-. -.-- ""TT ----

; r

/ IT CorporatioJJ

• • the 'I; grouP.

312 Directors Drive Knoxville. TfV 379234799 TeL 865.690.3211 Fvc. 865.690.3626

•

•

January 30, 2001

Mr. David Dorian U.S. EPA Region N

· Sam Nunn Federal Center 61 Forsyth Street, SW Atlanta, GA 30303-1304

1l M.:mlwr of'IM IT Cmup

Subject. Fonner Scovill-schrader Automotive DMslon Site, Dickson," Tennessee U.S. EPA ID Number TND 002 591 311 Conceptual Groundwater Model

Dear Mr. Dorian:

Enclosed are two (2) copies of the draft ConcepJual Groundwater Model for the Fonner Scovil-Schrader Automotive Division Site, Dickson Tennessee. This model was prepared using regional, local, and site-specific geologic and hydrologic data collected and compDed over the past" several years.- We have also Included In this conceptual model our Interpretation of contaminant

· mlgratlon pathways based on our understanding of the groundWater flow system. Anally, we Incorporated Information that was specifically requested during the COnference call between USEPA and IT on October 20, 2000 (see attached teleconference meeting minutes).

We are submitting this mod at for yooc review prior to submitting the final RA report so that comments can be Incorporated Into the ftnal RF1 report. If necessary. To facftltate your review, we would 6ke to schedule a conference can with you, Dave Jenkins, and Roger Donovan, at your ear11est convenience, to present the specific features of this site conceptual model.

Should you have any questions. please call me at {865) 690-3211.

·Sincerely,

-P~~'---Patrlcla Thompson

, Project Manager

Enclosure

cc: Roger Donovarl. Tennessee DMsion of Environment Charles Peny, Hunton and Williams Nick Bauer, Saltire Industrial Inc. Kathleen Huber, IT Corporation

•

•

•

• • • the'-~'

IT COTJ10TIItlo11 312 Directln Drive Knoxville, TN 37923 Phone: 865-69«):.~] 1 Fa: 865-690-3626 ~ lgroup

TO:

FROM:·

DA'tE:

REF:

MEMORANDUM

David Dorian. Region IV; EPA Dave Jeuldns, Region IV,HP A Nick Bauer, Saltire Industrial, Inc.

Plrtr:icia Thompaoo.IT Corp. Katbleen Huber, IT Corp.

Roger Donovan, 1DEC

October 2~, 2000

Meeting Minutes for CoDfereaeeCaD bela oo October 20, 2000 to discuss the· approach for presentation of the .hydrogeologlc:.fnformatfoa for t1te :Filial RFI Report, Former Scovi!J.Se.hndcr Automotive Facility, Dic:ksoD, TN EPA ID# TND 002-591-311

...

A coofercnce call was held OD October 20111 with David Dorian, Dave Jenicms. Kathleen Huber, and Patricia Thompson to discuSs the bydrogcologic informatiOD. that EPA considers important for inclusion in tho Final RFI report for the filcility. It is IT's understanding that the questions raised by EPA were based on their · review ofth«? Draft Groundwatr Summary Report, Grourrdwatr J,fonttortng Program. Jllllfl20, 1000. The topics discussed are SIIDllll8rized below.

OveraO Gro1111dw~ Flow System

c Dave JeukiDs made introduCtory statomcn18 expressing his concerns that tho levels 1111d axta;1t of · . contamjnatjon had not been clearly defined for the site and that it is impoJ'IBJlt .to find "lbo other end of the flow system" at the Dfclcsoo site (ie., ideutf~io1fsll8 groundwater tlowpa1bs and receptors). Ho indicated dmt bo is wading on other sites in Teuneaee where karst is an issue and that theSe systems C8Il bo defined through groundwater studies. Dave commcmtcd that he has not seen a thRo dhnensiooal cbaraCtcrization for tho ftow system or a detailed conceptual model for the-site that shows the . relationship of groUDdwater ftow regimes within a regiOnal picture. He-did not recom.mend detailed modeliog, however, he indicated that putting all avaUable·data Into a conc:eptual. model1hat iocluded the three dimensional informBtiini is imperativo fur the design of a long term monitoriDg program. He

... ooted that Saltire needs ta present a detailed oonccptual model to support the contmldon that groundwater migration is being contained either with tho Clttl'eDt interim DlC8Sili'CS groundwater 1IeatmeDt system or tbiough com:c:tivc measures that would be initiated for tho site over 1he long term. He indicated that without acceptBDce of a cooceptual mOdel that proves Saltire's undemaDdiDg of the· hydrogeology (locally and regionally), EPA may question whether Sa1tire has defined the nature and extenfof contamination which may lead to an alternate concJusion (by EPA) that contamimurt ~e is occurring but not being detected doe to an insufficient monitoring network.

Pqel of3

•

•

•

a KatJileen Huber then provided a gencml statement couc:eming the euviromnental WOik completed at the site over the last 10 years. SbC pointed out that Sa1tire has completed ClldeDsivc site cbuacterization studies and the data presented in tbe grotmd water monitoring program summary repoit did not include these da1a simply because this was not the focus of that particular document. saltire•s conceptual model of the site was reported in several RFI and related reporta that have been submi~ to USEP A in the past The dye traCe study, which was a very comprehensive stUdy provided Saltire with a good~~ of the hydrogeologic conditions specific to the site vicinity. Kathleen S"tmmarized the basic featmcs of the site CODCeptual model: a low-penueabili.ty residuum bvcrJying a fractured karst bedrock where voea in the .site soun:e areas are 1umg up in the sa,tmated zone of the residuum. slowly bleeding down iDto tbe bedrock fractures. Xathleen also DQted that in the past EPA had requested Saltite to ieview the enviromDeDtal data geoexated ~the Lewisburg site and other sites in Tennessee so these sites could be used as a model for sito investigations at Dicbou. Saltire did look at these data, however, the lattst euviram:Dc:ot at Lewisburg is quite c:tiffi:reDt from what is present at Dickson due to an entirely diffen:nt type oflimestoDc (ie., clean Ordovician limestones in Lewisburg versus n:mddy Missiasi:P:Pian limestones in Dickson) and that comparisons between the two ~systems are not appropriate. · ·

c David. Dorian asked about the ISM ~why had ODly 10 pcnmda beenleCOVCinlci from the system as per the June meeting. He asked rrs opinion on that.

a· Kathleen briefly described ISM operation and the startup period, the role of the thick residuum at the site and how the residuum slows the migration of contamimmta into the bedrock aquifer. and a1so into tho recovery wells. She noted that the iDterim groundwater 1rea1:alelit system contiimes to remove mass ftom the source areas but the limiting factor is the tigbtncM of the residuum llllil She .further mentioned that defining a "capture ZOJlll' in this type of enviromncm iS not possible due to the non-porous flow cOnditions beneath the iesiduum unit and that the idea of "capture,. should be carefully considered when discussing the remedial goals for tho site. Tho issue of flow nets was brought up during this discussion and Katb1ccn stated het' reluctance to present ftow nets in the quantitative sense due to the compll5ltities of the site because they -would not accurately depict site conditions. Flow directious can and have been approximated on existing site figures.

' . . . a Dave Jenkins agreed that trad:itiona.l flow nets mi&htnot depict an BCl;Uratc picture for the overall flow

system. ~owcver. Dave re-stated that a very good case will need to be made by Saltire regarding predictions of where contaminated'gl'O'DDdwater should be expected (both on-site and off-site) and where Saltire would expect to see discharge (based on a sound'conccptual model).

Nlllllral Attenulltlolt

a Dave leulcins made die statemeDt dtat based on the data ho .bad seen, it appears the natoni attenuation is going an to some extent aDd that SaJtire needs to pieseul data that would show this and pe:rfonn an evaluation fur this' in the RFL He suggcstetl plotting dissolved oxygen levela across the site and off site springs IIJ1d use this as a preliminuy base to see how much attlmuation Jill}' be going on. He also stated that SaltDe needs to collect data for other pai3IIlCtcm for natmal attenuation evaluation suph as nitrates. su1fa1cs. iron. etc. These data arc required before EPA will accept monitored natural attenuation as a remedy selection and EPA believes that natural attemlation will be a component to any JODg tenD COJ:rCCtivo measure. He stipulated that four quarters of monitoring are required to establish whether natural attenuation is occuaing and how eflective it could be as a coaectivc measure. David Dorian agreed that these data should be collected to expand our site chamcterization through geochemistry.

o Patricia TbO!q)son stated that an evaluation of natmal attenuation would most likely be done during coxrecti.ve 1IIC8SIIte studies and that we do not want to hold up the RFI to include this type of data in the report David Dorian indicated tbat this would be acceptable •

Paplofl

•

•

•

tJ Kathleen Huber asked Dave Jenkins to provide a reference list ofEP A guidance documents and other research papers that he considers essential. fur the evaluation ofnatunl attemtation.· Dave will provide refexenccs for this. · ·

S1111tmary

tJ In BWlliDII!Y, Dave Jenkins suggested that in order to finalize the RFI Salt:irO will need to 1) nail down discharge areas based on the regional conceptoal mode]; 2) show our predictions of degradation vs. . distance to support the fact that by tbl! time any cOutaminat:ion could reach a discbargo point it 'WOuld have minimal to no advenc en¥iiOIUDCDfaJ cfli:Cta; 3) show that DNAPL is being COD1aiucd by virtue Of the site- conditions, i.e., residnal DNAPL trappod in a tbict residuum; and 4) lower CODCCDIJ:aiiODS . of dissolved COiq)OWldlme bleeding of( but don't necessarily preseut adverse afl'ectl to o1f site mceptom.

o Kadllecn Huber rccommt:ndm that tho USEP A receive tho conceptual model portion of the final RFI before the entire docamc:at waa submitted so tbat any outs1andiDg issues could be sortod out prior to . the docum&mt completion.

tJ At the CODClusiOD of the confcri:ncc Q]]. Dave 1eokins ·suggested that EPA and Saltiro have a ""WOikina ·meeting where Saltire. could present the ccmceptual model with suppoxtiDg data such that EPA could pri:rride suggestions and ar gaidance" to ensure tbat the informatiOn~ would be su.fficitmt to finalizll tbe RFI. EPA indicated that the Jut two weeks in No~ would be convementtbr th~

tJ Patricia Thompson told EPA that we would keep them abreast on the progress of tl:ie development of tbe conceptual modeL Sbe a1&o stated 1hat data collection for natural auenuadon would be perfonncd during the CMS .

Pagt3 ofl

DSw'll"', L.'tC-.... ; .

Hydrogeologic Conceptual Modei .Packet

Draft

Former Scovill-Schrader Automotive Division Facility

Dickson, Tennessee EPA ID No.: TND 002-591-311

Submitted to:

v~'-¢0 S7".ot~

i ~ =u i~~ ~ - "(' ~ ~~ ~l. pR(j'.~r§""

U.S. Environmental Protection Agency Region IV

Prepared for: Saltire Industrial, Inc.

Prepared by: IT Corporation

312 Directors Drive Knoxville, Tennessee 3 7923

January 200 I

Fif!lln 1. PhysiograPhic Map o{Te1!llm¥,

As shown in this figure, Dickson County is located on the Western Highland Rim of Tennessee. The Highland Rim is an escarpment facing inward toward tho center of the Nashville Dome, an anciant feature, the center of which is now eroded away to fonn a topographic basin. On the surrounding Highland Rim, strata dip gently away from the fonner dome center. In Dickson County, northwesterly dips are very slight, on the order of less than I percent (Marcher, et al., 1964). Mississippian-aged limestone fonnations are exposed at the surface on the Western Highland Rim. ~ limestoMformations-tlre SL Lords, Wanaw, and Forl PayM formatiolf.t---fUe all cluuacterir.ed by aburultutt c.hert and relatively high Clay content These "muddy" limestones result in an very diffl2'ent type of karst system than limestones in other areas of the Central Basin. For CQIDple, t1ic Lewisburg, TN site, was cited as a comparable geologic/karst location for Dickson (USEP A, 1997; meeting notes); however, it is located in "clean" Ordovician-aged limestones which fonn more classic karst features (e.g., sinkholes, swallcts, losing streams).

DJUrr-Hydrogeologic Conceptual Model P~-&hrader, Dicbon, TN 12/lS/00

FIGURE 1 PHYSIOGRAPHIC LOCATION MAP•

DICKSON ON .. WESTERN HIGHLAND RIM

SoYrce: "Ground-Water Geology of the lllcklon. lmrrtnceburg, and Waverly Area• in the • . Wntom Hlghlond Rlm. Tonncsoco", 1 ,.. ~----~-~-·-J-g6-4. _____________________________________ ~ ... ";~~

NOT TO SCALE

•

• '.

•

Figure i Geglny Map in Site J1cjnitv.

This figure shows the outcrop areas of the primary geolo8ic formations in the vicinity of the site. The oldest unit p~ is the Fort Payne fonnation and it is approximately ZSO feet thick. It outcrops only in the deeply incised valleys.ofthe two major rivers draining. the area, the Piney River tO the west and Jones Cree.c to the north and east. The Warsaw Formation is the next oldest unit. It coilfonnabiy overlies the Fort Payne Fonnation .and . it is approximately 100 feet thick in the site vicinity. The Warsaw outcrops i~ most of.the intermediate-sized stream valleys and muc,-.·.ofthe highland area between basins. It iS the formation where the majority of springs issue due to it's relatively lower percentage of chert (versus the Fort Payne and St. LOuis fonnations). Therefore, the outcrop iuea of the Wtii'Saw Fontllllion is tlie :.one of groundwater discharge fur the aquifer spte:hL The St. Louis formation is the youngest unit exposed in the.site vicinity, however, it-is difficult to identify-due to the presence of a highly weathered residuum. The resiCfuum unit is a mantel of deeply weathered limestone that"(l} is very clay and chert rich. (2) is · dense, (3} exhibits a very low penneabilicy and; ·( 4}includes a cherty rubble zone just above competent bedrock:. · ·· · · : · · ·

J

D.RA..FT-Hydrogeologic Conceptual Model Packet-8covill-Schrader, Dickson, TN 12/15/00

•

•

•

.. ,.:

FIGURE 2 GEOLOGIC MAP OF DICKSON AREA

87'27'30" zs· 20'

36'07'30"

05'

02'30"

36'~----~~~~~~~~~~~~~L-~~~~~~~------~--------~~~

"Ground-Water Geology of the Dickson, Lawrenceburg, and Waverly Areas in the Western Highland Rim, Tennessee, • Plate 3, USGS, 1964. LEGEND:

CRETACEOUS CT j":.:.:;;j TUSCALOOSA GRAVEL

MISSISSIPPIAN

MsL D sT. LOUis LIMEsToNe: Mw I?Ij WARSAW LIMESTONE MFP ~ FORT PAYNE FORMATION

0 a.ooo

SCALE:

• ~----------------------m-e~~OUP.

f -N-

16,000

Ft:ET ~ rr Corporation ~~-5885 ninity Par.bvay Suite 120 O=utcr!Ville, VA20120 ~ .. -.,._tra... __ ...

Figun 3. GeneraligdStratigraphic Column and Cbnuptual Wemluring ProfilL

As shown in this figure, geologic formations present in the vicinity ofDickson have been subject to erosion and weathering, resulting in a complex boundary between the competent St. Louis Formation and the overlying residuum derived from the chemical weathering of the St. Louis. For this reason, the St. Louis formation is difficult to identifY based on geologic boring data alone. It's presence is estimated using both geologic data and structure maps provided in Marcher et al. ( 1964).

The tJUcknas of the resitbulm overburden ranges between 10 to more than 80 feet as determined from borings completed as part of the RFI procesa. It is believed that competent bedrock encountered at the site belongs to either the lower St. Louis Formation or the upper Warsaw Formation, depending on the elevation at which it was encountered.

DRAFT-Hydrogeologic Conceptual Model Packet-Scovill·Schrader, Dickson, TN 12/15/00

1..

STRA TJGRAPHIC COL~

FIGURE 3 GENERALIZED STRATIGRAPHIC COLUMN

AND CONCEPTUAL WEATHERING PROFILE

• ~----------------------------------'M~~OUP.

IT Corporatioll 5885 Trinity ParJ:woy Suire J2lJ =treviJJ~ VA 20120

.... ..........,.otn.rr~

Figure 4. Shaded RtlkfToeographv Mqp.

This figure shows the regional topography in a shaded relief format so that the landscape surrounding the site is more easily visualized1• The approximate site location is shown in outline to the center right of the figure. Beaverdam Creek, Spicer Creek, Will Hall Creek and Willow Branch are labeled The approximate location of the regional surface water divide is indicated with the yellow dashed line. The approximate outcrop of the lower WBISawlupper Fort Payne formations is depicted by blue shading. Thil stratigraphic Werval i8 belUI'Bd to be the bottom of the hust sptem due to (1) t1u abUIJlimu:e of chert in the Forl Payne fot:mo.tioll (I.~, effective pumeabllJJy oftlu unit i8 gretltly reduced from t1u Warsaw Fm. abo.ve), and (2) t1u tdnmll bed elevations, which serve 118 tlralnsfor the aquifer S]ltem, tUe primarily within t1u oulcrop oftlu Wanaw Formation. Therefore, the extent of~ aquifer system in the vicinity of the site is approximated by the non-shaded areas, signifiCant partions of which arc not hydraulically related to the site due to the presence of the surface water divide. (In general, flow directions Will be at right angles to this divide line; therefore, flow away from the site is not expected to occur parallel to this divide line).

Unear features are evident in the landscape depicted on these maps, most notably as the stream valleys. Large tributary streams of the Piney River (lower left of the Dickson map) enter the main stream at nearly right angles, suggesting a fracture origin for the stream . bed. Fractures along the regional surface water divide are not easily observed due to the lack of stream incision and the masking of fracture patterns by a thick residuum overburden.

Site groundwater flOw appeon ttl be limiJed toward twt1 main aretU: (1) to the east and IWrth in the vaUep of Beaverdam Creek and Will Hall Creek; and (2) to the wen in t1u valle] of WUlow Branch. Groundwater discharge points in these valleyB inckule sprlnp, seeps, and IHdrock parllnp In ths slrelmi bed.

1 Two digital elevarloa JDOdela are combiDod to create lhU figure, the Dicbon topograpbic quadrangle on lhe left and tbe BlllllJ 10p0grapbic qlllldranglea on tho right (USGS, OEM fiJeo doWDioaded from lhe USGS web site). The black daabcd lino down the middle of the figwc iJ the boundary between tho two quadransJea.

DRAFr-Hydrogeologic Conceptual Model Packet--Scovill-Scbradcr, Dicbon, TN O:UOI/01 .

6000

FIGURE 4 SHADED RELIEF TOPOGRAPHY MAP .

16000 I

sc...u: HET

• ~-----------------------------------tM~~UP. IT Corporatloa 5885 Trinity Pomov Suite120 OmtnwiliD, VA 2QJ20 .. .-.....rn.na...

···~

Figure 5. Regional SW-NE Geologic Cross Section.

Approximately 10 miles acrosl, this generalized geologic cross section shows the nearest Fort Payne outcrops to the Sit~ear mile 23 of the Piney River on the southwest and near the fish hatchery in the northeast. Geologic unit thicknesses and structure are obtained from regional references (Marcher, et al., 1964 and Bradley, 1985).

This section illustrates the vertical geologic conditions in the vicinity of the site. It is reported that the Warsaw Formation is the dominant water-bearing unit in the vicinity of Dickson (Bradley, 1985). Groundwater is collected in the upland areas, is stored in the residuum overburden, migrates downward to the fracture/solution network in the Warsaw, and is discharged in local streams via springs and gaining reaches. This figure sht11f1slum the St. Louis/Warsaw (ie., residuum/bedrock) tUfUifer system is perched on top of the Fort Payne Formation. Deeper j101r into the Fort Payne is not likely due to both the abundaitce of chert itt tJds unit and the location of significant tUfuifa drains (ie., local streams) at hight!T elevations.

The following three cross sections provide a more detailed look in the area of the site.

2See cross section location map as Attachment A.

DJUPT-Hydrogeologic Conceptual Model Packet-Scovili-Schnuler, Dickson. TN 12/15/00

•

:; C4 ::!!

900

800

700

:U fiOO w !!;

z 0 j: w ...1 w

400

300

. 0

FIGURE 5 REGIONAL SW·NE GEOLOGICAL CROSS SECTION

BETWEEN NEAREST FORT PAYNE OUTCROPS

1.0 2 . 0 3.0 4.0 5.0 11.0 7.0 8 .0 HORIZONTAL SCALE (MILES)

~~ . GEOLOGIC CONTACTS ESTIWAT£0 rROW th£1 ~"·

'-----S•T-RU•C•ru•R•E--MA_P __ P_RE•S-EN•l•E•O•IN __ B_RA•O•~-Y--(1•9•8•5) __________________________________ ,,~()ltj?

9.0 10.0

rr CorporatioiJ 588.5 Tr:ill/ty P'aaway Suitr: 120 OonU.rilJc, VA ·20120 "'~flll'n..rro.o..

Figure 6. S-N Cross Section through Site,

Roughly parallel to regional divide trace, this section depicts the greatest lateral extent ofthe St. Louis Formation-which is believed to exist primarily as weathered residuum. As shown in this figure, WJ.lluw Braru:h and West Spicer Branch have eroded through the St. Louis inJo Warsaw Foi-numon.

DIUFT-Hydrogeologic Conceptual Model Packet-Scovill-Scbrader, Dickson, TN 12/15/00

1 ..

• • •

:J C/)

~

900

800

.... w 700 w u.

z 0 . i= 600 o( > w -' w

500

400

I

FIGURE 6 SOUTH·NORTH GEOLOGICAL CROSS SECTION THROUGH SITE

APPROXIMATELY ALONG SURFACE WATER DIVIDE .

SITE

:1: 0 z

"' 0:: lXI :I: u z < a: lXI

0 0.6 1.0 1.5 2.0 2.5 3.0 HORIZONTAL SCALE (MILESt

t • NOTE: J 4(1 GEOLOGIC CONTACTS ESTIMATED fROM l. £e A STRUCTURE MAP fRESEN"fED IN BRADLEY (1985) 1grouP.

IT Corporation ~~ 5885 Trinity Par*"vay 0 Suite J2f) 1 Centreville, VA 20120 l!! "'~•n.aelroclp

FiKHre 7. SW-NE Cross Secdqn through Site.

This section roughly traces the lower reaches of Willow Branch, across Site, then on to headwaters of Will Hall Creek. Springs shown issue from the Warsaw Formation. Willow Branch appean to flow within the Wanaw formation from lleiU' its ~adwaters to its confluence with the East Piney River.

DRAFT-Hydrogeologic Conceptual Model Packet-Scovill-Scllrader, Dickson. TN 12/15/00

• • • FIGURE 7

WSW-ENE GEOLOGICAL CROSS SECTION ALONG WILLOW BRAf'ICH

aoo.,

800

:J (/) 700 :i 1-w w LL -600 z 0 i= "' > w 500 ..J w

400

300

.... ..

- ,---

lol .... .... )£ a: .... u

w a: UVl

::l:

• • •

Figure 8. NW=SE Cross Section through Site

This section extends rougWy along Beaverdam Creek from northwest of the northern portion of Site. The large springs shown issue from Warsaw Formation.

DJUf'T-,-Hydrogeologic Conceptual Model Packet-Scoviii-Schrader, Dickson, TN 12/15/00

r '·

:J CD :iii

•

900

800

... w 700 w ~ z 0 i= 600 ~ w .J w

500

400

• FIGURE 8

NORTHWEST-SOUTHEAST GEOLOGICAL CROSS SECTION ALONG BEAVERDAM CREEK

SITE

:.: w w IX u ~

"' Q IX ~

"' w

•

•

•

Figure 9. Spring Location Map from Karst Survey.

This figure shows all the springs within a 2 mile radius of the site as located during karst survey in 1993. Of the nearly 200 springs located during this survey (including those beyond 2-mile radius), more than 80% issue from the Warsaw fonnation, roughly 18% issue from the St. Louis residuum or the contact between the Warsaw and St. Louis formations. The remaining 1% issue from the Warsaw/Fort Payne contact. This distribution of springs supports the notion that the Forl Payne Formation is not a nugor groundwater flow unit as reporled in Bradley (1985). Therefore, based on regional reports and data collected as part of the RFI process, deeper groundwater flow cells in the Fort Payne are not likely to occur .

DRAFT-Hydrogeologic Conceptual Model Packet--Scovili·Schrader, Dickson. TN 12115100

•

Lflil!ND: 0 3000 6000 ~ i 0 SPIIHQ l.CCAliON I t -N-SCALE FEET ~ • ITc...,..,e I the f.{; groUp . Rl:'1'rlolv ~ Nt.UD I \ ~~VAJIJ.W

figure JQ. Coru:epmal Cross SecJion Slwwinr Lerge-Scok GWFluw Svstem.

This figuro illustrates tho groundwatc:c flow system in the vicinity of the site. The residuum unit (chemically weatheced St. Louis Formation) collects and stores precipitation, then slowly releases groundwater to the fracture/solution channels of the Warsaw Formation below. Groundwata in the WarsiiW migrates laterally to discharge points in local streams within the solution network. The Fort Payne Formation, tb4e to abiUidmu:e of bedded chert, doe$ rwt contribuU signijiCIUitly to 1M groiUUlwater flow syatem, although flow along the contact between the Warsaw and Fort Payne is suggested by Bradley (1985).

DRAFT-Hydrogeologic Conceptual Model Packet-Scovill-Scbradu, Picksoo, TN 1?./15!00

900

800

::! rJ)

2

:U1oo w ~ % 0 j: 800 w ..J w

500

400

0

FIGURE 10 CONCEPTUAL GEOLOGICAL CROSS SECTION

NW-SE ALONG BEAVERDAM CREEK :SHOWING LARGE·SCALE GROUNDWATER FLOW SYSTEM

SITE

~

0.5 1.0 1.5 HORIZONTAL SCALE (MILESI

2.0 2 .5

INSIGNifiCANT fLOW, ESPECIALLY WITH INCREASE IN DEPTH

NO SIGNifiCANT fLOW

~~ . GEOLOCIC CONTACTS ESTIMATED FROIA the( ,.. IT C4rporatioa 5885 TriDity Patbn]' Suito J20 '-----S•T-RU•C•T-UR•E--MA•P--PR•E•SE•N•TE•O--IN•B•R•~•U•Y--(1•9•8•5) ________________________________ ,,~1? O:na=i/1.., V.A 20120 A........,.•n..rr~

• • •

Figure 11. Conceptual Cross Section Showing Site-Wide Contaminants.

Groundwater monitoring data show that the highest concentrations of VOCs are hung up in the s~rated zone beneath source areas at the site. 11te low-permeability, high natural organic content of the residuum has trapped the source area constituents within the upper $aturatell zone. Constant flushing of rainwater through the source area serves to "slowly bleed" the VOCs into lower units, i.e., the ruqble zone at the base of the residuum and the Warsaw Formation below.

DRAFT-Hydrogeologic Conceptual Model Packet-Scoviii-Schrader, Dickson, TN 12/15/00

• •• FIGURE 11

CONCEPTUAL CROSS SECTION SHOWING SITE;.WIDE CONTAMINANTS

SITE

~ SEASONAL

WATER TABLE~ EXTREMES

MOST AREA . SPRINGS ISSUE 1

FROM WARWA W I FORIIA TION

•

MOST AREA SPRINGS ISSUE

FROM WARWAW FORMATION

•

~ ....... "' ·• 0. ~

rr CorporatioD ~~ 5885 Trizlity Pa.Tlcwa.r tl} Suite 120 I OmtreviJJc, VA 20120 If A ,..._.o1a. rrca...p

the'f;groul!

• •

Figure 11. Offsite. Groundwater La•el Map.

Generalized groundwater levels in vicinity of site are drawn from area wells and spring elevations. While contours are not corrected for surface streams, this figure shows generalized radial flow pattern away from the site. Preferential flow toward Beaverdam Creek drainage is suggested by more shallow gradient in that direction (i.e., shallower gradient suggests higher effective Kin fractured rock' whereas steeper gradient suggests lower effective K).

I

DRAfT-Hydrogeologic Conceptual Model Packet-ScO\ill-Scltrader, Dickson, TN 12115/00

•

f,,

j

FtGURE 12a GENERALIZEO OFF-SITE GROUNOWATER ELEVATION CONTOUR MAP, JUNE 1999 LEGEND:

o SPRING LOCATtoN SAWPL£0

• P«NATE wru.s s.wr~ • LOCATIOii or OFT-SITE &EDROCK

WONITOIUHQ W(U. 00\otW2.

~ LOGAnON Of ~-SIT!: WELL CWST£•

QACCES5 OOfr£0 8'1' PROPEJfJY OWHU fOI SPRINO/USiotNU-.L WELL 5-AMPUNQ

b~~~~R 1 ~~;~ .. AJ,.VD"Jo~ -11JQU1N£ 1 ~;g) --~ QROUiocOWATER £U:VATICN

CONTOUR

- -llfJO- - INtER.REO CAOUNOWATER . " tl.iVAnON C~TOUR

* 4T Of'F-Sitt WEll.. CLUSTERS, ONL.Y THE OROU~,41ER !l..£VATIONS FOR AESJOUUW-BE:OROC::X WEU. INCLUOED · IN CONTOURING •

.. SAWPUHO LOCAnoN OESTJqOY£0

NOTES: 1. SPRING NID WE:'U. LOCATIONS WEA(

SURVEY£0 IN D~CEUBEA 1997.

2 . GROUNDWATER WV.t.TIQN CONTOURS ~C 11AS£D OH WA.T£. lLV£L SURVt'Y PERrORWED lfol JUNE 1099.

3. OAOIJNOWATER (l£VATIONS or SOtr.l£ UICA OOHS W(RE NOT IHCt..UOEO lH COWTOURINQ. TlCO'it. LOCATIONS .ul[ SHOWN lN Q.RCY,

•

•

•

~·

Figure 13. Onsite -GW Level Map.

The groundwater level contour-map at the site has remained roughly the same since. water level measurements have been collected .. Groundwater flows from higher areas of site (near building) toward offsite natural drains (i.e., springs and seeps) in the tributaries of Willow Branch and· Beaverdam Creek. The seasonal range in·water levels varies across the site betv.reen 3 and 383 feet. The average seasonal water level change is approximately 1 5 feet.

3 Two wells (MW35 & MW31) exlu'bited a much greater seasonal range in water levels that the remaining · site wells. When these two wells are eliminated from consideration, the range in water levels at the site is 3-24 feet and the average is approximately 14 feet.

DRAFT-Hydrogeologic Con~tual Model Packet-8covill-Scbrader, Dickson, TN 12/15/00

• FIGIURE 13 GENERALIZED ON .. SITE GROUNDWATER ELEVATION CONTOUR MAP, JUNE 1999 l /

. I

~i . /s./ I 'vr" . . I

h

f

T

1. SURVEY OF AlL W£U. LOCATIOHS WAS Pt:RfORMEO IN NOVEII8£R 1997.

2. CDIITDIJR IHTERVAI. z S fT. ·3. GROUNDWATER ELEVA'llOH

VALUES ARE IN FEET ABOVE jj!JIN S£.1. LEVEL

4. THIS F1DURE IS BASED ON A SUflf[R IIAP AMO IS HOT DIRECTLY COIIPARABl£ TO CARl. Y MAl'S. IT WILL PROV10£ A ·MORE COMPARABlE BASE loiN' fOR 0111' DATA.

5. ClROUNoWAT£11 ELEV4'!10HS Of 5014£ WELLS WERE !lOT INCLUDED IN CONTOUR!NII. THOSE ELEVATIONS ARE SHOV

•

•

•

Figure U. Site TCE Plume Map.

Highest concentrations ofVOCs are centered near two main source areas: the Northern Disposal Area and the South Building area. Lateral spreading of plume is apparently limited tD preferential flow pathways (e.g., between MW36 and MWI5). Daughter products are pre5ent only within the most concentrated portion of the plume, and vinyl chloride is not accumulating.

Highlighted wells indicate locations where TCE daughter products are present .

DRAFT-Hydrogeologic Conceptual Model Packct--8covill-8chrader, Dickson, TN 12/15/00

u :

•

h

FlGURE 14 .TCE IN GROUNDWATER • DECEMBER 1998

rv I ' ' I

*

-~ -N-

~

I. RESIDUUM-BEDROCK 'NEU.S ARE SCREENED IN RESIDUUU JUST ABOVE RESIDUUM-BEOROCK CONTACT. FOR DESCRIPllONS or D'ni£R WELL DESIGNAnON CRITtRIA, REFER TO THE WELL CON-STRUCTION OETAII.S TAill£ IN AI'PENOIX C.

2. EXTII.lCTION WELL EWt IS SCREENED FROM ~30 Fm

. AOOvt THE RESIDUUM-BEDROCK CONTACT DOWN TO THE TOP OF BEOROCIC.

LEGEND: - --- PROI'tRTY 90Ul«!ART --rENe£ - ···- ··· -I~MimNT SlAEAII

OR DRAINAGE CIJAijNU

..t. SPRING LOCAnON

+- SHAll. OW. RESIDUUM WEU.. + . RESIDUUiil wt:LL e RESIOUUW-BEOROCI< WELL

+ SHALLOW B~ROCK wt1.L -100- TCE CONCENTAAT1014

ISOCDNTO\Jll

- -100--INrtRRED .TCE CONCENTRA-TION ISOCONTOUR

TCE CONCENTRAT10NS AT LOCATIONS loll'/39, IIW.O, IIW41, AND IIW42 AR£ FROW JUNE 1999,

N1J COI!IPOUN!I HOT · DCTiiCT!D,~t OR ~lldl/£ TCE- .QIII.NnT.IJIOH .uodT

·- - ----· 0 2SO 500 -SCUE rEEl • IT Corpozwllflll

11ta4/)I{I'Oup sn>nuq.........,. SW..UD ~~YA~J'II

! I

•

•

•

Figure 15. Plume Cross Section in Northern Disposal Source Area.

Highest VOC concentrations hung up in saturaied residuum, deeper portions of plume bleeding into bedrock Concentrations relatively stable over 10 year monitoring period. Transport in bedrock unit is lateral, with dilution being predominant attenuation mechanism .

DRA111'-Hydrogeologic Conceptual Model Pack:et-Scovill..Scbrader, Dickson, 'IN 12/15/00

aoo

FIGURE 15 CROSS SECTION IN NORTHERN DISPOSAL AREA

1200

POCI(!:TS OF TIW'P£0 (R~IOUAL) ON.u>l NEAR SOURC£ AJIEA

700

... LEGEND: .•

• ~ ~~~P~C=~rtH(~~O~i~;~ IN GROUNDWAT£R • At \.ocA.TlOH'S ~Al WtR£ troT ~g J)l JUHt. 1998,

TME 0£C£WIIIOl 1981! COMCOITRAnON IS SHOWH.

~ WEAJ HU£'0 ft£51DUUW: SILT't' CLAY WITH ~ CHarT NOOUU:S. rl,t.GJ,i[HTS AND SAND

~ CM[R'N zo~oet: ex.TENT OF rei! • ~ SAPA.Ol..ITE IN GROUNDWA. TEll: ~ (CONCENTRATION/J SHOWN ~ ~~S'TQfl4! IN, PARTS nR ~II..LJOffJ

·ro t.r~EL ra >loo.O100 -+" =~ JUNC 19'39 ---tiO • NO.T OCT£C'IYD

•

1&'111>10,00<

Ia >1 ,000 liilllll'l > 10

. ~!!a > 1

===~=~==========rro==. ====~----~~~~~------~-------------------~M~~

•

•

•

Figure 16. Plume Cross Section in South Building Source Area.

Hig}:lest VOC concentrations hung up in saturated residuum, deeper portions of plume bleeding into bedrock. Concentrations relatively stable over 10 year monitoring period. High concentration plume is no longer migrating laterally, rather, it iS bleeding off into the deeper bedrock where lateral movement and dilution are occurring.

_____________ ":-. .. ,:- -···-

DRA17-Hydrogeologic Conceptual Model Packet-8covill-Schrader, Dickson, TN 12115100 . .

FI(3URE 16 PLUME CROSS SECTION IN SOUTH BUILDING SOURCE AREA

e()(l 180 1100 1050 ut~o tsao 1&40 . usso · tSoo 1850 . a10o uao 2400 auo 21oo

;~ t'J,luO: .. :E NOT 'SAWPlED EW9£R 19518

'"· SJLTY cur Wl11'l

EMTS AHD UNO

HORIZONTAL SCALE

?OCKOS Of TJW'P£0 (RESIOU.l) 0"-"PL ~tAR 50URCE . AREA

• :;::,:,-wrJ:-~ BEDROCIC \i) ftOIOOUW-e[DROC'l W£U

-$- SMAWlW IIEDIOOCio . wt1.L ~ .. ,ouuv wtlJ. • SOIL. ICUttH.CJ L.DC.t.T10H SCAU

·.·_1. .. ·

·-~· · - .

000 1200

ra:r

•

• . .

·Figure 17. Multiple Grgphs, VOCs through Tiine at Onsite Locations.

voc coneentrations lurve been relatively stable through 1 iJ year monitoripg period. Daughter products are present only near Jouru areas and JIC is not aceu1Tilll41ing • . VOCs are trapped in saturated portion of the residuum, slowly bleeding off into lower bedrock system Dilution in bedrock solution system increases with distance from site. Concentrations believed be entering bedfock at site are on the order of 100-1 OOOppb,. whereas SS and S6 total VOC con~ntrations are on the order of

e Figure 17. VOC Graphs In Select Areas. MW32 VOC Concentrations

c 1000000.0 ·--=---=- £ I .. ~ . 0 100000.0 :;::; ! :--"-"" e- 10000.0 l ___ ,...

-.Q _ . ..,. co. 1000.0 ...--- • .._._,_,IX:I!f

• Figura 17. (can't) VOC Concentrations in MW26

s:: 10000' 0

~ :l :t:l· 1000 res· ... : ~ cis-1,2 DCE .. -...... .a ca. 100 i

-tt-PCE G)Q. u- 10 -di-TCE c

$ ............. I 0 1 (.) Sfl/90 1131/93 10/28/95 7/24/98 4/19/01

Sample Date

'

VOC Concentrations in MW31

• c:::: 10000 0 .......... :;:::; 1000 .~ ..;.!, .A -.r-TCE ~:c

100 I _,_ cis-1,2 DCE cc. Q)Q. v V' . .......,. -+-1,1,2 TCA u- 10 s::: 6 -+-VC 0

1 (.) 5/7/90 1/31/93 1 0/28/95 7/24/98 4/19/01

Sample Date

VOC Concentrations in DMW38

c: 10000.0 0 ..,._ ;:; 1000.0 -~- l1l' -.c

100.0 I~Dral c c. CDC. u-

10.0 c 8 """'

, ~ , --I-·

10128/95 3/11/97 7/24/98 1216/99 4/19/01

• Date Sampled ~\lii'T',.....C' --~ n-··-L.A.-- .. f ••~ •-

•

•

··- .. · ... , .. _c_·_

ATTACHMENT A CROSS SECTION LOCATION MAP

20'

05'

3~~---~~~~ZL--~~~~~~~~~~-=~------~------~~~

LEGEND: CRETACEOUS

Cr 12] ruscALoosA GRAVEL

Source: "Ground-Water Geology of the Dickson. Lawrenceburg, and Wcverly Areas in tha Western Highland Rim, Tennessee." Plate 3, USGS, 1964.

---a----MISSISSIPPIAN -t .

• MsL D sT. LOUIS LIMEsToNE Mw lf.IH WARSAW LIMESTONE . MFP ~ FORT PAY.NE FOR~ATlON

0

SCAL£

-N-

8,000 1 6,000

fEET

rr Corporatlon $885 Trinity .Pukway Suite 12!1

l

• • FIGURE 18

AL GROUNDWATER DISCHARGE . AREAS FOR THE SITE

SCAU: FEET

Of NO OFFSITE GROUNDWATER I.IIGRATIOH JT CorporatJOD

•

t T

I the A Suit.e J20 PRESENCE Of' REGIONAL SURF' ACE WATER OIVIOE ~· 588.5 TJ:iJJity Pariw.ey

[ · --==-]APPROXIMATE !AREA WITHIN DOWHGRADIENT AREA . 1m-oun Cc.otmvilla, VA20lZO -·~-= ·- Of' SIT£ UKE~Y TO RECEIVE FLOW fROM SITE er . 7::. A-.r:n.JTa...,. --~

' . '

t ,.

I

•

•

•

------

HYDROGEOLOGIC CONCEPTUAL MODEL SUMMARY

(J The geologic framework in the Dickson area is characterized by muddy Mississippian-aged limestones which form a much different karst system (in terms of dissolution features) than middle-Ordovician limestone units elsewhere in Tennessee.

Q Karst features in the Dicks~n area are characterized by: (1) a significant residuum overburden unit derived from the chemical weathering of the St. Louis Formation, (2) small-scale dissolution features, (3) rare, large-scale dissolution features such as caves and active sinks, and (3) an aquifer system that is drained by springs feeding local and main stem tributaries. These features are described in regional geology reports for the Dickson area and were verified by the extensive field karst survey completed as part of the Phase II RFL . .

0 The Warsaw Formation is the primary aquifer unit in the area. Most lateral groundwater flow occurs within the Warsaw.

0 The top Fort Payne Formation effectively represents the bottom of the aquifer system due to the abundance ofbedded and scraggy chert in this formation preventing the development of significant dissolution pathways for groundwater flow. This is further supported by the fact that the vast majority of springs identified in the area issue from the Warsaw Formation. Another important feature is the presence of · effective drains for the aquifer system stratigraphieatly above the Fort Payne Formation in the form of springs, seeps, and gaining reaches of streams.

0 The residuum overburden unit developed from chemical weathering of the StLouis Formation. Due to the depth of the weathering front, the St. Louis is difficult to identify at the ground surface. Residuum thickness ranges from a few feet to over 80 feet in the site vicinity.

0 The residuum overburden unit serves to stabilize groundwater flow by (I) providing a storage reservoir for infiltrating rainwater, and (2) slowly and steadily releasing groundwater to the underlying Warsaw Formation.

1:1 Based on data collected from site well clusters, VOCs are trapped within the saturated residuum in each of the site source areas.

0 VOC migration from the source areas is similar to that of groundwater-slowly . migrating downward to the Warsaw Formation, where lateral flow to local springs and streams predominates.

0 The discharge area for the site is limited to the areas perpendicular to the regional surface water divide, within the outcrop area of the Warsaw/Ft Payne contact.

0 Given the differences in VOC concentrations between the site and offsite springs, it appears"tba:t soirre-a-ttemnrtitTirinmmratlyuk1ng place m the aquifer syStem. Therefore, monitored natural attenuation will be carefully evaluated as part of the upcoming CMS .

. .. DRAFT-Hydrogeologic Conceptual Model Packet-scovill-schrader, Dickson, TN 12/15/00