' f /,J,fg,e.,' I f,' W' ,3 a 3 Y.. 1 s

• ',/*., h N $fhh l.

e'fE O '"; ^ y'f'f.qf' ,]...p$$},L. ? s. v., - i 1, t a;.e, C1,s e;%. 4

w..

y v. O. ,*1 s

. y4 y.T $($'i..f, ' 'M.

.1

g.i i p.,v

s

- lk,,; . jc - 3<,n TW!. f' _ ~ 5,' ,t ,, 8 / w.. - 1,.n -,3),vg g..<:> . h ',,, ib.: 'g J-[k,.s '$t,*r Y .A l (4 i ' /q* $s s. : s t. l'*v,] r *'.>

• yh p

h; ' ($. l y f y, y< : c.' k.t' T . e. -. g 4 . a.. g:. j) . g# d.! f". ',d.. !.:,Rf> l1;. g 1 's . t.. . > A:.

v..e

,\\ p a ' ')p..4.f .-l. u -3 t; l R, .7 l, ', 1 .f a N, *[$ g, < 1 3 i { '* !

• t,. Q,1. ' j *%..'.,'

) V s, ( o a i / 'h.- 4

1. w, l5 *v, 'l4Y, t e Ac..$.'.0, h N j

.w e e 1 J l

?Q~V..

1, s 5$p.692W@*{y.eh:n.,;g2

' WW.TGWb'. ' 4 ' 'hfi,W T' $.E sf ~ y;...a.. v: .. e g. .i .d ' ' ~

.; i ' ~7 ga:.f

-g.i,D h .s?.

• ?.

fif. N, s'k'li. I, ,Q prap aprons,lto'p,rovidel*addltional 00E has' agreed to redesign th i..- t sedimentation analyses-for outlets, and to provide thEthe, channels.'7.to redesign or,' relocate the channel RA!P;.hThese!. visions 111 be reviewed when. 4 submitted. '.j - ~ r 'i$, M i' ' '4 i .. k. d W T' ~ W A%[:,4' 'y 4 m.p:i! a.'y,,:k((.- y , S. ' '.. ;.;jEg

e

'N, ' .g { _ ['Y jg,'[lf: y- ..,b ' ' EY^ , $) . :... e.a. ... h,.Q .s ,, AA.N.e ' , O,L,, w' b' E., ,;N'w au 4 ... m. $.*,..,; .,+ r '* *'),p,) v. '3 1 [, '.', ^ _y .M ,d.. g g. ez h .,my& -Q(;4;, ., ) k E[; g y.. .s'r : r ( >., 2 .a 21k.' n ' ,',I i,,k, {. - l,'$ M %((-)fe,: :...:f 3.- .s.) ):p a r, ,p '( , Mk * ' g 1, 4, .k' '.*&n. '. ~ Vl ,,,, ;(y 'rli[. % .h ?),j:, . g p.c'...U..l%*J. i'{f '- 9 l$,.;.. " -l.yt n). . w p, g);: t nA

1 v.,,,,'P n,e.

r. -

.' c,h *./.l:. N. N.s'. & (..). :{c'.l.'#h':, . a. ';j '

.. n.

u s., n , e 's

. ')

i \\' .=l.' x, .(([,. i .,.",JFl,. g 1' l l = y $ Mt,

., ~.4 p,.h.ty Q[,. '

'l . )k q % f j. w v ? p p. 1. .~ ~ $p .f'O 9',9 g,p..ji ",- '. 7.e., 4 .. qe.

z'. -

" 't yQ[f h. A ',- r. bEb # .,. % $s .,,. ~ y.t,n,, p../i '* ;;,. ?

n..,

- 3 .w. , y 3. } i+ If % (*$ ' '. j'ig ' d . '.u.g ;f h,* f/5'T [I s .- J. g*.j.c:. (? a r,; yv,. q./ + Yr e t v ,,/a. 9 c E 'O n

,s. f...

....'

i,,, d. s. .w.

0. ~

s. Y'. s ( b,.,2N, .'Z ' i! ' 4' lq it nW j ,+. w F. t N?[ 5.4 f 3

1 .g 6.0 RADON ATTENUATION AND' SITE' CLEANUP % +.' H f 6.1 Introduction This section of the TER documents the suff.evalifatt'on of the radon ^ attenuation design for theidisposal. cell prepared Corporation (DOE,1995) aid the processing-site ci[by Umetco Mineralennup plan foi action at the Naturita, Col'orado. UMTRA Pro, ject s' te, described in the DOE Remedial Action Plan (RAP)' tion includes review of the material characterization,h(DOEOl994) A The eval adon barrier design,. Its radiological characterization, proposed remedial ~ actions, and ' site cleanup verification plan to ensure compliance w'ith the ~ applicable EPA

• standards. The review followed the NRC Standard ' Review Plan.for UNTRCA itle I sites (NRC, 1993).

+- l

n..4.w g

q .,y %gr[.hM. p C $$; J 6.2 Radon Attenuation

,e f:n

a. ? -

j To meet the EPA standards 'for long-tem'contiol of radiation and limiting release of radon from resi' dual radioactive materials to the atmosphere, the contaminated material will' be relocated to the Upper Burbank Cell, in Uravan, Colorado, and covered withithe following layers: 03-foot radon barrier, 5.5-foot frost protection, 6-in'ch bedding, and 12-inch riprap. The radon barrier radon from the disposal cell to meet the EPA flux standard of 20 pCi/m]/s andlayer to attain released radon levels as: low'as reasonably achievable (ALARA). 5 . " Mgy %; A $ Because radon (Rn-222) is n'.g:s with a short half;11fe (3.8 days), the amount of radon from uranium all15ta111aos reaching the steosphere is reduced by restricting the gas movement long'e augh so that 'adon decays to a solid r daughter which remains within the dis wn! cell. (The physical and radiological parameters influencing ito amount of: radon available to the soll pore spaces and its movement are incorporated into.a computer code. The staff evaluated the estimation oftthe long-term (at least 200 years) average (over the entire cell surface, over at least 1 year) radon flux from the disposal cell cover by utilizing the RADON computer code. rMost of the code input values are parameter values, derived during characterization of the various materials that will make up the cell and/or based on conservative estimates. The combination of input parameter values and underlying assumptions comprise the radon flux model. Staff also reviewed the construction specifications for materials, placement for consistency with the radon flux model assumptions (material sequence and compaction) and evaluated the other layers of the cover for their ability to protect the radon barrier layer from drying and biointrusion, p.,

3 3

An additional aspect of the staff review considered that the radon barrier layer is also designed to satisfy criteria for construction and infiltration of surface water. In addition, the potential for! cracking of the barrier layerduetosettlementor'heavingwithinthecellqandthepotentialfor freete thaw and erosional damage to the cover wereievaluated. These aspects of.the cell design are evaluated in sections 3 and14 of this TER. e .a Evaluation of Parameter ValuesW, y3..vd s,2.1 2,. 9 The staff's review addressed the adequacy of the parameter values (i.e., code input)byevaluatingthejustificationorassumptionsmadeforeachvalueto 9' ,i tl , m;, ' e thh,y:.h b o

s J g 5: confine that each value waifrepresen'tative'of the material or a conservative estimate, consistent with site construction specifications, and based on long-tern conditions. Design patameters of.the contaminated and barrier materials that were evaluated includh. material placement s'equence, layer thickness, bulk density, specific granit radon diffusion coefficient $ y, porosity..long-terin moisture content, and In addition l.the rad' lum concentration and radon emanation fraction of the co'ntaminated materials we're evaluated. The sampling nd testing methods for the)' materials were also re91ewed to determine their appropriateness and to ensurs.that the.-data',were a'dequate. A. Contaminated Materials Wtg pql< T@39.,. E + 4 ~ The Naturita processing sit'Q. W'M.he no longer' as a'k N n tailih s site contain residual contafa'ination.s These areas i a:plie, but areas of the ..~ H N. S h.. %. t! 1. former tallings pile area >'A N alli yard and ore buyin'g' station (27. acres, attthe lowest elevation) 2. (14 acres, onihigher terrace) 3. cre storage area B ' '(12' acres, we'st of Highway 141) windblown / waterborne materialsf (196. acres),0;. 4, N vi m. Umetco estimated the volumejof contaminated material to be 548,000 cubic yards (cy), including approximately 8000 cy of processing site debris to be placed in the disposal ce110': If. the application of supplemental standards (no remedial action) proposed by'00E is approved, the volume of disposed material would be reduced by approximately 147,000 cy of mainly windblown material. The proposed application of supplemental standardstis discussed in i Section 6.3.2 and the impact of.its approval and implementation on the radon t barrier analysis is discussed in Section 6.2.2.0 P, The disposal cell design indicates that'm %:. N O m ~ contaminated material from the alli yard Cincludes retention basin and ditch soils, vicinity property materials, demolttion debris, and are buying station) and oreistorage area will be placed at the bottom of the cell and covered by soll,from;the former tailings pile area. Windblown material will be pieced last (Umeteo Figure 1.6). The windblown material comprises'approximately 39 percent of the material to be disposed, if ths supplemental standards applicatien is approved, and 55 percent, if the a,ao11 cation is not approved (DOE > calculation 17-737-01-02). m,o.w fb.e. a S Umstco derived radon model parameter ~ values'for 'he contaminated material from DCE rap calculation 17-741-02.- Most of these values were determined from laboratory testing performed on material compacted Proctor), as the material it to be placed in the ce;to 90 percent (Standard ll at this level of compaction. Staff is concerned that some of the contaminated material parameter values are based on limited testing or non-conservative estimates. However, staff's concern is mitigated by the proposed thickness of the radon C barrier and frost protection layers of the cell cover, so that the contaminated material parameter values are not an l'ssue [see TER 4 For documentation purposes, the ey#aluatLon of each parameter Section6.2.2), value follows. O

d. 4,

1. 9 vz g

Avoragemaximumdrydensitylandspeelficcravitymiasurementsforeach mater < al type was used to calculate porosity. Theqvalues for mill yard and g. u,. y 2 6.p l(,g 7 [ ) c' + $w$a hl k ' m.e ~, e .u

m a ~ ore storage area materialsfw.;:23%d% w.a & materials will be combined'during placemen(volume' weighted) because these ere combined' t:in-the cell.. The resulting parameter values (below) a acceptabletostaff.p?g Nts Material Densit/ # T ts C Porosity Nb Nhh "." alli yard / '2.69h .40 4311 M ore storage 1.60 . 2.63 d,' 10 tallings area 1.61 9 74 ~ 7 .39 .42 .y!!FMh !2.58 4 '6 windblown 1.51 qy tyy .n. f The long-ters moisture content parameter v%: alue of teach material was based on j the average of minut 15-bar# cap 111ary moisture test results. The tailings i J pile area (five samples)10;8 percent, lone ore stotage sample yieldedaveraged natorial (five samples)lownimaterialf 11.1 percent, and windb four. samples) averaged 13.0 percent. ~ The high moisture value fodwindblown(solliis cons'idered reasonable because of l the high clay content 3 samples averaged 26 percent) of the material. To be conservative, staff use(d a* windblown material moisture value of 9.0 percent in its radon model. 6 e:g g g.;j Umetco indicated that a sin'le soil sample ea'm:ch'fr'on the mill yard, ore .,s r g storage, and windblown contaminated material.was m'asured for radon diffusion e coefficient. Each sample was tested at five diffe' rant moisture contents and a best fit curve prepared. Thediffusioncoefficienfvaluecorrespondingtothe diffusion coefficient of 0.0188'/s was derived forathe wind material was selected. An average radon cu' layer in the cell and 0.0136 cu A value of 0.025 cu'/s was estimated for the tallings. pile area. Because of the limited data, staff used the more conservative code-calculated diffusion coefficient value for windblown material.(0.031 cs'/s Umstco stated that 12 radon; emanation fraction m.1% ~) h A:n,mut9filW: easurements on contaminated material ranged from 0.06 to 0.35 4The volume weighted average emanation fraction for the mill yard / ore storage soils (fouri samples each) was 0.33, and for windblown material No tests were performed on former pile (four samples) the value w.s 0.22. area soll, so the alli ya assumed to be representative of this material.MThese values appear reasonable, but staff utilized more conservative values in its model. . wwx& : % DOE determined the Ra-226 content of the'contiminated materials primarily by gasma spectroscopy. Incremental Ra-226 depth profiles were constructed for each sensurement grid point. The averafie Ra-226 concentration was determined for each subarca by integrating the profiles over the volume, based on the excavation depth. Based on.these data, volume-weighted average Ra-226 concentrations were calculated for each, layer in the disposal cell as follows: k. ?I.h@d.'@ 'sk .y ff. [., oh M E h. 0 h

1. f lb

%m et to swant nomiten( if;,.' { $[ 'J h Q.

i:

&t z ___6

@y ' '+ >,r ~ '. :,"?.6 ~ it Arla Ra-226 (oci/01 i Samoles ' Yblume (x1000 cv) Thickness (cm) , Mill Yard / debris / ) I78 'VYh Ore Storage Q133 W.,y J27;j'% q }. a - 134.6 119*, 122 ty90',.C['281.b. 56 j M Y 116.1 113 tallings pile ' (* if supplemental standards applied)T@.y L 38 h( .M.y 294.1 171*, 402 Windblown a p, b. Nt ... dW. W - ] ' A value of 59.5 pC1/g 'w,hhher Ral I Umetco chose to model 226 values th n the average measured values. as used for windblown uterial and 94.1 pCi/g for former tallings pile so}1. This approach ;is conservative and, therefore, is acceptable to staff. ". i 9 #n..y r 1., 'g ?. M le g ZQ,k

j' S.

y g B. Radon Barrier b .%y Q

• j The parameter values fd thiradon" barrier maNrlal were selected by Umetco 1

based on the results of: laboratory testing ofisamples from the Club Mesa borrow site. Umetco indicated that additional / testing is underway and the results will be available in January 1996. .0 y s.%., w~; p. Q. i. The summary table of geotechnica% ign parameters in the Umeteo RAP ldes { calculation 010 (sheet!8) indicates that radon l barrier material has an average of KO percent fines. However, the construction specifications do not contain requirement for a mih'imum percent fines (material passing the No. 200 sieve) for radon barrier soi1Nalthough calculation 006 (sheet 107) indicates that theradonbarriersoillwillhaveaminimumof.50percentfines. In addition, the actual material te'sted contained an average of 86 percent and a minimum of 80 percent fines (calculation 006 sheet.108).uStaff is concerned that radon barrier soil placed in!the cell may be differeht in character than either the design or the tested material on which the radon flux model is based. This construction issue is discussed in Fection 3. y \\ , O. c we g Staff notes that the barrier material Vas' tested at 95 percent compaction but will be placed at 100 percent according to the design and construction specifications. The moisture test value derived from material at the lower compaction may not be conservative for the radon flux model moisture parameter because the looser soil would hold more water.'1 Therefore, staff modified the value for this parameter in its flux model., A The density and porosity parameterTvalues' for.W:.

. W %.p.q 1

.the radon barrier material appear to be based on th6 average measurements of two samples (calculation 010 sheet 6). However, the Umeteo RAP indicates that 10 samples were tested, although the summary o(Section 6.3.4) lts (calculation 006 sheet f soll test'resu ) 100) indicates that 5 samples tested for maximum density averaged 1.60 g/cc (99.64 pef) and 4 samples tested for specific. gravity averaged 2.70, resulting tn a calculated porosity of 0.41.ZCalculation' 010 sheet 6 states that the density value for design is 1.50 g/ce, but the:Umeteo radon flux analysis uses the less conservative 1.60 g/cc value. Staffs. utilized the conservative values of 1.5 g/cc and 0.44 for density and porosity 3in its model. a .p The long-t'erm moisture content parameter value was based on two minus-15-bar capillary moisture tests that averaged 19.4 percent. At this moisture value. M ^f,: 4 a .n t {

• l'

..( 3 ..,s. f ~

l M ?.-

• g.

,i e m g. s Arn Ra-226'(oC1/0) 1 S anles M Yblume (x1000 cy) Thickness (cm) Ihhhf8 '9 . Mill Yard / debris /

1. 133 W 5 27;y; Q.

119*, 122 Ore Storage J 90 4 d,? 281.} [M ' M 134.6 b 56 M 116.1 113 tailings pile ~ 1 38 i.

h. Y Windblown 294.1 171*, 402 a

(* if supplemental st.ardsapplied) Umetco chose to model h he,r,.n..ie.&.. W"];ic A W l A value of 59.5 pC1/g 'Vas us. Ral 226 values thhn the average measured values. ed for windblown n terial and 94.1 pCi/g for former tallings pile so)1. This, approach;is conservative and, therefore, is acceptable to staff. uf,. .y > g3.;g7, jp \\ q paw p,g y.L gh,',y. B. Radon Barrier V Mi ,h y W.y; ,g The parameter value's (Er the radon"t, barrier material were selected by Umetco L based on the results of' laboratory. testing of tsamples from the Club Mesa I borrow site. Umetco indicated that additionalftesting is underway and the results will be availab'le in January 1996. .f y. %.3. y.: w p C. f.e \\ The summary table of geotechnical design parameters in the Umeteo RAP J { calculation 010 (sheet 8) indicates that radonjbarrier material has an average i of 50 percent fines. However, the construction specifications do not contain o a requirement for a mih'imum percent fines (material passing the No. 200 sieve) for radon barrier sot 14although calculation 006 (sheet 107) indicates that the radon barrier soil;will have~a minimum of,50 percent fines. l In addition, the actual material te'sted contained an average of 86 percent and a minimum of 80 percent fines (calculation 006 sheet.108).gStaff is concerned that radon barrier soil placed intthe cell may be different in character than either the design or the tested material on which the rad 6n flux model is based. This l construction issue is discussed in Section 3. y .y2.,c. wy e Staff notes that the bar;rier material Vas'testid at 95 percent compaction but s will be placed at 100 percent according to the design and construction specifications. The moisture test value derived from material at the lower j compaction may not be conservative for the radon flux model moisture parameter because the looser soil would hold more water.V Therefore, staff modified the [ value for this parameter in its flux model.4'..WO. i ,, M%% tuq The density and porosity parameter.' values'for.the radon barrier material a) pear to be based on the average measurements <of two samples (calculation 010 siest6). However, the Umetco RAP indicates that 10 samples were tested, although the summary o(Section 6.3.4) lts (calculation 006 sheet f soil test'resu indicates that 5 samples tested for maximum density averaged 1.60 g/cc 108)64 pcf) and 4 samples tested for specific, gravity averaged 2.70, resulting (99. tn a calculated porosity of 0.41C Calculation' 010 sheet 6 states that the density value for design is 1.50 g/ce,'but the;Umeteo radon flux analysis uses the less conserystive 1.60 g/cc value. Staff. utilized the conservative values of 1.5 g/cc and 0.44 for density and porosity 3 n its model. 1 a + The long-t' era moisture content parater value was based on two minus-15-bar i capillary moisture test.s that averaged 19.4 percent. At this moisture value, U

l

c l 4 a

r

.n t d

V

. i f, y c.. a radon diffusion coefff tent of 0.0063 cu'/s w"s derived. However, Umetco chose to use a code-calc'iilated.value of 0.0021 m /s'in the flux analysis, NW which is not tonservativ'e'. Staff normally woul use the Rawls and Brakensiek y'; equation to actinate the content of the barrier s'hlong-term moisture con [ tent, but data on the organic oll were not available.f Staff is concerned that the 7, material tested had a high fines content which liay produce unreliable or unrepresentative capillary moisture test resulti.. Therefore, staff used a moisture value of 16 percent,-~ deemed reasonableifor'/s th clayey soil, and a calculated diffusion coefficient.value of.0.0097cca its model. The thickness of the radin barrier layer is setpbyShe' L%etco design at 3 feet A x % D dp W 5t.c h - J to satisfy ALARA considerations. 4 This is~acceWable to NRC staff because this I thickness provides reasonable assurance that,t f long-term radon flux standard canbemet,asdiscussed1below.wM e Evaluation of Rad @on Attenuatlo Mode 1N ;&' wak ,3. 6.2.2 n .q.m u. y y y Umetco provided two radre flux models ~and util1[m! the RADON computer code to evaluate the radon attenhation capacity of the b'arrier. One model assumed that the upper 16 feet windblown material, and(500 cm) of contaminatedlmaterial was composed of the other assumed this Jayer was composed of material from the former tailings of the cell have the same; pile area.4 Both modelsiassume that the side slopes i layer sequence and thickness. To be conservative, neither model considers the radon attenuation ablility of the frost protection r layer. Each model was run first to estimate th(' average long-ters radon flux froma3-foot-thickbarrierandsecondtpdeterminetherequiredbarrier thickness to achieved a flux of 20 pC1/m su Thef resulting values are: 2g.#. uffedB Flur'Inti/m,3 g g,g.y.p"aYrier Thickness contaminated Laver $ b nches (6 cm) Windblown Material 1.3 NY [4 .q % h" %.4 WyQ, 8.5.g 38{91 s. c. inches (21 cm) Tailings Pile Area M1,1 Staff modeled the more conservative ~ parameter va;y lues discussed in previous sections and used a conservative, but. realistic: layer sequence and thickness. ( The layers modeled were: n!) 4.5 feet'(138 cm) mill yard / ore storage i material; 2) 3.5 feet (106 cm) tallings pile area soil; 3) 4 feet (122 cm) windblown material; and ) 3 feet of radon barrier. This model reflects a minimal amount of materi(al with low-levels'of Ra-226 (assumes th 7 supplemental standards are approved) and, therefore, there is a higher total l Ra-226 concentration than'is expected to be present. resulted in an estimated.long-term radon flux of-!!.8 pCi/m}/s.The s aff's modeling i. This provides reasonable assurance thatithe Umeteo radon barrier design meets the EPA radon u 4 flux standard. y ' p. Wp... g$ 3.Q. n. 6.2.3Durabilityofthe}adonBarrla,' gig ij, One aspect of barrier dutability that can'be eviluated by radon flux modeling is freeze-thaw damage. Umstco calculated that the frost penetration into the cover would extend 44 inches (conservatively modeling the frost protection layer at its long-term opsture content). With{heproposed66-inch-thick y

1. s a y

f $h h

l.k

,h,. l,'.,_

.t , b.v.s h'3.e LQ s .. b.. ...yn 'i hh Y frost protection layer a$d the'18-inch-thick NIdtn ff' radon barrier soll would' Hot be affected by freeia g and riprap cover, the i thaw events (see further t discussion in TER SectiorM3.3.4).9Therefore, flux modeling with parameter b values altered to reflectCa frost damaged radon barrier was not necessary. W Evaluation of the potenti'ai for cracking of the tr'adon barrier due to desiccation or cell insta'bility is addressed.in 'Section 3.4. W vy:Mm:im y es,. Another aspect of the evaluation'of the long-term integrity of the radon i; barrier is estimating thdplikelihood of intrusich by burrowing animals or

q$

deep-rooted plants. Umeteo did not address this aspect of cell design, but staff considers that biolytrusion of the radon b[arder will be restricted by +; the unfavorable environment of the rock layer in}the final cover. Although it' M'r is recognized that some volunteer plant growth Will occur on the cover, f,,r[ significant root penetrat'Ibn 7 feet deep to the j'adon barrier, is not p anticipated. ff.;, g/g'.4p -${4 g,p g w e - j,? Site Cleanun .y . ' u., g, 3 A .s m 6.3.1 Radiological Site Charac.terization T O:. ,i. 4 9 3 g Field sampling and radiological surveys at the Naturita site identified contaminated materials covering 133 acres at the! Processing site and adjacent areas. In Section 6.5.1 of the DOE Remedial Action Selection (RAS) Report, thediscussionofsitecharacterizationmentions(thatdatatodate(March 1994) have not indicated af potential for preferential mobiltration of Th-230 at the site. Staff considers that characterization of soil Th-230 and U-238, as described in the RAP, is inadequate. NRC staff estimated from the coordinates given in Tablet B-3 of Calculation 171730-01-01 that only one i sample each from the former tallings pile and ore storage areas were analyzed n for Th-230. However, DOE" states'(RAS Section 6.5.1) that further characterization of Th-23p;and uranium will be performed in conjunction with test pitting of the cobblytsoil. DOE should provide the additional Th-230 and U-238 data obtained with the cobbly soil study ti substantiate that adequate characterization of these radionuclides has been(performed. This is an OPEN c ISSUE. / G ^ Soll background levels of3Ra-226 were measured in: the Naturita area and DOE stated that the average value is 2.3 pCi/g. Thelvalue is based on four samples ranging from 1.1 to 3.4 pC1/g. This limited sampling is not i acceptable to NRC staff because the soll cleanup 1 criteria include this background value. In addition, RAS Table 6.1 indicates that the range of Ra-226 background values was 1.8-2.0 pCi/g whichicould not result in the L stated average value of 2.3 pC1/g. DOE should provide more soll background Ra-226 data and correct the soil Ra-226 data in RAS Table 6.1. This is an OPEN ISSUE. d ", e: 2 u Cleanup Standards [.0,}.l rWf n

j. yl%g,: j

.') 6.3.2 .s .m o.n.., i U DOEcommittedtoexcavatefcontaminatedsolltomeettheEPAstandardof 5pC1/g(surface)and15pC1/g(subsurface)'plus[backgroundforRa-226insoil and to piace the contamina ed materials in an engineered disposal cell. Excavationwillbemonitor)edtoensurethatcleanupeffortsarecomplete. The p ..c 2 !g g.*' { s. en. y

, \\

f Yl &$$ h &&%ihML f y -a,

y w,, e .pyw .1W&$ l'R!.' alli yard and begin site exta.the"debr s and icinity property material in the cleanup plan is to stockpil's vation at.the higher elevations. r A m 3.34pw-wwA i, a I .O All buildings on the site will-be demolished,' and all contaminated debris will be disposed of in the disposal cell.*,Therefore, deanup of buildings is not E hh hfhh. \\ 00Estatedthatsubsollcodditionsinthetailings,'fpileareagenerallyconsist of a high percentage of cotibles and gravels greater that a number 4 sieve. Therefore, DOE proposes use7ef the generic proceduve for detemining the bulk Ra-226 and Th-230 content 9'd on the generic procedure but stipulated that a f cobbly_soll for excastion control and verification. NRC concurre report detailing the site-s should be provided for NRC reviewinaseparatereport*pecificproceduresuse crinthe.CompletionRpport. Staff's evaluation gg 9;. and approval on the applicht' ion of the cobbles procedure would be given at y., that time. i > mj ~ l ( Tdq W }W,~ c/g u, SUPPLEMENTAL STANDARDS i+ ' f.[p c ' /4 k% ' h. DDE submitted (RAS Appendix'A,' including cal ati.n 17-730-02-02) an application for supplemental standards to exclude temediation of some windblown tallings areas and tallings buried around a gas line (see attached map). DOE also provided the'same information by 1 and NRC staff responded on Apr11.19,'1995,1with tw;etter dated October 7,1994, o general and six specific comments resulting from its' review of.the application.. DOE has not provided the additional infomation requested in those connients. A sumary of the application and NRC staff's revised cosmients. follow.- %3O y igwnyyr.W.. The areas under consideration'for supplemental standards total approximately 142 acres and contain approx'inately 119,600 cubic yards of contaminated soll. The general justification for not excavating these[ areas is that cleanup would be difficult and costly, and the material does notipose a significant health risk. The application of the supplemental standard of "no remedial action" fortheseareasisbasedon' meeting 40CFR'192.21.('a-f)criteriathat represent circumstances that would result due to remedial action. The Part 192 criteria applied by; DOE are:gA ~ S 4p' H t,s se M 1' clear and present risk 'of. injury to war ers;..or:the public; a. ,e environmental ham thatif s exce$.ww.v.wifellon"g-term, manifest)comparedtoth b. health benefits; and 'j termbenefits'g3. {. a;'gg high cost relative to l'o'ngials do not pose a clear present or futureat a Vicini c. residual radioactive mater hazard. g{g; g. l., The type of area and the abo've criterli~that DOE applied to each area are: River Front Wetlands (Former' Pile Area and Area E)?. b; Former Ore Storage l Area Steep Slopes - a, b, c'i Steep Slopes with Windblown (Areas B, C, D, E, F, G1,andG2)-a,b,c. %o j ' y% g~ :.. y g High-Pressure Gas Line - a,f v. b, c a f 1 0, ' ej. m 2

c m q

.,g' d{,3 F,'

..' s G NRC staff provided specific ^ comments on"each ar' a ' type, as well as the general e .E comment that all areas proposed by DOE for.the ips11 cation of supplemental 1. standards include the crfterion of environmental harm without adequate S. justification. DOE needi'.to address how partial remediation of the areas proposed for "no remediat' ion" would cause,"envi(onmental harm that is clearly 7 excessive compared to thW! health benefits to persons living on or near the site, now or in the future", as required b 40 CFR 192.21(b). This is an OPEN IS$UE. 3 @ M 4dj;h$. 4lt.4pt n U .cN a6 f$ DOE indicates that the r er. front wetlan (3.4 'cres) are under the a jurisdiction of the U.S. ray Corps of. Engineers', and areas with cottonwood and willow seedlings tha ' occur next to the rive'r are to be protected. However,theUMTRAProje'lhasreplacedwetlands)vegetationatothersites, c DOE also indicates that spillage of. contaminated material during excavation s f could contaminate the river, but some of this ma'terial probably enters the e river with each flood. As indicated above, 00Efneeds to address how ( remediation of the river 0 front wetlands "would.fhotwithstanding reasonable 9 measurestolimitdamage,hdirectlyproduceenvironmentalharmthatisclearly excessive compared to the health benefits to people living on or near the site, now or in the future,l",as required by 40 CFR 192.21(b). j .;uuin W.u & h..h for areas of steep slopes [fDOE indicates that there are relatively flat .y i thelong-termbenefits.!However,.thebenefitsa}renotdiscussed. portions, but that the Note 2 on Figure 3 (Appendix A),16dicates that relatively) flat areas on the steep banks will remain unexcavated because disturbance may cause erosion problems and the areas will be recontaminated from the steeper slopes above. The erosion and recontamination problems:and.the cost of solutichs are not discussed. DOE mentions that alternatives to conventionalsremedlial actions, presumably to reduce risk to workers, would elevate costsp but}the costs are not discussed relative to benefits. D0,E needs to. provide.a' di'scussion of costs of remediation versus benefits for the clean-up,ofithe areas of steep slopes. This is an OPEN ISSUE. f gg,gj i: The steep slopes of Areas 8, C, and E' ford'/ the^ east side ;f the highway, e Figure 3 indicates that, Area D covers'114.5 acres on the west side of the' typically,$ will not be excavated. it i S.. highway. It appears that.some portions of these: areas could be remediated l without excessive e,ost or risk. assuming that the elevated radon readings are I not due to natural (in situ)the DOE contractor l'n the field, DOE should deposits.3.Because f.inal excavation limits for all areas are determined by ~ provide guidance (possibly a reminder on the extent of excavation map) j indicating that the remediation will.come as close to meeting the otherwise v l f,' OPEN !$5UE. applicable standards as is' reasonable under the Eircumstances. This is an . s.; i c

avy up -

In addition, there is the potential ~for construction of a golf course on the I east side of the highway!at the Naturita site (A vehicle park on the westiside of the highway (Are(ea E) and a recreational a D) (February 22, 1995, letter from D. Crane,for #aturita). Chairman, Naturita Citizens, Group to W. Wood Project Site Manager There is no7ponsideration of these possible future uses of the areas in the informa~ tion provided by DOE. DOE rn y v _I' ,\\' $.hp f,k 4 h "3 .w y y 2 'f Li* >s f! f,h.f' ,f ,ji bhW h. O M [ ^

W~. ..jg Ep, r,, c -

a. ;. ;

y-

• l c,.

i y c u should provide a discussionfor,possibi futuref$ set of all the supplemental standard areas in the discussion of. potential health risks to persons that sight occupy the areas.. This is an OPDI ISSUE. &$). 6. t::Wg & h g.E W One steep windblown sha(. bled G2,4as an, average Ra-226 concentration of ( 53 pC1/g (10 samples) and W highest value was 205 pC1/g. DOE should indicate why the Ra-226 levil in Area G2 is higherythan typical windblown n areas and discuss the potential long-ters health benefits if the area is remediated. This is an OPEN ISSUE For the gas line, DOE proposes that 'a 5-foot-wide area on either side of the i G line remain unexcavated. Notes of a conversation (alth an employee of the to Natural Gas Company indicate? that excavation must be by hand for 3 feet on [ either side of the line. DOE indicates that the cdst and time for hiring specialized workers or provi'dtng workers extra training and special equipment ^ for work near the gas line imay not be warranted beiause of the low risk of public radiation exposure, but specialized workersfer training does not seem necessary for work several feet away from the line'.t. Also, DOE does not provide information to indicate what er.tity (the State of Colorado. City of Naturita, or Natural Gas Company)ld the gas,line be' dug up.would have the of the contaminated material = shou In addition,. Ra-226 data was not provided'for the assessment offpotential health risk because DOE assumed the remote location would prevent public exposure. DOE i needs to reconsider how much> removal can be performed around the gas line without increased unit cost $ Also, DOE needs te di'scuss what entity has the futureresponsibilityforanycontaminatedmaterialbexcavatedfromalongthe gas line and show that the designated entity has kriowledge of its responsibility. In additioni Ra-226 data for the a~rea along the gas line should be provided. Remediation. near the gas line jis an OPEN ISSUE. ^ N.:F A A record of a conversation Mth a representative of/the Natural Gas Company is the only land owner comment:provided by DOE.tDOE's'isubmittal states that a record of negotiations with!the other ownerslis dochmented in the Remedial Action Agreement, but no copies of these agreements;or discussions were provided. As required by 40.CFR 192.22(c) DOE should provide NRC with copies of any coments from land owners regarding the proposed application of Y supplemental standards to portions of.their property. This is an OPEN ISSUE. RadionuclidesOtherThanRa-226[;,y- - u J, n-2: p.; y, ', DOEstatedthatforTh-230contaminatinTsu..,n.:pplemenjal standards will be based on the NRC-approved generic. thorium policy,oUranium concentrations, after the Ra-226 has been removed to miist standards, will be lassessed by a pathway analysis of potential enviro) mental and health impacts. If the analysis indicates that remedial action,for uranium is neede'd, DOE will propose a supplemental standard. Thes's approaches.are' accept lable to staff, if DOE and its contractors reduce the cintamination to. levels A[.LARA, as require CFR 192.22 (a). f.g v.7 5 .glj; r t di J. y 6.3.3 Verification .. -Qo fh 4

t]d.,1.$j'y% k%

t-y e n

4.. e,

.h ( G r {vco ,i T y f f

m The final radiological verification survey for land! cleanup will be based on 100-square-meter areas. The standard method for analysis of composite soll s'akples by gamma spectromlRa-226 verification is etry, but DOE may use several other measurement tedhniques, depending on putticular circumstances. DOE indicates that the nine-pbint composite gamma measurement technique or the RTRAK detection unit may be used in the windblown afeas. NRC has previously agreed to the use of these prbcedures with adequate? uality control in specific cases. t{, 3.. DOE stated that verificationCfor Th-230 will folloM he generic thorium policy. Also, any uranium cleanup verification will.:be derived as part of a supplemental standard. .t1

n. w@gW o A.

G WMie: 'ij.. lA No on-site structures at the$ roc ~eising siteYllf riq{Iire radiological verification, because all structures will be ' demolished and the debris will be buried at the disposal cell. y gfjj,p g,q/ g r O 6.4 Conclusions j .).h Based on review of the desigrPand analy'ses' presente in the RAP and associated documents, NRC staff concludes that the Umeteo radon attenuation model has 2 some short-comings, but model barrier design is conservativing by the staff demonstrates that the radon e. Radon attenuation provided by the frost protection layer was, conservatively, not considered in either model. In addition, DOE and Umetco havej committed to do further testing of materials and DOE will perform a final radon flux analysis with asibuilt parameter values to verify the design. This fina) analysis will be provi,ded for NRC review as part of the Completion Report? Therefore, assumingi&11 cell stability issues will be resolved, staff is aspured that the averageisurface radon flux will be below the EPA standard. y 'w. a 1 g w .,;y g The staff finds that the radiological characterization program, and the proposed processing site cleanup and verification plans require more information, or modification.1 The OPEN. ISSUES concerning site cleanup to be addressed by DOE are: 'J ' t.. .S .f ./$ 1. DOE should provide the additional Th-230 and U-238 data obtained with the cobbly soil study to substantiate that adequate. characterization of these radionuclides has been performed. p g M e!w.- ii'. ,,,wgs d y. 2. 00E should provide more soil backfrdun[Rb226 data and correct the soil Ra-226 data in RAS Table 6.1. d_ ,p l d.kh,lc,.W h 3. 00E needs to address how partial remediation of xthe areas proposed for "no remediation" would cause fenvironmental harm that is clearly excessive compared to the health benefits to persons livini on or near the site, now or in the future" as required by 40 CFR 192.21(b). n$ Chh?.' T,E Y ms 6 4. 00E should remove the former ore storage area from consideration of supplemental standards u(der Part 192.21(c). m[ n 'V 4 6 e a s. e Ik.. '..M, h6[id M k 'A ) b a

1 4 g j ~ 5. DCE should provide guida'ocf('p'ossibly a reminder on the extent of excavationmap)indicatthgthattheremediation?willcomeascloseto meeting the otherwise applicable. standards as s reasonable under the Ciremstances. ,, J4 :... m.W. 9Nt -W we v Age;',4 b,N4yr#w% + 6. DOEshouldprovideadislussion'c poss'ible fut"ure uses of all the supplemental standard are'as in the discussion of potential health risks to persons that might occu' the areas. f.g g &g g~ f;,A p pf 4Q DOE should indicate why' he'Ra-226~ evel'In A Na[G2 is higher than typical 7. windblown areas and disciiss the potential long 't,erm health benefits if the area is remediated. p p DOE needs to reconsider ow'auch removal can befperformed around the gas 8. line without increased unit cost.M Also,' DOE needs to discuss what entity s.. has the future responsibility for any contaminated material excavated from r along the gas line and sh'ow that the designated? entity has knowledge of its responsibility. In' addition,' Ra-226 data f6r the area along the gas line should be provided.d.t A% . 4 -QiMidh't!2 E i As required by 40 CFR 192.22(c)', DOE'should prov,I F. Ms c 9. ide NRC with copies of any a commentsfromlandownersTregardingtheproposed@pplicationof supplemental standards t'olportions of their pro 7erty. W j, fh i((y, k2.V L.SCM MINOR COMENTS 4 ' 'f . kN.'.% V y p I. Umetco's Section 6.3.1 iddicates that the long-term moisture value for windblown material was 1210 percent, but the tab'le of code input values in calculation 010 (sheet 5)iindicates that 13.0 pe,rcent was used. Umetco should correct this discrepancy. ' gh g' t) e .i>, The geotechnical data for' contaminated material (qpresented in Uravan RAP 2. Sections 6.3.1 and 6.3.4!do not reflect the current information provided in DOE's Naturita RAP calculation 17-737-01-02 3 Sheet 8 of Umetco's calculation 010 does pres ~ent the current data which was used in Umetco's modeling. The Uravan RAP: sections should be corrected. .,., A7.cy;.W.. y 3. The Supplemental Standards Application indicates;(page 6) that steep areas of the foneer ore storage area are eligible for! consideration of supplemental standards under Part 192.21(c) A Hoeever, the ore storage area is part of the designated processing site and Criterion (c) is only applicable to vicinity properties.y. - %.Page 6.shouli! be corrected, . N C.h Appendix A of the Supplein' ntal Standard Application contains duplicate 4. e Ra-226 samples in the average'value for the cre @torage area, and Areas B, AreaDeitherhaveatypo)graphicalierrorinthe:h226samplelocationsin D(fourpairs),E(steep),andF."Also,fourRa north coordinate (N49500 and N49200), or the samples were not taken in Arda D. At location N54200/E49315, there is either a typographical s'rror on the east coordinate, or the sampleiwas not taken in Area:D. h! i fl. WG 9

y... tim &'

4 n ,W i.d M A g r,:7 M(' ' e ' (,; w k *..

1 OE'** wJ.s+ff<jts, her. A v' O :- hi%' i di ' (:; f W !p $n$ Q V j @p f. /c cg:!'gs $ ? ', - X 22 a ;.4 2 g h M: 6.5 References

r.,.; &g. otu; y

.m.

x., : q DOE (U.S. Department of Energy).' Reinedial Action Plan for Naturita, March 1994 l

. k-d). W m.?aW !g..; - i Remedial Action Plan and Site'besign for Stabilization of the Naturita j Rate,rlais at Uravan, Colorado,; Final Draft, Novemy9%... .-1995. i

q,:

v NRC (U.S. Nuclear Regulatory,Commiss%).%,b.:y?;s. ion)f Washington D C i

Regulatory Guide 3.64." Calculation-of Radon Flux At'tenuation by Earthen l Uranium Mill Tallings Covers," June 1989.'MjP,f.. '!/).. i . mccswpr go.g.y g- --, Division of Low-Level Waste'Managisiesand Decommissioning, " Final Standard Raview Plan for the'; Review of Remedial Act~ ion of Inactive Mill Tallings Sites under Title I of, Act, Revision 1," June 1993.'j(pg,,the-Uranium Mill, Tailings Radiatio l {.;. ?gg.pg{in. ] T. >$ 8hs v.:,5$f.Y.,.'},.: .? V.;',jJl}y[5,f' h,z.'. aw.;f '- . 7 es u. ,t 7.14 ft [k, k.

h..

I '. 'i ~ *;' omyh . w[sp f, e.f.,. J, A..,, l n!./ /,. . a" 3. 'o. , \\,. .4,.. ,, i 4 34.s . (. i " 6.h'p,,,, d) I., ) ~ - .% ; 0.W '*%*g

• $fll'$5lOll;:b!

f(. ll.e

?

W {:,q. 0 ' ' '&) !

• N f..l.f)Yh)Q 1

f. ,y[*M: O

2... Q,c

,n... ' f.S...,. (.. cu.? :df. ,f. Uis n.4.... 'M - \\. X)." b l

?^ { '*,*.;*..G f.:

k % ' :: ;;1.:%- r,'.>;,Mu.ti;:.,1l +(Q ',', ' -).7 .O.. ..* w. .[y b

.g e.; i;,:'r '^l;f'My ; *j.n

=,d;, -

l

n ' '-

. r. 'fd N"7i$;r%-&x.ps.? j ii '. y b h *,.Q, $ .h w{ . <.:.3 n wl% jI ', yu I.T'.;.0.,l .c.f. e :L;..,{; r. < 3' -p g. ,s v \\ ',3.,T..l [..* ..{ $ ' \\, j ,,s< 3 f ]. 'y[.=,.A.... @ 'y' ,s, e .....v,e : r. '*1 .c' 8j s

• c f

p I M-gd Ix., ] :y -w- .'Sv s ,I

• f; '

.A '- , h{ l *,1 L*; l e t .f.. s f lf, ft N.! s s' ( l 'J+ "/ w.g,b' t;

• G

~,...a, " s ..g i n. e c.;' ..44 J., Q sy. y n!, n 1 12 a. L t p., , c. e.,. ..e. 4 6 [ h> : gi* .h, 1_. a f,- l$ l.. a l' %,. '/f. h g4 ',,..: A,'h

.5.s.\\., h>

c(J 'dc. w#ft g p .i', e. v ai

! l Anne,R:mirez-naturita Feb TER.wpd Page 1l Draft 02/26/99 l i NATURITA TER l 5.0 WATER RESOURCES PROTECTION l 5.1 Introduction The Naturita processing site is located in the San Miguel River Valley and is underlain by unconsolidated alluvial floodplain deposits and fill material. The alluvium is the upper-most aquifer at the processing site and is contaminated from former processing-site activities. The alluvium is underlain by the Brushy Basin and the Salt Wash Members of the Morrison Formation. The Brushy Basin consists of interbedded shale, sandstone, and conglomerate lenses. The Salt Wash consists predominantly of sandstone with some shale. The Brushy Basin and the Salt Wash have not been affected by uranium processing activities. Existing groundwater contamination does not presently represent a risk to human health or the environment. However, DOE is required to demonstrate that cleanup or control of existing processing-related groundwater contamination at the Naturita site will comply with the EPA groundwater l protection standards in Subpart B of 40 CFR Part 192. Groundwater cleanup at the former j l processing site will be addressed under a separate DOE program and a National l Environmental Policy Act process; using strategies and options outlined in a programmatic l environmentalimpact statement that has been developed for the UMTRA Project. The need for and extent of groundwater cleanup at the Naturita site will be evaluated based on the extent of existing contamination, the potential for current or future groundwater use from the uppermost aquifer, and protection of human health and the environment. At the Upper Burbank Disposal site, DOE must comply with the final standards issued by the EPA on January 11,1995 (40 CFR 192.20). From a review of the information submitted;it i appears that the site will comply with the requirements of Subpart A of the EPA groundwater ] protection standards. However, the NRC staff has yet to reach a determination on the long-term surveillance plan which the DOE has not yet submitted to the NRC. 5.2 Hydrogeologic Characterization I 5.2.1 Identification of Hydrogeologic Units i A. Processing Site l At the Naturita processing site; unconfined groundwater occurs within the alluvial flood plain deposits from 3 to 18 ft (0.9 to 5.5 m) below the land surface. The saturated thickness is approximately 15 ft (4.6 m)in the vicinity of the site. The next deepest aquifer at the site is the Salt Wash Member of the Morrison formation, which consists predominantly of sandstone with some shale. The Salt Wash aquifer is separated from the alluvial aquifer by the Brushy Basin Member of the Morrison formation. Under the site, the Brushy Basin Member is considered an aquitard and consists of thick, laterally extensive, interbedded shales with some sandstones, it ranges in thickness from 110 to 105 ft (33.5 to 50.3 m). The top of the Salt Wash aquiferis approximately 130 to 165 ft (39.6 to 50.3 m) below land surface near the site. The total O O \\

f-l [ Anne Ramirez - naturita Feb TER.wpd Page 2 j Draft 02/26/99 I thickness of the Salt Wash aquifer has not been determined, but is at least 80 ft (24.4 m) thick in the vicinity of the processing site. 8. Disposal Site Five principal hydrostratigraphic units occur within the upper 800 ft (240 m) of sediments beneath the disposal site. From the land surface down, these are: (1) sandstones and shales of the Salt Wash Member of the Morrison Formation; (2) shales and siltstones of the Summerville Formation; (3) sandstone of the Entrada; (4) sandstones of the Kayenta; and (5) sandstones of the Wingate Formation. The Salt Wash Member directly underlies the disposal site and has a thickness of about 120 ft (37 m). This unit is predominantly comprised of sandstone with some interbedded shale layers. The Summerville Formation underlies the Sa,t Wash Member. This unit is considered an aquitard and is 90 ft (27 m) thick. It is composed of massive clayey mudstones, silty shales, clayey siltstones, and minor, interbedded sandstones. The Entrada Formation underlies the Summerville Formation and has a thickness of app'oximately 160 ft (49 m). The Summerville Formation is a sandstone. Underneath the Summerville formation is the Kayenta Formation, which has a thickness of approximately 180 ft (95 m). This formation consist of interbedded layers of sandstone, siltstone, shale, and some conglomerate. Below the Kayenta Formation is the Wingate Formation. This formation is about 250 ft (76 m) thick and is a sandstone. The Kayenta Formation together with the Wingate Formation form the first saturated aquifer beneath the disposal site. Below the Wingate Formation is the Chinle Farmation. The Chinle Formation is about 400 ft (120 m) thick and is predominantly a siltstone. Because of it's low permeability, the Chinle Formation acts as an aquitard to vertical groundwater movement. 5.2.2 Hydraulic and Transport Properties A. Processing Site The occurrence of shallow groundwater in the alluvial aquifer is limited by the lateral extent of the alluvium in the vicinity of the Nattrita processing site. The average hydraulic conductivity for the alluvial aquifer is 3.0 ft/ day (0.001 cm/sec) and the average linear groundwater velocity is 0.06 ft per day (2x10-5 cm/sec). The groundwater flow direction in the alluvium is subparallel (northwest) to the San Miguel River. Groundwater from the alluvial aquifer discharges into the San Miguel River northwest of the site. The Salt Wash aquifer is a major regional groundwater system in the area. The potential area of natural discharge from the Salt Wash aquifer is the San Miguel River northwest of the processing site. For th.e Saft Wash aguder, hydraulic conductivities averaged 0.06 ft/ day (2x10-5 cm/sec) and the averzge Eneargmundwater velocity is estimated to be 0.002 ft/ day i (7x10-7 cm/sec). The S'afte Wash aguder is separated from the alluvial aquifer by the Brushy Basin Member, which loca'Irris considered an aquitard Groundwater in the Salt Wash aquifer is confined, and has a pctatmetit surf ace that is higher in elevation than the water table in the alluvial aquifer. Therefere, if any significant flow were to occur between the Salt Wash i ~

[ Anna R.:mirez - naturita fib TER.wpd Page 3 { 1 i Dr 3 02/26/99 aquifer and the alluvial aquifer, water movement would be upwards from the Salt Wash aquifer through the Brushy Basin Member and into the alluvial aquifer. B. Disposal Site The disposal site is underlain by approximately 600 ft (180 m) of unsaturated sanc. stone, siltstone, and shale. The Summerville Formation, composed of shale and siltstone, is about 120 ft (37 m) below the site and has a hydraulic conductivity of less than 0.01 ftlyr (1.0x10-e cm/sec). Thk 90 ft (27m) thick layer functions as an aquitard and should prevent any potential groundwater contamination from the disposal site from reaching the Kayenta/Wingate aquifer. The Kayenta/Wingate Formation is the uppermost aquifer beneath the disposal site. Only the Kayenta/Wingate aquifer is saturated beneath the disposal site. The aquifer is unconfined and at a depth of approximately 600 ft (180 m). Average hydraulic conductivity of the Wingate formation is 0.12 ft/ day (4.2x10-5 cm/sec). Groundwater flow beneath the repository is toward the north at velocity of about 8 ftlyr. Primary recharge to the Kayenta/Wingate aquifer is northeast of the San Miguel River along the Uncompahgre Plateau. Secondary recharge to the Wingate portion of the aquifer is from the Paradox Valley south of the site. Discharge from the aquifer is to the San Miguel River. 5.2.3 Extent of Contamination A. Processing Site To determine whether uranium processing activities at the Naturita processing site have influenced groundwater quality in the alluvial aquifer; DOE collected semples from on-site and down gradient monitor wells and analyzed these samples for the constituents (including lead, nitrate and silver) listed in Table 1 to Subpart A and Appendix 1 of 40 CFR Part 192. Based on this analysis; arsenic, cadmium, chromium, fluoride, methylene chloride, molybdenum, selenium, strontium, thallium, tin, toluene, uranium and vanadium, radium-226 and -228, and gross alpha and may be contaminants in the alluvial aquifer groundwater. Of these constituents; arsenic, cadmium, molybdenum, selenium, uranium, radium-226 and -228, and I gross alpha were found to exceed the EPA maximum concentra: ion limits. Based on uranium concentrations, a contaminant plume extends at least 1500 ft (460 m) down gradient from the former mill yard and has a maximum width of approximately 900 ft (275 m). Contaminated groundwater from the alluvial aquifer discharges into the San Miguel River, where no impacts on the river water quality have been observed to date. j I Groundwaterin the Salt Wash aquifer was found to not be contaminated as a result of milling operations at the processing site. . B. Disposal Site Groundwater at the disposal site is presently uncontaminated by the disposal of contaminated material. ~ i

r l Anne Ramir;z-n turita Feb TER.wpd Page 4 l l Draft 02/26/99 5.2.4 Water Use A. Processing Site Eight wells are located within 2 miles (3.2 kilometers) of the processing site. Water from these wells are used for domestic purposes. Five of these wells are located up gradient from the processing site (four obtain water from the alluvial aquifer and one from the Salt Wash aquifer). The two remaining wells are located down gradient of the site, but are located on the i opposite side of the river. There should be no threat to up gradient wells from contaminated groundwater from the processing site in the alluvial aquifer as groundwater flow is in the opposite direction from these wells. Groundwater pollution in the alluvial aquifer should not be a threat to down gradient wells because the river represents a discharge point for the alluvial aquifer. Since, these wells are located on the other side of the river, groundwater contamination in the alluvial aquifer should not reach them. No impacts have been observed or are expected to groundwater in the Salt Wash aquifer from the processing site. No impacts to surface water quality have been observed from contaminated alluvial groundwater at the site. Future use of groundwater in the alluvium for do;nestic consumption is not expected. The is because the alluvial aquifer has a very low potential for use as a source of water, since it is limited to the small area of alluvium in and adjacent to the San Miguel River. Alternative supplies of reliable, good quality water are available from the town of Naturita, from surface water. and from deeper groundwater aquifers. l e3. Disposal Site Four wells produce groundwater within two miles (3.2 kilometers) of the disposal site. All of these wells are owned by Umetco Minerals Corporation. These wells are no longer being used and will be plugged prior to deeding the land in and around Uravan to the federal government. Five additional wells are within a radius of five miles from the site. All of these wells are up-gradient of the disposal site and therefore cannot be impacted by the site. Future use of groundwater beneath the Upper Burbank site is limited by the poor water quality and low permeability of the Wingate Formation and significant depth [600 ft (180 m)] to groundwater. In the future, land will be deeded to the federal government, which will limit the development of groundwater resources by the general public. There are no agricultural or domestic surface-water users within a 2-mile (3.2-kilometer) radius of the site. Umetco Minerals Corporation does have water rights on the San Miguel River for industrial uses. Within a 5-mile (8-kilometer) radius of the site, water use is limited to springs used for stock water. The closest spring used for stock watering is located about 4 miles (6 kilometers) northeast of the site. 5.3 Conceptual Desiga Features tb Protect Water Resources A. Processing Siter Groundwater contamincion does notrepre ent a risk to human health or the environment 1

l Anne Ramirez-naturita Feb TER.wpd Page 5 { Draft 02/26/99 because there is currently no consumption of the contaminated groundwater in the alluvial aquifer. The distribution of hazardous constituents in groundwater will decrease with time, because contaminated material is being moved off site to another location and because the alluvial aquifer discharges to the San Miguel River. This means that existing contamination in alluvial aquifer groundwater will eventually be flushed out of the aquifer and diluted by the water in the San Miguel River. Groundwater cleanup at the former processing site will be addressed under a separate DOE program and a National Environmental Policy Act process; using strategies and options outlined in a programmatic environmental impact statement that has been developed for the UMTRA Project B. Disposal Site The climate in the vicir.ity of the disposal site is semiarid. Under natural conditions deep percolation at the disposal site is less than 0.01 f 1/ft /yr and may for all practicable purposes ) 3 2 be zero. The lack of a perched zone under the disposal site and the lack of springs and seeps along the canyon walls further indicate that the site has a very low infiltration rate. The disposal site was modeled by DOE to determine possible infiltration rates after cover j 3 2 construction. Modeling results suggest'a very low infiltration rate of about 0.028 f t /ft /yr i 3 through the cover. This infiltration rate equates to a flow of approximately 0.1 gpm (9660 ft /yr) through the base of the 8-acre disposal cell. This indicates that very little deep percolation should occur underneath the disposal cell. Travel time for liquid from the base of the disposal cell through the Summerville Formation was calculated to be in excess of 1,000 years. The residual radioactive materials that will be disposed of at the site are principally contaminated soils. The contamination of these soils should be relatively low due to the mixing of the original tailings materials with surficial soils. Batch tests performed on this soil material confirm that this material has relatively low concentrations of radionuclides and heavy metals. Geochemical attenuation of any leachate from the disposal site would occur as contaminated water flows through the bedrock formation. 1 5.4 Disposal and Control of Residual Radioactive Materials 5.4.1 Water Resources Protection Standards For the Disposal Site The groundwater standards (40 CFR 192.02) require three basic elements for setting the groundwater protection standards. These are (1) determination of hazardous constituents; (2) proposal of a concentration limit for eech hazardous constituent found to exist in the tailings or leachate; and (3) specification of the point of compliance. The DOE analyzed groundwater samples from the alluvial aquifer beneath the Naturita processing site and conducted laboratory batch leach tests of contaminated soil material from the Naturita processing site. Based on these tests, DOE identified 25 hazardous water quality parameters that are reasonably expected to te in.cc derived from residual radioactive material to be disposed of at the site. These parameterswe? selected to be monitored at the point of compliance and are presented in Table 5-1. Fcrthase pararreters. the DOE has established concentration limits (Table 5-1). The proposed oorcentration fimits are the either the maximum concentration limit or for those hazardout constitue;ts webout maximum concentration limits, the statistical 9

f' l Anne Ramirez - naturita Feb TER.wpd Page 6l l Draft 02/26/99 maximum of background groundwater q'uality derived from water samples collected from Wingate well CM93-1 and CM93-2. Concentration limits for strontium and tin will be determined by routine sampling of CM 93-1 and CM 93-2 during long-term site surveillance activities. Since DOE will sample for many other parameters other than tin and strontium, the ground-water protecticn program should detect any potential contamination in the ground water from the disposal site, during the period when tin and strontium concentration limits are being established. Wingate well CM93-2 is designated as the point of compliance for the disposal site. This well ils immediately down gradient of the disposal cell. 5.4.2 Performance Assessment for the Disposal Site DOE must demonstrate that the performance of the disposal unit will comply with EPA's groundwater protection standards in 40 CFR 192 Subparts A and C. The disposal cell design should minimize and control releases of hazardous constituents to groundwater and surface water to the extent necessary to protect human health. The following are important to performance of the disposal site: 1. The uppermost aquifer, the Wingate Formation, lies approximately 600 ft (180 m) below the base of the disposal cell and is hydrogeologically isolated from surface recharge or initial transient drainage from the disposal cell by low permeability shales and mudstones overlying the aquifer. 2. The Summerville Formation. the principal aquitard beneath the site, is approximately 90 ft (27 m) thick and effectively isolates groundwater in the I underlying Kayenta/Wingate aquifer from potential contaminants in the disposal cell. 3. Geochemical properties of the bedrock materials attenuate hazardous constituents possibly associated with leaching of the residual radioactive materials. 4. The multi-layered cover reduces the infiltration rate and minimizes long-term seepage from the cell. 5. The disposal cell will be contoured to provide efficient drainage of precipitation away from the disposal cell and to minimize excess moisture in the cover and associated infiltration. 5.4.3 Closure Performance Demonstration for the Disposal Site DOE must demonstrate that ttre proposed disposal design will (1) minimize and control groundwater contaminatico,(2) minimize the need for further maintenance, and (3) meet initial performance standards of cce design,in acccrdance with the closure performance standards of 40 CFR 192.02. The csposd cett design uses a multi-layered cover to reduce the infiltration rate and minsize bg4cen seepage from the cell. The disposal cell will be I e

' [ Anne Ramir:z - niturita Feb TER.wpd _ Page 7] Draft 02/26/99 contoured to provide' efficient drainage of precipitation away from the disposal cell and to minimize excess moisture in the cover and associated infiltration. In addition natural stable material will be used in constructing the disposal cell to minimize the need for further maintenance. L 5.4.4 Groundwater Monitoring and Corrective Action Plan at the Disposal Site

The DOE is required by 40 CFR 192.03 to implement groundwater monitoring during the post-disposal period for the purpose of demonstrating that the disposal cell will perform in accordance with the design. 40 CFR 192.04 requires the implementation of a corrective action i

program if the monitoring shows an exceedance of concentration limits. The monitoring plan required under 40 CFR 192.03 should be designed to include verification of the site-specific assumptions used to project the disposal system performance. Prevention of groundwater contamination may be assessed by indirect methods,'such as measuring the moisture migration within varus components of the cover, tailings, or beneath the tailings; as well as direct groundwater monitoring. j At the disposal site, DOE will monitor potential repository seepage using wells at the contact of the Salt Wash and Summerville Formations near the disposal cell for a period of time following i completion of remedial action. Monitoring any perched groundwater on the top of the Summerville from the disposal cellincludes well BR95-1, BR95-2, and BR95-3. If seepage is detected in these monitor wells, performance monitoring of wells CM93-1 and CM93-2 will be conducted. 5.5 Clean-up and Control of Existing Contamination at the Processing Site The DOE is required to demonstrate that cleanup or control of existing processing-related j groundwater contamination at the Naturita site will comply with the EPA groundwater protection standards in Subpart B of 40 CFR Part 192. Groundwater cleanup at the former ) processing site will be addressed under a separate DOE program and a National Environmental Policy Act process; using strategies and options outlined in a programmatic environmentalimpact statement that has been developed for the UMTRA Project. The need for and extent of groundwater cleanup at the Naturita site will be evaluated based on the i . extent of existing contamination, the potential for current or future use of groundwater from the i uppermost aquifer, and protection of human health and the environment. 5.6 Conclusions The staff concludes that the proposed remedial action for the Naturita sites will acceptably comply.with the EPA groundwater standards, with the exception of the following open issues which may be deferred: 1. DOE must demonstrate compliance with EPA's groundwater clean-up standards in 40 CFR 192, Subparts B and C at the Naturita processing site (deferral provided by the UMTRCA amendment of 1982). 2. DOE must provide ttte details of its groundwater monitoring program (sampling I

l AnnQ RamirGz - naturita Feb TER.wpd Page 8 l Draft 02/26/99 frequency, etc.) for the disposal site to demonstrate compliance with 40 CFR 192.03. This information can be included when DOE submits the long-term surveillance plan to the NRC for review. e

[^nne Ramirez - niturit3 F;b TER.wpd Page 9] i i Draft 02/26/99 l . Table 51 Hazardous Constituents and Concentration Limits for the Disposal Site Constituent Concentration Limit Aluminum 0.1 mg/L (background) Antimony 0.1 mg/L (background) Arsenic 0.05 mgIL (maximum concentration limit) Barium 1.0 mg/L (maximum concentration limit) Beryllium 0.05 mg/L (background) Cadmium 0.01 mg/L (maximum concentration limit) Chromium 0.05 mg/L (maximum concentration limit) i Copper 0.02 mg/L (background) ~ Cyanide 0.01 mg/L (background) Fluoride 5.9 mg/L (background) Gross alpha (excluding uranium and radon) 44.7 pC1/L (background) Lead 0.05 mg/L (maximum concentration limit) Mercury 0.002 mg/L (maximum concentration limit) Molybdenum 0.1 mgIL (maximum concentration limit) Nickel 0.05 mg/L (background) Nitrate (as N) 10 mg/L (maximum concentration limit) Radium-226 and -228 5.0 pCl/L (maximum concentration limit) i Selenium 0.01 mgIL (maximum concentration limit) Silver 0.05 mgIL (maximum concentration limit) Strontium 0.1 mg/L (background) Thallium 0.01 mg/L (background) Tin 0.005 mg/L (background) Uranium 0.044 mg/L (maximum concentration limit) Vanadium 0.05 mg/L (background) Zinc 15.5 mg/L (background) j O}}