ML20217H457

Forwards Preliminary Draft Comments Resulting from Review of Riverton,Wy Draft Ea.Submittal Does Not Constitute Final Review of Reverton Draft EA

ML20217H457
Person / Time
Issue date:	12/08/1984
From:	Higginbotham L NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS)
To:	Themelis J ENERGY, DEPT. OF
References
REF-WM-60, TASK-TF, TASK-URFO NUDOCS 9710160047
Download: ML20217H457 (36)

Text

__--

4 9 Clog #

f"

! y'

/ #o,

• , UNITED STATES l+ 3 $ NUCLEAR REGULATORY COMMISSION wAsMiwotos. o. c. gooss

\,....* / -

NC$  %

A I Mr. John G. Themelis, Project Manager Uranium Mill Tailings Project Office

^N Albuquerque Operations Office (d

' U. S. Department of Energy P. O. Box 5400 Albuquerque, New Mexico 87115

Dear Mr. Themelis:

Enclosed are our preliminary draft coments resulting from our review of the I;*

Riverton, WY Draft Environmental Assessment (EA). It is my understanding that our draft comments may assist in minimizing delay in your schedule for the Riverton remedial action. However, you should be aware that this submittal does not constitute our final review of the Riverton Draft EA.

Should Flory on you FTShave any coments regarding this submittal please contact Claude A.

427-4554 F

{SifF ly.

% ' A" k Leo B. H botham, Chief Low-level Waste and Uranium Recovery Projects Branch Division of Waste Management

Enclosure:

As stoted 9710160047 841200 PDR WASTE WM-60 PDR o' __

Nj k

Mbtw) buy ) lA*)' (o O

. .: - ste i m nE uz DISTRIBUTION:

Robert E. Browning WM William Ford, WMGT Michael J. Bell, WM Ted Johnson, WMGT Joseph 0. Bunting WMPC Jose Valdes, WMGT Malcolm R. Knapp, WMGT Michael Weber, WMGT Lake H. Barrett, WMEG Dan E. Martin, WMLU John Greeves, WMEG Roger A. Pennifill, WMLU 3, Myr. ore S. Nataraja. WMEG Claude A. Flory. WMLU  ;

Timothy Johnson, WMEG Nrk Haisfield, WMLU ' '

Myron Fliegel, WMGT Yvonne Young, WMLU R. John Starmer, WMGT Robert Fonner, ELD l s Philip S. Justus, WMGT R. Dale Smith, URF0 n Daniel-Gillen, WMEG Dennis Sollerberger, WHLU V

Steve Smykowski, WMEG Kitty Dragonette WHLU Banad Jagannath, WMEG William Ford, WMGT Michael Blackford, WMGT Mark Larson, WMGT William Dam, WMGT l

From: Giorgio Gnugnoli, WMLU i

-m

a ga seg E

'd 8,, UNITED STATES n

g NUCLEAR REGULATORY COMMISSION WA&&HNQ TON, o. C,20696 Mr. John Themelis, Project Manager Uranium Mill Tailings Project Office Albuquerque Operation Office U. S. Department of Energy P. O. Box 5400 Albuquerque, New Mexico 87115

Dear Mr. Themelis:

The NRCTwo Enclosure staff has reviewed contains our detailedthecomentsRiverton grouped Draft Environmental Assessme by subjects. -

pd Enclosure One sumarizes the detailed connents for certain subject areas. As explained in the enclosed connents, we are continuing our review in the areas of geomorphic stability and radon dispersion in air. We will submit coments on these areas at a later time.

Because the draft EA inoicated that some of the environmental impacts are stim under study, we expect to review additional infonnation prior to publication of the final EA. This additional needed information is identified in the enclosed coments.

review As this needed additional information is submitted to us we will it expeditiously.

Sincerely, Leo B. Higginbotham, Chief Low-level Waste and Uranium Recovery Projects Branch Division of Waste Management

Enclosures:

1. Sumary Coments - Riverton DEA

2. Detailed Coments - Riverton DEA i

l l

l l

l J

1. .

e' ENCLOSURE ONE i

i MAfT .-

l

'SU W RY COMMENTS NRC REVIEW OF THE ~

RIVERTON ORAFT ENVIRONMENTAL. ASSESSMENT I, m

i n

2-DiUH GE0HYDROLOGYANDHYDROLOGl The Draft Environmental Assessment fails to adequately define baseline groundwa*,tr quality. This is important, because without an adequate  ;

determination of background, the extent of groundwater contamidation Cannot be i detennined. Present groundwater quality maps not only show groundwater contamination moving towards the south into the Little Wind River, but away from the site towards the north.

No indication is made of what will be done with the contaminated groundwater and geologic materials in the unconfined aquifer. This is important, because the unconfined aquifer may be used as a future groundwater resource or fsN penetrated by wells seeking water in deeper aquifers. Additionally, the -

V unconfined aquifer is a known source of gravel, which may have been contamir.ated.

The preferred alternative of stabilization-in-place in conjunction with emplacement of a slurry wall is based on the assumption that there is no hydraulic connectie between the unconfined aquifer and the deeper confined F sandstone aquifers. However, this lack of communication is not supported in

the Draft Environmental Assessment (DEA). For example, the elevated concentrations of total dissolved solids, sulfate, uranium, and manganese which are present in the first confined aquifer indicate comunication between the unconfined and confined act.ifers, in addition, details of existing local domestic well completions are not presented. Depending on the type of completion (gravel packs, grouts etc.), these wells may be providing avenues for contaminant migration between aquifers.

The Draft Environmental Assessment does not indicate how the restored tailings pile will be protected from future gravel mining activities around the pile.

W The text does not explain what type of geologic material the slurry wall will to completed in. This is important, because if the slurry wall is completed Into the gravel or sand of the unconfined acuifer contaminant pathways under it would not be blocked, in order to be effective, the slurry wall must at least be completed into the confining material beneath the unconfined aquifer.

The text contains " considerable confusion about the confined aquifer. Data in the Draf t Environmental Assessment indicate that the confined aquifer is really made up of more than one aquifer, whereas the text treats the confined aquifer as one unit.

incorrect conclusions about the confined aquifer system.The text is '

y a%

k

.? .

\

, %,. q ,

The eohydrologic data for the Dry Cheyenne Site is inadequate if this site shou d become the preferred site.

'ih*

l.ocal hydraulic gradients within the confined aquifer and vertical gradients between aquifers have not been determined. These are important in order to evaluate the extent of contamination and the effect of remedial action alternatives.

We question the validity of statistical flood estimates for this site, due to the limited data base which was used to extrapolate the data.

_ GEOCHEMISTRY I The Riverton Draft Environmental Astessment presents geochemical data which indicates that there is interconnection between the contaminated alluvial aquifer and the deeper confined aquifer system. However, the DEA seems.to draw exactly the opposite conclusiun.

The text is deficient in several areas including no analysis of potential f organic :ontamination from the solvent extraction process used at Riverton.

• Organics are known to mobilize heavy metals. Although the text indicates that a carbonate leach process produced carbonate mill tailings it does not describe these tailings at all. Because migration of heavy metals is greater under alkaline conditions, carbonate tailings may be a source of significant impact on groundwater resources.

The DEA suggests that geochemical computer codes can greatly facilitate remecial action analyses. Results are presented from codes that are not documented or validated with known field results. Limitations and assumptions of the codes are not discussed. Furthermore, the text does not state how the modeling results will be used to support a remedial action decision.

GEOLOGY AND GEOMORPHOLOGY c

In many respects, the relative merits and shortcomings of the processing site versus the alternate site are not accurately portrayed in the EA. For example!

(a) hazards at th9 Ory Cheyenne site from ephemeral stream erosion and flooding, as well as the potential need for related protective measures, are not properly addressed, (b) the conclusion that the Dry Cheyenne site would have a greater impact on m(rf 'al resources than the Riverton site does not follow from the infonnation and arguments presented. (c) the conclusion that the Riverton site would not impact on a floodplain seems to be contradicted by the observation that the site is on a floodplain, p

,4, l

i i

lhe stratigraphic characterization of the Riverton site, as portrayed in the EA, is not adequate. Dctailed stratigraphis cross sections and supporting  !

i infomation (well-location maps, well logs, detailed petrologic-mineralogic core description) should be presented.

CertainaspectsofgeomorphicstabilityattheRivertonsite(i.e., river ,

encroachment and PMF flows) are described in the EA as still undergoing analysis.

Conenents on tentative observations in these areas, therefore, have been curtailed pending finalization of the same. in addition, no consnents are offered in regard to observations and statements for which consultants' reports are referenced reports which is incluced were with the not available detailed to the reviewer. A list of these coments.

D .

(V Reference is made to the Riverton PSCR review for generic comments (e.g., the need to supply applicable to theNRC EA. with all geologic logs available) which are equally SEISMOLOGY y .

Seismological aspects of the environmental assessment (EA) are addressed in the segment entitled " Slope stability and seismic risk" of section 2.2, Stabilization in Place, in the segment entitled "Riverton tailings site" of section 3.5, Surface and Subsurface features, and in Appendix B, section B.2.5.5, Slope stability / seismic risk. Tworeferences,(ANL,1979)and(SHB, 1983), are cited in the three parts of the EA described above.

The second reference, the Sergent. Hauskins & Beckwith (SHB) re90rt, contains information on the derivation of the MCE and ground motion to be expected at the Riverton site, however, discrepancies between this document and the EA make it impossible for the reviewer to substantiate adequately seismological parameters cited in the EA.

The Nviewer does not take issue with the acceleration value determined for the site but he is concerned with the manner in which the value is presented, indeed, the reviewer calculated peak acceleraticn values for the 5 hypothetical earthquakes presented in lable 1 of the SHB report and arrived at values that adequately approximrted e

table values. He utilized relationships developed by Joyner and Boore (1981). The confusion that exists between the information presented and Table 1 of the SHB report and the infomation presented in the EA needs to be resolved. The use of effective peak acceleration in lieu of other methods of determining peak acceleration, which are accepted in earthquake engineering, needs to be explained.

e ENCLOSURE TWO DRAh DETAILED COMMENTS NRC REV!EW OF -

RIVERTON DRAFT ENVIRONMENTAL ASSESSMENT k

k l

r1 .

}

, GENERAL The following data and references are requested for NRC review VATA NEEDS

1. All punp test data and plots.

2. All well completions and surveyed elevations.

3. Tabulated water level elevations.

4 Tabulated groundwater an'J surface water quality data with cation / anion balance.

\Q REFERENCES NEEDED temputer inputs ano outputs from the PHREEQE and DYNAMIX rnodels for all y*

simulations.

A computer listing of the DYNAMIX source code and data base files.

A computer listino of the latest version of the PHREEQE code used in these analyses and a listing of the new thermodynamic data referenced on page 0-137 of the DEA.

LBL,1984 " Hydrology and Geochemistry of the Uranium Hill Tailings Pile at Riverton, Wyoming" Draft report, Referenced on page 0-164 of the DEA.

W CSU (Colorado State University).1983a. " Characterization of Inactive Uranium Mill Tallings Sites: Riverton, Wyoming," draf t report prepared by the U.S. Dept. ofNew Albuquerque, Energy, UMTRA Project Office, Albuquerque Operations Office, Mexico.

FBOU(Ford, Bacon &DavisInc.),1983."EvaluationofAlternateDisposalSites for the Riverton Site, fremont County, Wyoming," unNblished preliminary draft prepared for the U.S. Dept. of Energy, UMTRA Project Office, Albuquerque Operations Office. Albuquerque, New Mexico.Section 1.1. 13. Fage 1; Section 2.1.3, 13, Page 20; and Section 3.2, 16. Page 73 4

tig1.1,13,Page1;Section2.1.3,13.Page20;andSection3.2, The first sentence in this section states that "most of the original mill structures and equipment" remain at the site.

On a recent site visit NRC staff noted that the main mill building is partly dismantled and most of the equipment is gone. The DEA should be revised to accurately describe the present conditions on the mill site.

Section 1.2, Item (3, Page 13 Given the assessments that the " consumption of borrow materials from the proposed local sources would have a ne cost of these resources in the area," (pg.146)andthat"Noneofthese gligible impact on the availability and alternatives would have an impact on other mineral resources in the area" (pg. *

(W

' 147), what is the basis for the conclusion in this section that "The Dry Cheyenne alternative would have a greater impact on ... mineral resources ..."

Section 2.2, 14 (Major construction Activities. Site preparation), Page 25 , , ,

v-This paragraph states that existing structures are to be demolished. The DEA b

should specify if the active sulfuric acid plant is included.

Section 4.4, Pages 146-148 The text indicates that gravel will be mined for cover material at gravel pit

• 2 next to the tailings pile and after mining will be not be restored so that the pit can be used to mine gravel in the future. Further, U.S. Geological Survey Topographic Maps show that the unconfined aquifer has been mined as a source of gravel in the past. This indicates that the unconfined aquifer may be used as a source of gravel in the future. However, the gravel between the pile and the Little Wind River has been in contact with contaminated groundwater and has probably become contaminatec through geochemical reactions with the groundwater. In addition, the gravel at gravel pit #2 may also have been contaminated in the past and will come in contact with contaminated groundwater as a result of dewatering activities associated with mining the gravel. In suunary if the gravel has become contaminated then contaminated cover material may be placed on top of the tailings pile, future gravel mined in the groundwater plume may be prohibited and any mines might have to treat water prior to discharge. The Environmental Assessment should indicate why contamination is or is not a concern either for using it to mine cover material or for mining the gravel between the pile and the Little Wind River as a source of gravel in the future. If there is a problem with the gravel the Environmental Assessment should indicate what measures must be taken to clean up the gravel or to assure that contaminated gravel will not be used in the

~

- ~"

DMFT future (for example by mining restrictions, or gravel washing af ter mining, etc.).

Section 4.4, 15, Pages 146-147 The last sentence states "... it is unlikely that any mineral or oil and gas resources beneath the sites would ever have a value sufficient to warrant development under the conditions that the Nuclear Regulatory Comission may place on development of these resources in order to ensure that the stabilization (sic) tailings were not disturbed by the mining operations."

There are two problems with this sentence. First, it assumes a negative cost / benefit for future mineral development. Second, it presents NRC license conditions, rather than the need to ensure stability of the pile, as a block to future mineral recovery. A suggested sentence is: "Any recovery of mineral, oil, or gas resources from beneath the sites would be governed by license (D

KJ conditions to prevent any disturbance of the stabilized tailings pile. If the cost be of avoiding disturbance of the pile is too high, resource recovery would precluded."

Saction 4.5.2, 14, Page 156

,f The first sentence in this paragraph states that "there would be a substantial decrease in the contamination of the unconfined aquifer". Migration of b cont 6minants into the unconfined aquifer, not the unconfined aquifer contamination, will be substantially reduced. This sentence should be reworded.

Section A.S. 12 Page A-8 The wording of the second sentence is not consistent with the EPA standard in 40CFR192.12(b)(1). It should be reworded to explain that. in cases where radon working levels are between 0.02 and 0.03WL, the government will make reasonable effort to reduce the working level to below 0.02WL.

RADIATION Section 1.1, 14. Psge 3

1) This paragraph identifies atmospheric releases of radon as the " principal hazard associated with the tailings". No other hazards are identified in this section, project Sumary. The ground water contamination under and adjacent to the site is significant; it should be mentioneo as a significant hazard in this paragraph or in a separate paragraph.

2) The third sentence implies that a threshold exists for radon health effects. This contradicts the concept of no threshold for radiation

. .5- l l

w.9 cts used in Appendix G. A suggested reworded sentence is: " Increased I

exposure to radon and its' decay products will increase the possibility of '

cancers in pert ans living and working near the pile."

Section 1.2. Table 1.1, Page 7 This table identifies public health impacts. Using the results in Table 4.1 (p.128)andTable4.2(p.132)itappearsthatthepublichealtaimpactforthe 1 SIP alternative includes the remedial action worker impact. The results for the Dry Cheyenne public health impact does not appear to include the impact '

from Table 4.1, " Excess Health Effects During Remedial Action". Also, the Dry Cheyenne public health effect over 1,000 years has not been multiplied by the 1,000 year term. Assuming that the numbers in Tables 4.1 and 4.2 are correct the entries for public health impacts should read "0.018 fatal cancers in 10

,Q years and 0.68 fatal cancers in 1,000 years" for the SIP alternative, and

• V "0.019 fatal cancers in 10 years and 1.31 fatal cancers in 1,000 years" for the Dry Cheyenne alternative.

Section 2.1.3, 15, Page 21 p.:

This paragraph should include inhalation of airborne radioactive particulates

• as a cause of cancer. It should also include contaminant migration into ground water as a natural process which increases the public health hazard.

Section 3.3, 14. Page 74 The DEA states that the predominent wind direction is " westerly." The PSCR states that the predominent wind direction is from the west (i.e., " easterly").

Section 3.8.1, Pages 100-101 This section should also give the background levels for radioactive particulates in air and radiation in surface and ground water.

Section 3.8.2, 11, Page 102 The Uranium and Th-230 concentrations in tailings should be given.

Section 3.9, tt, Page 106 Locations of residences relative to the tailings pile should be shown. This could be done by pictting residences on Figure 3.2 (p.67) as was done for St.

Stephens.

I'

4

)

DRAFT Section 4.1.1, Pages I?3-127 l

1 This section should explain how the radiological impacts from the remedial action are incremental to present conditions at the site. The exposures due to radon are several orders of magnitude greater than exposures fron other pathways because of the large radon flux which predominates over Aher source 4

terms prior to, during, and after remedial action. Although the radon released oue to construction activities is small, relative to the radon releaseo prior to remedial airborne action. it is still a dominant source tenn over direct gama and particulates.

i However, for completeness this section should discuss the relative increases in direct gama and airborne particulate source terms.

Increases of gamma exposure rates and airborne particulate concentrations S111 be significant when compared to their levels prior to renedial action.

Airborre particulates will increase from essentially zero cencentration to

• measurable levels of concentrations. There is a possibility of significant s concentrations in the case of adverse meteorological conditions and ineffectual *

' mitigative measures. Direct gama exposure rates will also increase 45 the pile is excavated and mounded. A qualitative discussion of the mechanisms causing incremental radiological impacts from radon, direct gama, and Airborne particulates should be added to the quantitative discussion on radon and direct gama impacts. Such a discussion would point out that any of the remedial , , ,

alternatives Would justify a finding of no significant impact from any of thf~

incremental radiological impacts. '

Section 4.1.2, Pages 127-130; and Section G.2.2, 14, Page G-17 This section should also estimate the number of excess genetic health effects as a quantifiable impact of the remedial action.

Section 4.1.4, Table 4.2, Page 132 The number of yearly excess health effects for the pry Cheyenne site is listed as 0.0013. This contradicts the value of 1.3 x 10' in the text.

Section G.2, 16, Page G-8 The DEA judges that the ingestion of plant or animal food products is insignificant because "the majority of land near the pile has no measurable contamination". However, the reference on which this conclusion is based (BFEC 1983a) found windblown contamination to the southeast up to 700 feet from the pile. It concludes that t predominent winds from the northwest caused this contamination and that a contamination in the agricultural land west of the pile has been mixeo into the soil during cultivation.

Radionuclides from the pile have entered the food pathway through this cultivated field immediately downwind of the pile. During remedial action the dusting of vegetation will increase. The DEA should quantify the health effect n'

. 7-DRAFT of eating during plant and remedial animal foods contaminated by tailings blown from the pile action. This contaminated food path may have a quantif%ble health effect because of the p-ocess components of the model.

greater retention values in the food and metabolic Section G.2, 13. Page G-10 '

The risk factor for excess fatal lung cancer, which in this EA is 100 x 10-6 deaths per person-ULM, is used for the general population tod for the remedial action worker.

The Evans et al (1981) reference, which gives the primary justification for using this risk factor, states that workers are at higher risk than the general population for equal exposures to radon daughters. A higher risk factor comparable to those recomended by UNSCEAR and used by the NRC should be applied to the remedial action worker.

Section G.3.1. 12 Pages G-13/15 Using of 210the parameters pCi/M in Table G.3.2 (page G 16) a radon flux from bare tailings

-sec is calculated. The reviewer, using the same parameters and )

Equation 3 in the NRC 84 reference, obtained a value of 250 pC1/M -sec. j

f. ,

Section G.3.1, 14, Page G-17

1) A wind speed of.2.7 meters per second is not an adequately conservative parameter for determining radon concentration on the pile. This wind speto should be adjusted to take credit for the calm conditions which prevail 30% of the time (see Table C.2.2 on page C-4). The appropriate wind speed would be 1.9 meters per second. 2) The methodology for calculating the radon concentration on the pile is questionable. The NRC is withholding judgement on this methodology until further review is completed. The 4.6 pCi/l obtained with this methodology may be reasonable based on data contained in the FBDU engineering assessment of August, 1981. However, the FBOU data is based on only two days of measurement and may not be adequate for judging the calculated radon concentration. Nightime concentrations of radon increased threefold during the FBDU measurements, presumably due to atomspheric inversion conditions and low wind speeds. Similar conditions may exist at times during the construction season and would significantly increase the on-pile radon concentration. (The FBDU measurements were made in October with temperatures ranging from 15'F to 50'F).

Section G.3.1. 15, Pages G-17/G-21 Because of it's relative importance in the determination of health effects from the remedial action, NRC is reserving judgement on the adequacy of radon

_ _ ~~

..' concentrations and working levels away from the pile. The calculational methodology used in the DEA is under review.

Section G.3.1, 15, Pages G-17/18 Evan's formula for calculating ingrowth of radon daughters should not be used for this case. The formula was developed for one-pass air conditions such as in a ventilated mine. In this case the air, being mixed by turbulent conditions, is a mixture of old and new air. The appropriate working level can be estimated by assuming a typicci working level ratio (WLR). Typical outdoor  ;

WLR's range from 20 to 50t.

Section G.3.1, 114 (Remedial action worker health effects from radon daughters g exposure), Page G 15

• The 20 percent redon daughter equilibrium ratio is not a conservative value.

This value is based on a single studyt typical outdoor equilibruim ratios have ranged from 20 to 50%. Furthermore, in the dusty conditions of remedial action on the pile, the ratio should increase as radon daughters attachment to particulates incr*3se. F-GEOTECHNICAL ENGINEERING Section B.2.5.3, 15, Page B-18 This section indicates that because of the extensive floodplain, the slightly elevated position of the tailings pile, and the protection offered by nearby road grades, flood waters from the Wind or Little Wind Rivers are not expected to reach the pile and no additional flood protection measures are deemed necessary. However, the flood analysis performed by Ford, Bacon and Davis indicates that flow from the Wind River woulo reach or approach the base of the tailings pile during a 1,000 year flood event. Studies by Sergent, Hauskins and Beckwith also indicate that the two rivers are becoming less stable with t1me and have the potential for meandering and migrating in the area of the tailings site. Partial or total inundation of the pile could adversely affect slope stability, and any infiltrating water through the pile could be retained by the slurry w=11 and cause a " bathtub" condition below or into the pile. The EA should include discussion related to the effects of flooding on the pile.

This discussion should include the effects on stability of the designed slope and the possibilit- of a " bathtub" effect within the confines of the slurry wall for flows ra ging frem the 100 year flood to the PMF.

n l

d

,. 9 Section 2.2,14 (Borrow Materials), Page 26 This section states that rock from borrow site 2 will be excavated and placed over the radon barrier cover of the tailings pile. Since this site is adjacent to the pile, there is a possibility that this material has been contaminated by seepage and may not be suitable for haulage road construction or for erosion protection of the pile. No data is presented to indicate that these materials are not contaminated. It is recorrrnended that the material in borrow site 2 be assessed for contamination.

2) Depending upon the distsnee of the borrow site from the pile, excavation of materials surrounding the pile could result in the pile becoming unstable. An analysis and/or discussion should be presented relative to the stability of the pile considering the planned adjacent excavation. This discussion should also include effects of -

Section 4.5.1,12_(potential future Stabilization miningPages in Place) in other areas surrounding the pile.

148-149 This section states that during remedial action, contaminated water resulting-from surface water run-off over contaminated materials and from activities such as the washing of equipment, would be collected and retair.ed on site in an evaporation pond, infiltration of contaminated water from the drainage ditchsis; and the pond could contaminate groundwater and the underlying soils. There ti no indication that the pond would be lined to prevent such infiltration. In addition, page B-9 (Site Restoration) contains no discussion of clean-up i

procedures for the contaminated soils from the drainage control areas and the pond once the water has been evaporated. it is recommended that this section include a discussion pertinent to the containment of the contaminated water and the clean-up of the contaminated soils.

Section B.2.1,14 (Groundwater Prottetion), Page B.8 kW inis section states that a 20-foot deep, 3-foot thick, bentonite slurry wall will encircle the tailings pile and extend 5 feet into bedrock. The i

stratigraphic column shown on pege 0-53 Indicates that the first bedrock unit below the aluvium is saturated sandstone, if the wall only extends 5 feet into this 12 to 14 foot saturated sandstone layer, and since this sandstone may be a relatively permeable fonnation, contaminants could migrate through the sandstone layer underneath the wall. The sandstone should be characterized to determine its permeability and a discussion should be included pertaining to the subsequent effectiveness of the slurry wall in preventing contaminant transport. In addition, the basis for proposing this design in lieu of other designs such as placing the tailings on an impermeable pad should be discussed.  :

i i

4

,_ _ ,, w-~

. Section B.2.5.6, 12, Pages B-20-21 This section states that the combined total settlements of the foundation materials, the pile, and the cover are estimated to be 2.1 feet and that most of this settlement will occur within the tailings pile. It also states that due to the sandy nature of the tailings, between 75% to 90% of the total settlement will occur prior to cover placement and that o1fferential settlement is, by engineering practice, considered to be 50% of total settlement. No basis is given for the conclusion that 75-90% of the total settlement will occur prior to cover placement. Table 8.2 in the Riverton PSCR (June, 1984) indicates the existance of areas of tailirgs that still have high degrees of saturation and would thus be susceptable to consolidation over a period of time exceeding the time for cover placement. Since settlement is time dependent, it is recornmended that an analysis showing the time for 75-90% settlement be made which would indicate if this is within the time frame for cover placement. The O basissettlement total for the statement that differential settlement is considered to be 50% of ~

is also requested.

k Section B.2.5.7, i'4, Page B-22 The last sentence in this section is unclear. Is something missing?

, GE0HYOROLOGY AND H(DROLOGY Section 2.2, 1b_(Description of Final Conditions), Page 29 The text states that "The underground slurry wall (3 feet wide and 20 feet deep) would extend 5 feet into the subsurface bedrock ano would encircle- the perimeter of the stabilizeo pile". The problem is that bedrock is not defined.

If the slurry wall 1s placed in the unconfineo aquifer or the top of the sand beneath the unconfined aquifer the slurry wall will not be effective, since the

' water will simply flow underneath it. The text should explain into what geologic material the slurry wall is to be completed.

Section 2.3, Page 38 Very little geohydrologic data exist for the Dry Cheyenne Site in the Environmental Assessment. Should this become the preferred site in the future more data will be 'needed.

Section 3.1, Page 70 Figure 3.4 on page 70 shows that Borrow Site 2 will be located directly adjacent to the tailings pile. Will mining in Borrow Site 2 force the irrigation ditch to be moved?

n'

Section 3.6.1, Figure 3.7, Page 83 Inspection of aerial photographs and topographic maps strongly suggest that the location of the irrigation ditches are plotted incorrectly in Figure 3.7 The irrigation ditch that enters the site from the north appears to be misplotted.

Rather than tieing into the main ditch in the middle of Section 4 it appears to connect with the main ditch in the west 1/2 of Section 4 from there diagonally and site, under beforethe formerthe entering C&WsiteRail Road in the tracks south and 1/4 of carallel Section 4 to the road north of the l Section 3.6.2, Pages 87-95, and_ Appendix 0 The confined aquifer needs better definit 4on in the text and figures. There is confusion in data presentation as to what is meant by the confined aquifer.

For example, in some situations d6ta from 200-300 depth are used to make -

O)

( incorrect statements about depths of 40-60 feet. Near surface aquifers (40-60 feet in depth) which may only have as much a 15 feet of confining material should be distinguished from deeper aquifers, which may have more than 200 feet of confining material.

the confined aquifer. The following 5 coments deal with further questions on y

Section D.2.5.2,17, Page 0106 This paragraph states that "An upward '

vertical hydraulic gradient protects the confined aquifer from future contamination due to comunication with the unconfined aquifer". No where in the text nr tables is evidence for this vertical hydraulic gradient found.

- Section 0.2 Neither the groundwater section of the text or the Appendix contain direction of groundwater movement data for the near surface confined aquifer. While, the text does make statements about the direction of groundwater movement in the deeper sands at 300-400 foot cepths, this data is unlikely to apply to the near surface sand at 40-60 foot depths. Data should be supplied on the direction of groundwater movement for near surface sands in the confined aquifer since, if there is any contamination in the confined aquifer it should show up in these sanos first. Furthennore, if the bentonite barrier wall works, this could become the preferred groundwater movement pathway from the restored tailings.

Section 0.2.3.?,16, Page D-56 and Section 3.6.2,16, Page 89 The Draft Environmental Assessment states that about 12 to 24 feet of unsaturated siltstones and shales separate the confined and unconfined system. This appears to be highly unlikely, since these units have probably had a confined saturated aquifer beneath them and a saturated unconfined aquifer above them for thousands of years. What is more likely is that they are too impermeable to readily yield water, even in core samples. Data should be provided to support this statement since it is cited as support for noncomunication between the unconfined and confined system.

?

,.- Section 0.2.4.1, Page 092 Table D.2.2C shows a zone of extensive fracturing in drill holes 3 and 4 at approximately 45 feet in depth, with hydraulic conductivities up to 40,600 ft./ day. This is at the depth of the near surface sands in the confined aquifer and indicates that fracturing and maybe faulting exist at the tailings site. Further, the packer tests indicate that these fractures can transport water very rapidly and therefore could be a fast mechanism for contaminants to move from the tailings pile.

Further information and discussions about these 4

fractures should be provideo along with the locations of drill holes 3 and Section 0.2.3.2,14 Page D54 This paraoraph states that "Interbedded layers and lenses of shale, siltstone, and mudstone confine the ground water in the sandstone beds, but the entire sequence behaves as a single G aquifer in response to hng-term stresses." This implies that the deep

• sands in the confined aqu.'fer ar.e hydraulically connected over more than V 200 feet of confining materiel to the near surface sands of the confined aquifer. How does this connection occur? This should be explained, since it indicates that contamination that enters the near surface sands of the confined zone would be comunicated to the deeper sands of the confined zone, f

Section L6.2. , Pages 89-94

1) There is no discussion of vertical hydraulic gradients that exist within particular aquifers or between the unconfined unit and the confined system.

i This is necessary to determine the amount of comunication betweer. the two aquifers.

2) The direction of groundwater flow in the confined aquifer has not been adequately determined. For example, on page 94 (Paragraph 11, Section 3.6.2) it is stated that the city wells are located to the north and northeast upgradient of the site. This contradicts the third paragraph on page 89 which states that the hydraulic gradient has been reversed to the northeast, thereby Hing areas northeast of the site downgradient.

Section 4.5.1, 14-7, Pages 149-151 and Section D.1.2.1, Pages D-14 to 0-15 Based on our experience and an expected PMF magnitude of over 400,000 cts in the Wind River, it is unlikely that the 1000-year and 10,000-year floods will have magnitudes of 16,370 cfs and 18,350 cfs, respectively. It appears that a direct extrapolation of historic flood data was used to estimate the flood magnitudes.

3 The NRC staff believes that such estimates are unreliable, in view of the limited data base which was used. Accordingly, the Environmental Assessment should discuss the uncertainties inherent in statistical flood estimates. The NRC staff believes that statistically-derived flood estimates for recurrence intervals of 1000 to 10,000 years cannot be reasonably computed using only 70 years of historic flow data. We believe that statistical 4

.. estimates of floods of such recurrence intervals are inappropriate using such limited data. At best, we believe that floods with recurrence intervals of 100-200 years could be estimated using this data, but only if a range of flows associated with appropriate confidence intervals for these recurrence intervals are presented.

Recognizing that detailed water surface profiles are being developed for the RAP, this data should be provided in the Environmentcl Assessment to determine the extent and frequency of flooding for a range of flood events (from a 100-yearflooduptoandincludingthePMF). This information is needed to assess stability and groundwater contamination concerns. Please make the necessary revision to the Environmental Assess, ment.

Sec tion 4.5.2,12, Page 155

(]'

The Draft Environmental Statement says that " clean up of the wind-blown tailings and other contaminated areas could require some dewatering of the unconfined aquifer. This dewatering would be minimal and for a short time, and the water removed would be used for compnction and dust control or evaporated."

The problem with this statement is that aewatering Borrow Site 2 and the area,..

for the slurry wall could generate a lot of water from underneath the tailin9F How much water will be produced? Will the contaminated water only be used on' the tailings or will it contaminate clean areas? Will the contaminated water be used to compact the cover material and if so will that contaminate the cover material? Where will the water be evaporated? If it is put in a pit in the gravel it will reenter the aquifer. If it is evaporated in a lined pit what will be done with the contaminated solids after the pile is restored?

Section 4.5.2, 15, Page 156, and Section D.2.7.4 The text states that "With the substantial decrease in the generation and migration of contamination from the tailings pile, the natural movement and discharge of the ground water in the unconfined aquifer would reduce the concentrations of the contaminants to background levels. The time required for this natural flushing action has not yet been estimated, but it is likely that it would take at least several decades." This will occur if the tailings are stabilized in place or if they are moved. However, the text makes no indication if the. groundwater in the unconfined aquifer will be restored or what measures will be taken to prevent the public from using the water without treatment or if any wells will be contaminated in the future. Since, this problem exists in different degrees of severity for all of the 3 alternatives it should be discussed, n'

,. g, ,,,

Section 4.5.2, 14, Page 156 It is stated that emplacement of the slurry wall would result in a lowering of the water table.- This is not supported in the document.

Section B-2, Preferred Alternative - Stabilization in Place Page B-4 Information should be provided regardin used for the erosion protection cover, gincluding expectedinformation sources ofon rock that rock could sizes be and rock durability. This will assist in the determination of whether adequate rock is economically available, and whether the pile can be adequately stabilized at a reasonable cost.

Section D.1.3.1, Pages D-16 to 0-23 bv -

1) Section 0.1.3.1 does not contain information on surface water quality down-stream of the pile in the Little Wind R'.ver. Surface water samples simultaneously taken upstream and downstream of the site during a time of low surface water flow would provide a measure on how the contamination in the 7 shallow gravel is effecting the Little Wind River.

2) Section D.1.3.1 does not contain any information on the surface water discharge at the time of surface water quality sampling. This is important because during high flows, streams contain high suspended solids and at times of low flows, streams contain high dissolved solids. This is because during times of high flows, stream flow is largely due to surface runoff, whereas

' during times of low flows, stream flow is largely due to groundwater inflow.

In addition, if flows are known at the time of sampling the ameunt of contaminants leaving the site by surface water pathways can be calculated.

Surface water flows should be presented in the text for the time of sampling to determine if the samples represent grour.dwater contributions or have been diluted by surface water runoff.

3) It is not clear where the gaging stations on the Little Wind River and the Wind River are located.- This information is needed to ascertain where the surface water quality sampling was done, upstream or downstream of the tailings pile. If possible this information shoulo be located on a map.

Section 0.1.3.1. 14, Page 0-21 The standing water referred to in the statement " Contamination of standing water at the base of a dike west of the tailings pile is most likely due to runoff from the tailings or to leaching of windblown tailings in the area.", is not included in Table 0.1.9.

..- - 15 Section D.2 DRMT The geologic investigation of the site is inadequate to properly assess the hydrogeologic character of the area and extent of groundwater contamination.

This lack of geologic infonnation is brought out in the following two coments:

Section D.2.2.2. Figure D.2.1, Page D-29 Figure 0.2.1 should specify the depths and aquifer sampled by each well.

Section D.2.3.1, 13, Page D-52 The stratigraphic section described on this page only refers to a small area (perhaps 50 feet across) in the northwest portion of the site. Based on the amount of stratigraphic variation in just this small area (Figure D.2.10), geologic cross-sections should be cor-tructed suspected. across the entire site and into areas where contamination is p .

Section D.2 The grour.dwater flow system in the unconfined aquifer has not been adequately defined for seasonal variations in hydraulic gradients, hydraulic properties, and the extent of groo dwater contamination. The following seven comments ar F

, in regard to this.

Section D.2.3,2,13, Page D-54 The wells used to construct Figure D.2.11 should be located along with the. corresponding water table elevations in those wells.

The water table elevations have not been determined in the areas to the north and west of the site. In addition, these elevations should be mapped seasonally to determine any change in hydraulic gradients.

Section D.2.3.2. Table D.2.3 There is no justification for the methods of 4 analysis used in arriving at the transmissivity and specific yield results for the unconfined aquifer presented in Table D.2.3 This is because the methods are normally only applied to confined aquifers and not unconfined aquifers, in addition, there may be problems with partial penetration effects in the analysis. Finally, in regards to Tables D.2.3 and D.2.4, the semi-log recovery method cannot be used to determine storage coefficient (Bouwer,1978).

Section 0.2.4, 15. Page D-89 Since the pumping well (RVT-112-83) 1s completed in both the gravel and the sandstone, the discharge comes from both units. However the monitor wells have been completed into either the alluvium or the underlying sand, but not in both. Since the monitor wells are very close to the pumped well (less than twice the thickness of the aquifer away) this means that partial penetration effects will be significant. Therefore the values detennineo from this pump test need to be justified or they should not be used to to predict environmental effects in the unconfined aquifer, v'

4 a.- Section 0.2.4.1, 12, Page D-85 This paragraph states that " variation in calculated hydraulic conductivities at the observation wells are a function of the low variable pumping rate and associated small drawdowns as well as the difference in well completions between the observation wells." This : atement, plus the wide variation in values in Table 0.2.17 indicates that these results should not be used to predict environmental effects.

Section D 2.4.1,15, Page D-90 A groundwater travel time of two years from the pile to the Little Wind River is calculated on the basis of the pump test results and tne existing hydraulic gracient. This corresponds to a seepage velocity of 410 meters per year. However, a groundwater velocity of 20 meters per year is used (based on the plume location) as input to the computer simulations described later in the text (page 0-129). This large discrepancy should be addressed.

(N V Section 0.2.4.2, 11, Page 0-93 Based on the chemical analyses presented in this section, it appears that other contaminants have been transported into the unconfined aquifer besides sulfate, molybdenum, and uranium. The text should discuss Other possible contaminants.

Section D.2.4.2, 12 Page D-95 In regards to the plume delineation maps f (Figures 0.2.13, 0.2.14. D.2.15), values plotted on the map do not all

• correspond to the values in Table D.2.9.

Section 0.2 The preferred alternative for the remedial action at this site is based on the

- -premise that there is no hyoraulic connection between the unconfined aquifer and the confined system below. This lack of connection is not adequately

' determined in terms of geochemical evidence or hydrologic evidence. The following three comments address this issue.

W Section D.2.4.2, 11. Page.D-93 There does seem to be evidence of contamination of the confined aquifer as indicated by the chemical

' analyses listed on Table 0.2.11 as compared to the eight analyses referred to as " background" in Table D.2.12. This evidence is as follows:

1. Three of the four 00E wells contain higher concentrations of iron than any of the background wells. These three concentrations are at least 907, higher than the Wyoming groundwater standard for domestic use.

2. The four 00E wells have concentrations of molybdenum that are between 2 and 110 times the highest background concentration except for one anomalous background well.

3. At least two of the DOE wells show nickel concentrations higher than any background wells. One of these 00E wells is at least 2.5 times higher than background.

i

4 Three of the four DOE wells show calcium concentrations higher than any background wells. One of these DOE wells is over 3 times the highest background concentration.

5. Two of the DOE wells are higher than any background well in terms of sulfate, one of which is over 2 times the highest background concentration. All four wells are higher than the second highest bachground well. All four wells are also above the Wyoming groundwater standard for domestic use.

6.

At least one of the DOE wells is 3 times higher in uranium than the hignest background concentration.

7.

All four DOE levels of TOS are higher than background concentrations, two of which are more than twice the highest background level. All four DOE wells are above the Wyoming groundwater standard for domestic use.

8. Two of the four DOE wells are higher in magnesium than the highest background conc 67tration. This is true for potassium also.

(V 9.

One of the DOE wells is 66 percent higher in nitrate than the highest background concentration. Three of the four DOE wells are above the Wyoming- groundwater standard for domestic use.

10.

All four DOE wells are between 2 and 81 times the highest background concentration in terms if manganese. All four wells are between 3 and 114 times higher than the Wyoming groundwater standard for F donestic use. '

11. One of the DOE wells was at least 18 times higher in the radionuclide lead-210 than the highest background level. This DOE well level is also above drinking the USPHS.USEPA Maximum Pennissable Concentration for water. The same DOE well was over 9 times the Wyomin groundwater standard in terms of gross alpha for domestic use.g Section 0.2.5.1, 13, page 0-100 The possibility exists that local wells provide pathways for contaminants to travel from the unconfined aquifer to the confined system. The completions of these wells should be checked, especially in regards to gravel packs and locations of screened intervals.

W Section 0.2.5.2.,17, Page 0-106 Contrary to this paragraph there is no evidence presented in the report that demonstrates the presence of an upward gradient from the confined aquifer to unconfinea aqu1fer. On the contrary, if the siltstones and shales that underlie the unconfined aquifer are partially saturated as is stated on page 0-52, then the pressure head within the siltstones and shales is less than zero. This indicates that a downward vertical hydraulic gradient trom the unconfined aquifer exists. The data that support an upward gradient from the confined aquifer should be presented.

18 -

e.

e, W

Section 0.2 ',1 . 4 The magnitude and direction of groundwater flow in the confined aquifer as well as the extent of contamination is not adequately determined. The following six comments make reference to this.

Section 0.2.4.1., 15. Page 0-90 No data is presented to support the statement that the first sandstone aquifer below the unconfined system is confined or that leakage is coming from an underlying, confining layer.

This is because the analysis of leakage does not indicate the source of the leakage. Therefore additional data should be presented to demonstrate the direction of leakage since this could indicate hydraulic connection with the unconfined aquifer.

Section 0.2.4.1.,18, Page 0-91 Storage coefficient cannot be calculated

p from recovery data (Bouwer,1978). Therefore these parameters should not

• be used to assess environmental impacts.

Section 0.2.5.2,14, Page 0-104 According to this paragraph, flow in the -

confined system is towards the northeast in the direction of the c'ity well

' field. If this is the case, then those water samples referred to-as "upgradient or background" in Table 0.2.12 may actually be downgradient F' -

the confined aquifer. An equipotential map in the confined aquifer is needed to evaluate possible directions of contaminant migration. The effects of the Western Nuclear well should obviously be included in this equipotential map.

Section 0.2.5.2, 14, Page 0-104 Two domestic wells should be investigated further to assess possible contamination of the confined aquifer from the tailings site. These wells (from Table 0.2.12) are referred to as Raphael Norse and Schlotter. The fonner is relatively high in sulfate and TOS while the latter is extremely high in molybdenum and vanadium.

Section 0.2.5.1, Table 0.2.2.1, Page 0-101 This table does not contain the location of the Raphael Norse well.

Section 0.2.3.4, Table 0.2.11, Page 0-75 Well number RVT-113-83 is extremely high in gross alpha and lead-210. This well is listed as being completed in the confined aquifer in this table, however according to Table 0.2.19.it is completed in the sandstone portion of the unconfined

' aquifer. This discrepancy should be resolved and the well should be investigated further due to the very high concentrations.

I Section 0.2 It 'll . =

The groundwater modelling that is presented does not include details that address the uncertainty of input parameters, boundary conditions, initial 4

. 4

A 4 -

DMFT conditions, and input data. The following five comments pertain to these concerns.-

Section D.2.6.5,11, Page D-115 The recharge controls and therefore the unconfined flow regime are likely to vary seasonally, This possibility should be discussed in supporting the assumption of a steady state (hydraulic heads are unchanged) flow regime. In addition, the period when the pile was active and water was being discharged with the tailings probably had a large influence on the present location and relative concentrations of contaminants probably noch different from whatinthey that are the now.

water table elevations were This period would certainly represent a significant deviation " rom " steady state."

Section D.2.6.6, 13 Page D-121 A discussion of input parameters and how they had to be modified during calibration of the TRUST .nodel would clarify this portion of the text. Examples of the parameters include

('N

\> moisture-content characteristic curves, unsaturated hydraulic conductivity curves, initial conditions and the boundary condition at the top of the tailings. Once calibration is completed, a sensitivity analysis should be perfortned to insure that uncertain input parameters do not affect the output to a large extent.

Section 0.2.6.6,114, Page D-128 Due to seasonal variations in snow melt,' .

irrigation and precipitation, water table elevations can change by six feet (p. D-54). This may very well explain the zone of leaching in the tailings referred to in this paragraph.

Section D.2.6.6,115, Page D-129 The fluid fluxes of 1 meter per year and 1 cm per year are most likely based on calculation of flux from modelled output of vertical head distributions in conjunction with unsaturated It would be helpful to have this hydraulic conductivity) values.information (h(2), K(pIn )regards to support thes to the groundwater velocity computed on p. D-90 of 1343 feet per year or 410 meters per year, the value input into the model (20 meters per year) is less than one-twentieth of the former value. This discrepancy should be expla1ned. The bsMs for the selection of the other input parameters should be presented as well.

Section D.2.6.6,121, Page D-139 The oscillations mentioned may be due to instability of the model. This might be remedied by decreasing the time step, increasing the grid spacing, or increasing the dispersivity. Since the dispersivity is an input paramter, it should be determined independently and not be changed. This would help to insure uniqueness in the computed solution.

Section D.2.2.2, Figures D.2.4 and 0.2.8, Pages D-40 to D-47 Neither Figures D.2.8 or D.2.4 contain depths and geologic units monitored by the Colorado State University and Sandia National Lab wells. This should be included if data from these wells is to be used in the Environmental Assessment.

Section 0.2.5.2, Figure 0.2.16, Page D107 Figure D.2.1.6 appears to have the registered wells near Riverton plotted incorrectly according to the convention of 1/4 see of the 1/4 sec of the Section. '

Section D.2.6, Pages D-108 to D-142 o -

1) The slurry wall is planned to be made of bentonite clay, which usually swells when it gets wet. Depending on the composition of the water, bentonite may swell or shrink. If this were to happen the slurry wall would not work.

Since the water chemistry inside the restored tailings will be very different' from that outside the slurry wall, two questions need to be answered; how wilk the water chemistry outside the slurry wall affect the slurry wall; how will

• the water chemistry inside the tailings effect the slurry wall?

2)None of the models or flow nets presented in this chapter take into account the effect that the irrigation ditches in and around the site have on the groundwater flow in the unconfined aquifer. Because the irrigation ditches are built on top of the unconfined aquifer; when they are flowing they are probably the major contributor of groundwater to the aquifer. Since the area receives very little precipitation during ths year this could be the major source of recharge for the unconfined aquifer around the tailings site. Indeed, this may explain the water table rise in the tailings in the summer months. Further, since the irrigation ditches bound the site on the north, the east and the west, they may act as groundwater flow barriers forcing the contaminant plume to flow towards the Little Wind River. Because the irrigation ditches may be a major factor in predicting the future perfonnance of any restoration alternatives the text should address the effect that irrigation ditches have on groundwater flow at the tailings site.

, Section D.2.7.1, 13, Page 0-144 There is no evidence or data presented that supports the conclusion that the water table would be lowered in the area beneath the pile once the slurry wall 1s in place. Also, there is no discussion as to how Figure D.2.27 (page D-145) was developed.

If

,r

1 GEOCHEMISTRY Section 3.6.2, Pages 87-95, and Appendix D The Riverton Draft Environmental Assessment fails to adequately identify the extent of contamination in the confined aquifer.

no evidence of contamination in the confined aquifer.The text states However, that there is data presented in the text indicates contcmination in the first confined sandstone aquifer located at a depth of acproximately 55 feet. The following three coments further detail this concern.

Section D.2.3.2.,16, Pages 0-56 and D-57 Tritium data presented on page D-57 indicates high values in the alluvial aquifer and relatively low values in domestic wells completed in the confined aquifer. The lower tritium values for the donestic wells were obtained from a sandstone p aquifer approximately 320 feet deep (White and Delany,1982 . This -

evidence is used to support the conclusion that there is no) connection V between the shallow and confined aquifers. However, because tritium was only produced in significant qnntities from bomb tests in 1954, the tritium indicator can be applied only to the last 30 years. However, it cannot be assumed that the tritium data from 320 feet is representative of 4 the entire confined aquifer. Therefore, the tritium data in its presentF-fonnat and thedoes not demonstrate confined aquifer. non-comunication between the shallow aquifef- .

The tritium data and discussions should be presented analysis. in a manner that more clearly demonstrates the relevance of this Section D.2.3.4, Table 0.2.11. Page 0 74 Dissolved constituents in the confined aquifer are elevated above background concentrations. Table D.

2.11 indicates concentrations much higher than background for the constituents Mo(1.1 mg/1), Ni (0.1 mg/1), Ca(260 mg/1), 50, (747 mg/1) and TDS (1450 mg/1) for well RVT-106-83 in the first confineo landstone aquifer.

The TDS is a factor of three higher than the Wyoming domestic groundwater standara as shown on page D-62.

Section D.2.3.4 Table 0.2.13, Page D-78 Table D.2.13 indicates that maximum pH values range between 7.50 and 9.55 in downgradient domestic wells within the confined zone. These wells are 200 to 250 feet deep and contain alkaline to very alkaline waters. High pH's suggest the confined aquifer has been contaminated.

i The mobility of uranium and other heavy metals is greatly enhanced under alkaline conditions and contaminant migration may be a significant problem in the confined aquifer. These high pH's may indicate contamination by leachate from the carbonate tailings. The occurrence of high pH's should be addressed in order to deteraine the extent of contamination.

_. ._ J

l Section 3.6.2, Pages 87-95, and Appendix 0 The text does not adequately characterize all the potential contaminants from the mill tailings. First, two types of 10 aching processes were used at Riverton, resulting in acidic mill tailings and carbonate tailings. However, the text does not adequately characterize the carbonate ta111ngs. Information concerning the amounts and locations of the tailings are needed, The pore water chemical data and geochemical modeling do not represent the carbonate tallings. Because the potential for migration may be increased under alkaline conditions, the carbonate tailings should be addretsed in detail.

Second, solvent extraction, which uses various organic solvents, was used at Riverton. Analyses are needed for at least total organic carbon (TOC) and total organichalogen(T0X)toindicateiforganicsaresignificantcontaminants.

O Third, the text describes solid radium concentrations in soils and sediments at -

Riverton. Although geochemical data for other potential heavy matal d contamirants in soils and sediments are available, the text does not address this. Heavy metals may provide a future source of contamination and may threaten human health in the future. The Environmental Assessment should discuss the significance of solid contamination by heavy metals.

,~

Sections D.2.2 and 0.2.3 Questionable analytical procedures that may affect the modeling results are discussed in the following two coments.

Section 0.2.2.2, 14, Page 0-30 The text states that pH, Eh, and specific conductance were measured within eight hours of collecting the samples.

In general, these parameters must be measured immediately and exclude air contamination and gas loss to obtain accurate results. A statement discussing the uncertainites of this data should be included in the Final Environmental Assessment.

Section 0.2.3.4, Table 0.2.8, Pages 0-69 to D-70 The relationship between Eh and dissolved oxygen is not consistent with expected results. Higher dissolved oxygen in the water suggests a higher oxidation potential (Eh).

The water quality data do not show this relationship consistently. For example, for well RB-3, dissolved oxygen is 3.97 mg/l and Eh is .037 volts. For well RA-2, dissolved oxygen is 0.84 mg/l and Eh is 0.500 volts. Several other wells show this inconsistency. Inaccurate Eh measurements may be siginficant because Eh is an important input variable for the modeling calculations.

9

Section D 2.3.4, Pages D.58 to 0-84, and Figure D.2.12, Page D-65 The following 5 coments should be addressed to clarify Figure D.2.12,

1. Figure D.2.12 shows extremely high concentrations of heavy metals in soil water to a depth of 4 meters. The text does not explain or evaluate the high concentrations.

2. Other important metals are missing from the profile, including arsenic and cadmium.

3. It is unclear wnere the location of the tailings pile is relative to the vertical profile.

4 Does ' soil water' represent the tailings pore water?

5. What is source of the data for Figure D.2.12? In particular is the data from one boring or a composite?

(

Section D.2.3.4, 12, Page D-64 and Table 0.2.7, Page D-66 and D-67

1) What are the detection limits for Fe, Cr, Se, Zn and Cu? Detection limits should be given because they can vary significantly with the analytical method that was used. This is particularly important when the detection limits are higher than water quality standards. F

2) A heading for cadmium (Cd) is listed in Table 0.2.7 without any data for' cadmium. Marcos and aush (1983) analyzed for Cd in the tailings pore waters. The Environmental Assessment should present concentrations of Cd for the unsaturated tailings pore waters.

3) The source (s) of the data for Table D.2.7 should be referenced.

Section 0.2.3.4, Table D.2.9 , Page D-71 What are the detection limits for Fe, As, Cd, Cr, Pb, Sc, and V?

!ection 0.2.4.2, 11 to 3. Pages D-93 to D-94 The conclusions regarding the tource of contamination in the groundwater appears to be based on geochemical modeling results. These ' conclusions' need to (II)be validated redox couplewith field is the evidence.

primary The text states that the Mn (IV) - Fe Eh control, ho evidence is given to support this conclusion. 'Unless an analysis was made for the different valance states of Fe and Mn, this couple cannot be determined. This redox couple is not consistent with the statement of page D-127 which indicates Eh was controlled by the Fe(II)/Fe(III) equilibrium, i n' -

.- Section D.2.5.1, Figure D.2.15, Page D-98 Molybdenum concentrations in the Riverton shallow aquifer are shown by contour lines without the well locations and actual data. Because contours are intrepreted from the data, Figure D.2.15 should contain the molybdenum data and well locations.

Section D.2.6 There are several limitations and assumptions of the geochemical modeling which are not discussed in the text. The following five conments illustrate this further.

Section D.2.6.1, 13, Page D-109 The text states that computer models can greatly facilitate site characterization efforts and remedial action -

(S impact analyses. It is not clear how the results will be used to support

() a remedial action decision.

Section 2.6.3, tl, Page D-110 The DYNAMIX code is still under development according to the author. The code has not been documented or validated with known field results. The text does not address the many y uncertainties in the code and how these uncertainities may affect the '

results. Thermodynamic equilibrium at each node along the flow path is one of the major assumptions of the model. This assumption is highly questionable because the contaminated groundwater is not likely to be at thermodynamic equilibrium. However, the text does not discuss this assumption in detail. Other assumptions of the code cannot be assessed at this time because specific infontation on the code is not available to us.

Section D 2.6.1,13, p. D-109 It will be difficult to support a remedial action decision based on geochemical computer modeling, due to the complexities of the natural physical system and the uncertainities in the codes and data. For example, the whole subject of Eh control in natural systems and its treatment in computer models such as PHREEQE and DYNAMIX has been called into ,uestion by the finding that equilibrium among redox couples is not reached in nature (Lindberg and Runnells, 1984).

Sections D.2.6.3 and 0.2.6.4, Pages D-110 to D-115 The importance of redox reactions should be evaluated in the context of the proposed remedial action. '

Section D.2.6.6, 15 Page D-117 The sulfate plume is assumed to represent the full extent of contamination.

The text does not provioe evidence to support this assumption. As shown by the plume maps, uranium anc molybdenum (and other reoox-sensitive metals) have e geochemical behavior significantly different than sulfate.

7.-

,.- GE01.0GY AND GEOMORPHOLOGY DRgy '

Section 2.3.2, pisporal at Ory Cheyenne. Long-term Stability, Page 51 It is premature ;o conclude without supporting data that the Dry Cheyene site would not be subject to flooding. In addition, risks to the site from fluvial erosion (streamdissectionandheadwardextension)shouldbeconsidered,given that: (1) the site appears to be on a drainage divide, at or near the head of several ephemeral streams, and (2) as observed on pg 80, the area around the Dry Cheyenne site is characterized by highly dissected terrain.

Section 3.5, 14, Page 77 O It is stated that " floodplain materials" consist of " eolian sand" -

k (alternatively alluded to as " soil" on page 0-50) and coarse grained alluvium."

What is the basis for the genetic classification of these deposits and what are their relative ages? What paleoenvironmental interpretation is placed on tha-deposits? Projections on the future geologic stability of the site should address these issues, y

Figure 3.6, Page 78

- Given the importance of accurately defining thz stratigraphy in the designated site area, simplified and generalized stratigraphic columns alone are not adequate. It should be noted that the " simplified" stratigraphic column on this page indicates that surficial gravels and cobbles overly an uppermost claystone 4

unit of the Wind River Formation, whereas the " generalized" stratigraphic i column on page 0-51, Volume 2, does not show this stratum and the gravel deposits are depicted as directly overlying an uppermost sandstone unit.

Detailed stratigraphic columns and cross-sections based on specific well and core logs are necessary to define lithofacies changes and identify potential pathways for contaminant migration.

Section 3.6.2, 13, Page 88 The verbal descriptiens of the lithologic variability in the designateo site

' area are not specific enough to adequately define the same. Detailed (basedon

! specific well or core log data) rather than generalized " aggregate" cross sections are necessary to spatially define the lithofacies changes in the area and thus identify potential pathways for contaminant migration. At least one north-south and one east-west cross section extending beyond the boundaries of the site should be provided. The depth of these sections should be determined by the need to characterize the hydrogeologic conditions at the site.

.' l Section 3.6.2, 15 Page 89 ERMT With regard to the siltstones and shales between the unconfined and confined j aquifers, what is the reference for the assertion that "co es from these strata revealed no evidence of fracturing the 12 to 24-foot interval." The reference

' cited at the end of the paragra not make any such observation. ph containing this stattment (FBOU,1981) does Section 3.13. Item #2, Page 115 i

lt is stated that public concern has been raised regarding the ability of the slurry wall to prevent contaminant migration "if the geological structures under the pile are sloping or if the bedrock consists of sandstone." This

' concern is not addressed in this EA, despite the statement to the contrary on

' pg.116, given that neither the structural attitude nor the lithology of the -

units under the proposed wall are adequately defined. Though it is asserted (pg. 77) by reference to FB00,1981, that the stratigraphic units under the pile are "nearly horizontal," the actual dips (ma various strata and how these data were obtainedinferences (gnitude and fromdirection) the regionalof the

' dip? dip measurements in cores? dipmeterdata?)shouldbestated. in regard to the bedrock lithology, detailed cross sections and petrologic-mineralogic At descriptions of the strata (e.g., degree of lithification, composition of #

authigenic and allogenic constituents) are necessary to define potential pathways for contaminant migration.

Section 4.5.1. ,15, Page 150 i

i It is difficult to evaluate the merits of the FBOU analysis said to indicate, that simultar.eous flooding of the Wind and Little Wind Rivers would not affect the site. It appears that such a conclusion is premature given that, as stated

> on page 115, the potential effects of a PMF in both or either of the rivers are still being assessed.

1 Section 4.5.1, Page 151 It is concluded in this section that flood flows from the Wind and Little Wind Rivers would not reach the stabilized pile and that stabilization in place would have no impact on surface water cuality. These conclusions are premature and incongruous in relation to: (1) the stated uncertainties as to whether a 1,000 year flood in the Wind River woulo reach or approach the base of the pile (pg.150), (2) the admitted potential for the channels of the Wind and Little Wino Rivers to move toward the site (pp. 150 & 151), and 3) the incomplete assessment of the potential effects on the site of a PMF in both or either river.

l 2

-_ f

The ergument is made that engineered measures to safeguard nearby developments would, in part, prevent floods from reaching the site. This contention is problematic. It should be kept in mind that current human structures and

-facilities may not be present or important 200 to 1,000 years into the future.

Indeed, nearby flood protection measures could also threaten the stabilized pile by directing flood waters toward the pile. It is unclear why benefits to the site from future human activities should be taken to be definable when, according to the rationale used in RAP's, hazards to the site from the future works of man are not definable. Therfore the assumption of coincidental flood protection should not be made. If credit for coincidental flood protection is taken, the remedial action design for stabilization of the pile may be inadequate for meeting the EPA standaros in 40 CFR Part 192.

Section 4.5.1, Wetlands Determination Page 154

% In light of the stated fact that "The Riverton tailings site is on the floodplain of the Wind River" (pg. D-50), the determination that "...no floodplains or wetlands would be impacted by either of the remedial action alternatives" is paradoxical. If different definitions of " flood plain" are being used, this should be made clear.

,a::.

Section B.3.5, 12, Page B-36 The statement is made that "The Dry Cheyenne relocation 61ternative would incorporate the same measures to assure long-term stability against water ...

erosion ... as discussed for stabilization in place." However, the measures referred to are only concerned with erosion from runoff on the surface of the pile and, given that aojacent ephemeral streams may pose an erosion hazard for the Dry Cheyenne site, such measures cannot be said to be sufficient for the latter pending further characterization of this hazard.

W Section D.2.3.1, 13. Page 0-52 Whereas it is indicated in this section that cores taken between the ta111ngs pile and the inactive mill show 12 feet of siltstones and shales underlying the uppermost sandstone unit, on pages 89 and D-56 these strasa are described as being 12 to 24 feet thick. What is the reason for this discrepancy?

Figure D.2.10, Page D-53 As it lacks a horizontal scale, this figure does not constitute a stratigraphic cross section.

shown on a mapThe locations drawn to scale. of If the wells and the line of section should be the wells connected are not in a straight line, the cross section should indicate where the bends occur in the section and the bearings of its legs. Alternatively, if the wells are projected onto a

' +

,.' straight section line, the direction of the projection should be shown on the index map and the cross section annoteted accordingly.

SEISM 0 LOGY Section 2.2,_SL0pE STABILITY AND SEISMIC RISK, Pages 35-36 In this segment entitled " slope stability and seismic risk", no references supporting the evaluation of the MCE or ground motion are cited. Also, no reference for the " generally accepted" static and seismic safety factors are cited that would enable an independent confirmatory review to be made.

Section 3.5, 18, Page 79 O

V In the sr 'nd complete paragraph on this page an MCE of 6.8 for the South Granite bantains fault system and a horizontal ground acceleration of 0.129 are cited. In Table 1 Selected Potential Seismogenic Sources - Riverton Site, of the SHB report the MCE for the South Granite Mountains is 7.2 and the resultant acceleration at the site is 0.10g. There is no explanation for this' discrepency. The MCE and acceleration values for the North Granite Mountains F fault system, which is closest to the site, are the same as the values cited I'n the EA.

Section B.2.5.5, pp. B19-20 for ti.. ection the coment made above for section 2.2 applies; there is no reference for the stability analyses or static and seismic safety factors. The use of effective peak acceleration is a controversial seismological topic with no generally accepted definition. Schnabel and Seed (1973), do not define effective peak acceleration, however, Bolt and Abrahamson (1982) proposed such a measure. It is suggested that more recent references than Schnabel and Seed be consulted in order to determine accelerations at the Riverton site.

Significantly more strong motion data has been collected since the work of Schnabel and Seed,

, REFERENCES CITED AHL-(Argonne National Laboratory),1979. Descri)tions of United States Uranium Re' source Areas, NUREG/CR-059CaiL/i5-75, prepared for the U.S.

Nuclear Regulatory Comission.

Bolt, B. A. ano N. A. Abrahamson,1982. New Attentuation Relations for

'peak and Expected Accelerations of 5trong Ground Motion, Bull. Seis.

Soc. Ariarica, vol. 72, pp. 2307-2322.

Bouwer,H. Groundwater Hydrology McGraw-Hill, Inc. 1978. p. 100. ,

Joyner, W.B. and D. M. Boore, 1982. Peak Horizontal Acceleration and Velocity from Strong-Motion Records Including Records from the 1979

. Og Imperial Valley, California, Earthquake, Bull. Sets. Soc. America.

Vol. 81, pp. 2011-2938

%/

Marcos G. and Bush K.J., 1983, Geochemical Investigation of UMTpAP Designated Site at Riverton, Wyoming UMTRA-DOE /AL-0ZZ9 Lindberg, R.D. and D.D. Runnells ,1984., Groundw6ter Redox Reactions: An F Analysis of Eouilibrium State ApplieTto Eh Measurements and #

Geochemical Modeling, Science, v. 255, pp. 925-927.

Schnabel, P.B. and H.B. Seed, 1973. Acceleration in Rocks for Earthquakes Effects, Riverton Site, Freement County, Wyoming, Urantum M111 Ta111ng_s Remedia,1 Action Project, unpublisheOeport prepared for the Technical Assistance Contractor (Jacobs-Weston Team), UM1RA Project Office, Albuquerque, New Mexico.

SHB(Sergent,Hauskins&Beckwith),1983. Evaluation of Potential Earth-ouake Effects, Riverton Site Fremont County, Wyoming, Uranium Mill Tallings Remedial Action Pro: unpublished report Technical Assistance contractoT ect,Jacobs

( - Weston Team) prepared

, UMTRA Project for the Office, Albuquerque, New Mexico.

White A.F. and Delany J.M., 1982,_ Chemical Transport Beneath a Uranium Mill Tailings Pile, Riverton, Wyomino, Symposium on Uranium Mill Tailings Management, fort Collins, Colorado UMTRA Project Office, 1984 Processirig Site Characterization Report for the Uranium Mill Tailings Site at Riverton, Wyoming, Draft, June,19d4 draf t docum:nt distributed by the DOE / Albuquerque Operations Office, Albuqueroue, New Mexico to the U.S. Nuclear Regulatory Comission.

,