/ DAMES & MOORE am A DAMES & MOORE GROUP COMPANY September 9,1998 REVISED DRAFT SITE REMOVAL EVALUATION MEMORANDUM DIAMOND AND GRANBY SUBDISTRICTS NEWTON COUNTY MINE TAILINGS SITE NEWTON COUNTY, MISSOURI Job No. 37182-001-030 ex [.,„ .mi mil nin 11 Suite 2500 633 Seventeenth Street Denver, Colorado 80202-3625 (303) 294-9100
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Appendix A Analytical ResultsAppendix B QA/QC Data Assessment Reports
ii
P:\NEWTONCO\SREM\SREM.RV1 DAMES & MOORE
1.0 INTRODUCTION
This Site Removal Evaluation Memorandum (SREM) includes a brief description of theinvestigations performed during the Removal Site Evaluation conducted at the Diamond and Granbymining subdistricts of the Newton County Mine Tailings Site (Site) located in southwesternMissouri. This SREM has been prepared by Dames & Moore on behalf of Blue Tee Corp. andASARCO Inc. pursuant to an Administrative Order on Consent (AOC), U.S. EnvironmentalProtection Agency (EPA) Docket No. VTI-96-F-0022, dated June 20, 1997. The SREM is adeliverable listed in AOC paragraph 27b.(l)(b)iii.
Data collected during the Removal Site Evaluation will be used to support an EngineeringEvaluation/Cost Analysis (EE/CA). If any removal action is warranted, the most appropriate non-time-critical removal action will be described in the EE/CA.
This SREM includes a summary of sample results, a Site-specific cadmium to zinc ratio, a qualityassurance/quality control (QA/QC) assessment of the data, and a discussion of any complicationsencountered or deviations from the approved Sampling and Analysis Plan (SAP). This SREM alsoincludes a list of residences where exceedances of the drinking water and/or residential yard soiltime-critical removal action levels defined in the AOC have been identified.
Section 2.0 of this SREM provides Site background and physical setting information and theRemoval Site Evaluation investigations are described in Section 3.0. A discussion of results isprovided in Section 4.0.
P:\NEWTONCO\SREM\SREM.RV1 DAMES & MOORE
2.0 SITE BACKGROUND AND PHYSICAL SETTING
The Diamond and Grariby mining subdistricts are located in southwestern Missouri in and aroundthe towns of Diamond and Granby. The general boundaries of the subdistricts are shown on Figure2-1. The towns of Diamond and Granby are located approximately 13 and 22 miles southeast ofJoplin, Missouri respectively. The Diamond and Granby mining subdistricts are among thosesubdistricts that make up an important lead and zinc mining region in Missouri, Kansas, andOklahoma known as the Tri-State Mining District.
2.1 SITE BACKGROUND
The Tri-State Mining District (District) was one of the foremost lead-zinc mining areas of the worldand provided nearly continuous production from about 1850 until 1970 with approximately 460million tons of crude ore produced (Spruill, 1987). The Granby subdistrict accounted forapproximately five percent of the total District production of lead-zinc concentrates and the Diamondsubdistrict accounted for less than one percent of the total District production (Brockie et al., 1968).
The majority of the mining in the Missouri portion of the District was by underground methods. Themined ore was hoisted from the underground workings and was treated at mills on the surface. Atthe mill, the crude ore was crushed and sized to minus 5/8 inch, and then concentrated using gravity-separation processes (jigging and tabling). After about 1920, the froth-flotation process wastypically used in the District to treat the finer-grained residues from jigging and tabling. However,this technique was not extensively used in Missouri because of the free-milling nature of the oresand the general curtailment of mining after 1920 (Dames & Moore, 1992).
The mined and mining-related materials on the ground surface consist of materials from mining,milling, smelting, and other operations. These materials may include development rock, chat, sands,fine tailings, and slag. Development rock is rock that was removed during excavation of the shaftsas they were advanced through the Pennsylvanian shale and/or Mississippian limestone to themineralized zones. Two types of mine material which are similar in nature to and sometimesreferred to as development rock include waste rock and overburden. Waste rock is oversizedmaterial that did not pass the grizzly (primary screening device) and did not meet grade based onvisual inspection by the mill operators. Because most of the waste rock was extracted from minelevels, it is composed mainly of Mississippian limestone and jasperoid. Overburden is limestone
P:\NEWTONCO\SREM\SREM.RV! DAMES &MOORE
DlamondMining
Subdlstrlct: -George
_ .-l^ashingto' ' Carver
Notionot-
MiningSubdlstrlct
o 1/2 i______!5iiiSCALE IN MILES
NEWTON COUNTY, MODiamond & Granby Mining Subdistricts
^5/ DW-43Rural drinking water and residential yard soil m Residential yard soflsampling location and Identification number sampling location onlyMetals concentration exceed AOC time-critical -^ .„„„„ ,un|«<i Rural Dlremoval action level(s) for drinking water * Access deni
DW-16 I
| DW-15
; DW-2
j 27i OW-3: •
w DW-61)S, • Beech Rd.
• X • •DW-90 DW-77
DW-4*L . DW-89"CharmingPRjS DW-23 -o
' 34 II §i Q-
l__ OW-60 Cherry Rd.
DW-6
DW-65
3DW-102 OW-9
—— 7OW-10i i DW-76\
-T "TTast (Market fel •I •
Ip^ DW-62— LL Diamond
DW-13DW-26 „ ^• in DW-10f—
f / *\y _— - — . — • 7=- •
DW-99 DW-88J DW-87 B
m^=:==:r--~J)\N-75/ ^DW-59 ^^^-'
/-///•
^/- _ v
•^ _l 1 5
" DW-91
Foliage Rd.ilitalsmiii iipppiSSS
DW-28 DW-29^^
DW-1in • ........ ...... „
i ' /\
ttix>l\1 \ ^-Genera!i )< Boundary%S \ of Diamondp l Mining District
I 'i spi \tw~5 \i ';» \1 ''I/WT~ •DW~8
|DW-D4
Bi-iDW-11 ^ / '
lDW-12 //w'~~~- —/— • — • — •
t^-——Vj-7- ""
1 DW~63
mm
rt"3""* \
1 / }v L/
— \ \
1 >i
: );/'I
!i**~* i
li /• ij \
l"""•»,
\s.
! %_^
^^^
"^>
1/>
\
\
\
1\ -
f 1;/
! V
}i 1
l~")\^-s
\
//
//
//•
— — „— ' ——
(D
j\%,r
""•v*
s\\11J
//1t\\1
/
NEWTON COUNTY, MODiamond Mining Subdlstrlct
Inking Water and Residential Yard Soil Sample Locations MapFigure 4-1
GW-555 QW-584GW-552 GW-501
GW-605GW-6O4
GW-558 GW_559• «
East Cemetery \GW-521 •
GW-590GW-588 GW-6O2
GW-522GW-561
GW-582 GW-575*x GW-594
GW-519GW-520GW-503
GW-577
GeneralBoundaryof GranbyMining District
GW-583GW_572
lark Springss
GW-515-GW-591 •GW-516
GW-593GW-592
GW-549
GW-511GW-510
GW-548 » I GW-567 \
1500 3000~
SCALE IN FEET
6000 EXPLANATKMM
GW-523•
GW-634Rural drinking water and residential yard soilsampling location and Identification numberMetals concentration exceed AOC time-criticalremoval action level(s) for drinking water
Access deniedNEWTON COUNTY, MO
Granby Mining SubdistrlctRural Drinking Water and Residential Yard Soil Sample Locations Map
Figure 4—2
or shale with scattered waste rock, which was excavated during surface mining in open pits(Dames & Moore, 1992).
The milling of the crude ores produced several types of waste products. The waste materials fromthe milling operations are classified into three general categories: chat, sands, and fine tailings. Chatis 1/4-inch to 5/8-inch gravel sized angular rock fragments produced by the jigging operation. Sandsare material ranging in size from #20- to #65-mesh produced by the shaking table separationoperation. The chat and sands were usually combined and stacked in large piles collectively referredto as chat piles. Fine tailings are fines removed from the gravity separation process and finesproduced by the froth flotation process. Fine tailings were usually deposited in bermed areas orponds. A sand size and smaller fine tailings was also produced by the washing and screening of chatto produce commercial aggregates (Dames & Moore, 1992).
Slag is composed of the oxides of gangue minerals which result from smelting operations. Slagseparates from the molten metals by gravity during the smelting operation. Another material similarin appearance to smelter slag is boiler residue, or "clinker," discarded from wood- and coal-firedsteam plants that powered the mines and mills of the pre-World War I period. Clinker is composedof dust particles and the insoluble portion of coal used for fuel in boilers and other operationsrequiring heat (Dames & Moore, 1992).
i2.2 PHYSICAL SETTING
The Diamond and Granby mining subdistricts are in the Springfield Plateau Region of the OzarkPhysiographic Province. They are located on the northwest flank of the Ozark uplift region whichhas plateau-like features (Shrader, et al., 1954; Smith and Siebenthal, 1907).
The geology, hydrogeology, surface water, soils, climate, and population and land use at theDiamond and Granby mining subdistricts are described in the following subsections.
2.2.1 GEOLOGY
The stratigraphic sequence in the Diamond and Granby mining subdistricts consists of Paleozoicsediments ranging in age downward from the Pennsylvanian system through the Mississippian,Devonian, Ordovician, and Cambrian systems as shown on Figure 2-2. The Paleozoic sediments
P:\NEWTONCO\SREM\SREM.RV1 DAMES & MOORE
i
FIGURE 2-2
Generalized Section of Geologic Formationsand their Water Bearing Properties
£ a 3 DErTHTOTOP« § g THICKNESS • OF FORMATIONv\ W 0 RDRMATION (FEET) UTHOLOGY (FEET) WATER-SEARING CHARACTER
Yields tittle wtttar wnara massive,but can yield over ICO gpm in bracciatad areas.
solution cnsmebi may yia4d large supplies
Generally yields adequate supply fordomestic and stock use, rarefy over
50 gpm. juppB*s many springs
" • Generally yields adequate domesticor yjxic supply, supplies many sprinos
Yields vary imafl quantitias of water
Aquitard
Generally doaa rot yfetd water
• Aquitard
Yields snail quantities of water
Yfelds smal quantities of water
Generally yields good luppry of water;most suppies betwaan 50-150 gpm
Yields small suppOes of water
Generally yWds good supply of water, especiallyfrom lower portion; b«twean 50-4OO gpm
YMds small iupp4es of watar
Yields vary consJdanoiy, formationmay be absent over Precambriab nigns
GonaraKy doas not yicM water
unconforrnably overlie Precambrian igneous basement rocks. The sediments lying below thePennsylvanian clastic deposits are largely carbonate rocks composed of cherty limestones, chertydolomites, and dolomites, which in the Ordovician and Cambrian sections are interspersed withminor sandstones and shales. The dolomites and limestones in the mineralized zones of thestratigraphic section are commonly cherty and fractured. The average depth to the Precambrianrocks in the District is 1,700 to 1,800 feet, but has been reported as shallow as 290 feet in someportions of the District (Brockie, et al., 1968, Hagni, 1986). Mississippian cherty limestonesconstitute most of the surface exposure in southwestern Missouri. Pennsylvanian shales, sandstones,and siltstones, and occasional coal seams unconformably overlie the Mississippian cherty limestonesand occur as scattered erosional outliers.
The regional dip of the Paleozoic rocks (Cambrian through Pennsylvanian Systems on Figure 2-2)is to the northwest at relatively low angles of 15 to 20 feet per mile. Regional structure insouthwestern Missouri consists of a number of gently folded anticlines and synclines which plungeto the northwest away from the Ozark uplift (Brockie, et al., 1968). Local structure related tokarstification is common in the Mississippian units. Dissolution of the carbonate units resulted insinkholes and brecciated zones. The open areas created by carbonate dissolution typically hostedthe ores of the District (Hagni, 1986). The highly permeable sinkhole and brecciated zones alsocontain groundwater from the shallow aquifer (Feder, et al., 1969).
The major mineralized zones include the Fern Glen, Reeds Spring, Elsey Formation, undifferentiatedBurlington and Keokuk, and Warsaw Formations; the total thickness of these formations averages400 feet. The overlying undifferentiated Mississippian (Carterville) and Pennsylvanian (Cherokee)rocks often contain mineralization.
The major metallic minerals present in the region are sulfides: galena (PbS), sphalerite (ZnS),chalcopyrite (FeCuS2), pyrite (FeS2), and marcasite (FeS2) (Brockie et al., 1968). The sulfideminerals occur as open space fillings in breccia and other dissolution features. Some finer-grainedsulfides occur disseminated in secondary jasperoid near coarser-grained lead/zinc minerals (Hagni,1986). Small, near-surface deposits of oxidized lead/zinc minerals (oxides, sulfates, carbonates, andsilicates) also occur in the area and were mined early in the District's history (Stewart, 1987; Brockieetal., 1968).
P:\NEWTONCO\SREM\SREM.RV1 DAMES & MOORE
2.2.2 HYDROGEOLOGY
Ground-water throughout the District occurs in two aquifers, the "shallow aquifer" and the "deepaquifer," which are separated throughout most of the region by confining layers. The principalconfining layers are the Compton Limestone and the Northview Shale, which may be discontinuouswithin Newton County. The sequence of formations located immediately above and below theseprincipal confining layers also has low permeability and may act collectively as an aquitard becauseof their argillaceous character. These other sequences include the Fern Glen, Chattanooga Shale,Cotter Dolomite, and Jefferson City Dolomite Formations. In places, one or more of the confininglayers may change in lithology or thickness.
The shallow aquifer primarily occupies permeable zones in any of the Mississippian Formations.Mining generally occurred above and within the shallow aquifer units. Water yields from theshallow aquifer are highly variable and are controlled largely by karst features and highly permeablebreccias, joints, and bedding planes. Other areas of the aquifer consist of dense, massive limestoneswhich are unaffected by significant dissolution and brecciation, and generally have low yields(Feder, et al, 1969). Individual mineralized breccia areas, which are isolated and do not havesignificant hydraulic connections to other breccia areas, are known as "pools" or "jugs."Historically, during mining operations, 200 feet of head difference was observed between pools(Feder, et al., 1969).
Groundwater flow is generally to the west and northwest, or off the flank of the Ozark uplift. Therecharge to the shallow aquifer generally occurs locally through infiltration from the surface and ismost significant in the more permeable brecciated areas.
Depth to groundwater in the shallow aquifer ranges from 60 to 150 feet with reported yields of upto approximately 100 gallons per minute. Residents outside the city limits of Diamond and Granbyobtain drinking water from the shallow aquifer (Jacobs Engineering Group Inc., 1995).
The groundwater in the deep aquifer is contained in the Ordovician and Cambrian dolomites andsandstones. The deep aquifer consists primarily of the Roubidoux, Eminence, Potosi, and LamotteFormations, and is characterized by heterogeneous vertical and lateral hydraulic properties due tothe presence of solution channels and fractures (MacFarlane and Hathaway, 1987). Transmissivitiesin the deep aquifer are also variable with the largest transmissivities observed northwest of the Site.Available maps indicate the regional gradient of the deep aquifer is to the west and northwest
7
P:\NEWTONCO\SREM\SREM.RV1 DAMES & MOORE
(Schloss, 1986). The only observable changes in the deep aquifer potentiometric surface are causedby groundwater withdrawals. No effects of local precipitation on groundwater levels in the deepaquifer have been observed (Schloss, 1986).
The deep aquifer is recharged primarily by precipitation on Ordovician and Cambrian rocks wherethey outcrop in the core of the Ozark uplift, southeast of the Site. The deep aquifer has no apparentdischarge to the surface in the area.
The primary source of drinking water in Diamond and Granby is the deep aquifer. Schloss reportedin 1986 that the total water use from Newton County, including surface water, was approximately2.9 million gallons per day (mgd). Cities or communities in Newton County that derive all or partof their water supply from deep aquifer wells are listed on Table 2-1 and include Diamond, Fairview,Granby, Sarcoxie, Seneca, and Stark City.
Table 2-1Municipal Water Supply Facilities
Newton County, Missouri
Diamond
Fairview
Granby
Sarcoxie
Seneca
Stark City
0.288
0.216
0.691
1.050
1.836
0.025
0.085
0.014
0.337
0.155
0.496
0.050
Iwell
1 well
2 wells
2 wells
3 wells
Iwell
No
No
No
Yes
No
No
(Source: Schloss, 1986)
2.2.3 SURFACE WATER
The Diamond and Granby mining subdistricts are in the Spring River drainage basin. The tributaryto the Spring River which drains the subdistricts is Shoal Creek.
P:\NEWTONCO\SREM\SREM.RV1 DAMES & MOORE
The Diamond mining subdistrict is drained by Carver Branch and Baynham Branch, ephemeraltributaries to Shoal Creek. Carver Branch and Baynham Branch flow to the west and southwest,respectively. The Granby mining subdistrict is drained by Dry Branch, Wolf Creek, and Gurn SpringBranch. Dry Branch is an ephemeral stream as it flows westerly across the center of the Granbymining subdistrict, but it is a perennial stream as it flows north along the western boundary of thesubdistrict to Shoal Creek. Wolf Creek and Gum Spring Branch are ephemeral streams that flownorth through Granby and join northwest of the city before the confluence with Shoal Creek.Ground elevations range from 1,200 feet to 1,000 feet above mean sea level. The topography slopesgently to the northwest, except in areas adjacent to major drainages where local relief may be asgreat as 100 feet, with steep slopes.
2.2.4 SOILS
The majority of the soils in Newton County developed from one of three sources: Mississippiancherty limestone, Pennsylvanian shale and sandstone, or loess (wind-deposited material) whichcovered the region during the Pleistocene epoch. The soils consist of silry loam with red clay andchert, especially in the Granby mining subdistrict. Coarser soils are generally located in areas ofhilly topography. The soils are expected to exhibit moderate permeabilities (Missouri Departmentof Natural Resources, 1989).
2.2.5 CLIMATOLOGY
Newton County has a humid continental climate characterized by hot summers, cold winters, andhigh humidity. Summers are humid with occasional afternoon thundershowers. Severe weatheroccurs mainly in the form of thunderstorms and tornadoes. Missouri experiences an average of 40to 60 days per year with thunderstorms and an average of six tornadoes per month in April, May, andJune. December, January, and February are characterized by alternating freezing and thawingtemperatures.
The monthly mean temperature at Neosho, the county seat approximately eight miles southwest ofGranby, ranges from 34°F in January to 79°F in July. Daily temperature variations of about 20°Foccur year round. The average annual precipitation is 40.87 inches. Of this, approximately 60percent falls in April through September.
P:\NEWTONCO\SREM\SREM.RV1 DAMES & MOORE
Average monthly rainfall measured at Neosho during the period 1952-1980 ranges from 1.51 inchesin January to 4.82 inches in June, with a mean annual total of 40.87 inches. About 60 percent of therainfall occurs during the crop growing season (April to September). An annual average of 12.2inches of snowfall occurs during the coldest months, December through March, based on data from1952 to 1980 (U.S. Soil Conservation Service, 1989).
2.2.6 POPULATION AND LAND USE
The population of Newton County is 44,445 (1990 U.S. Census). In 1980, 35 percent of thepopulation lived in the urban centers and 65 percent in the rural areas (Schloss, 1986). Thepopulation of Diamond is 775 and the population of Granby is 1,945 (1990 U.S. Census).
Outside the city limits of Diamond and Granby, the primary land use within the mining subdistrictsis agricultural, including crop farming and pasture grazing. Industrial activities present within theDiamond and Granby mining subdistricts are related to aggregate production, light manufacturing,and construction.
10
P:\NEWTONCO\SREM\SREM.RV1 DAMES & MOORE
3.0 REMOVAL SITE EVALUATION INVESTIGATIONS
One objective of the Removal Site Evaluation was to characterize mining-related cadmium, lead,and zinc concentrations in drinking water and residential yard soils to support the EE/CA. Anotherobjective of the Removal Site Evaluation was to identify residences where time-critical removalaction levels for cadmium, lead, and zinc in drinking water or the action level for lead in residentialyard soils have been exceeded. The AOC provides the following time-critical removal action levelsfor drinking water:
• 5 micrograms per liter (p.g/L) for cadmium• 15 u,g/L for lead• 5,000 ug/L for zinc.
For soils located within defined yards or play areas of a residence, the AOC time-critical removalaction level is a soil lead concentration greater than 2,500 milligrams per kilogram (mg/kg).
To fulfill the objectives of the Removal Site Evaluation, rural drinking water and residential yardsoil investigations were conducted in the Diamond and Granby mining subdistricts. The drinkingwater component of the Removal Site Evaluation included sampling the potable water supplies toresidences outside the Diamond and Granby water supply district service areas. The residential yardsoil component included sampling of soils at the following:
• All occupied or habitable isolated residences• 20 percent of urban or grouped residences• All licensed day care center play areas.
For the purposes of the Removal Site Evaluation, an isolated residence was defined as an occupiedor habitable residence at least 500 feet from any other occupied or habitable residence as describedin the approved SAP (Dames & Moore, 1997).
3.1 RURAL DRINKING WATER INVESTIGATION
To identify residences outside the Diamond and Granby water supply districts, Dames & Mooreinitially contacted officials in the towns of Diamond and Granby to define the boundaries of the
11
P:\NEWTONCO\SREM\SREM.RV1 DAMES & MOORE
water district service areas. The City Clerks for Diamond and Granby defined the water districtboundaries as the city limits. According to Joanne Lamp, Granby City Clerk, the Diamond andGranby water districts are the only water districts that exist within the general boundaries of theDiamond and Granby mining subdistricts. The closest rural water district, outside the generalboundaries of the Diamond and Granby mining subdistricts, serves the Stark City/Newtonia area.
Dames & Moore attempted to contact the owners of all residences in the Diamond and Granbymining subdistricts, outside of the city limits of Diamond and Granby but within the Diamond andGranby mining subdistricts, to identify candidate wells to sample and to obtain permission to samplewells. If a resident was not home when contact was attempted, a written message was leftidentifying the date, time, and reason for the attempted contact along with a name and localtelephone number to call regarding sampling. If field personnel were unable to make contact witha property owner after two separate attempts, the residence was dropped from the list of candidateresidences.
At each residence where permission was granted, Dames & Moore collected a sample from the wellin accordance with the procedures specified in the approved SAP (Dames & Moore, 1997). If morethan one residence was connected to a private supply well, only one sample was collected.Permission to sample, along with domestic water use information (type(s) of water use, availabilityof alternate water supplies, consumption of bottled water, installation of point-of-use water treatmentsystems, etc.) was documented on field worksheets. If the homeowner or other responsibleindividual refused to allow samples to be collected, the refusal was noted on the field worksheets.Field worksheets are on file at Dames & Moore.
Approximately 180 rural residences were identified in the Diamond mining subdistrict as candidatesfor sampling. Dames & Moore sampled 100 domestic wells, which service 129 residences or 72percent of the candidate residences. Twenty homeowners (11 percent) in the Diamond miningsubdistrict refused to permit sampling. These refusals are likely the result of concerns regardingother on-going actions at the State and local level. Contact was not made with homeowners aftertwo attempts at 31 residences (17 percent). The locations of these residences were submitted to EPAat a meeting on March 4, 1998.
Approximately 180 residences were identified in the Granby mining subdistrict as candidates forsampling. Dames & Moore sampled 108 domestic wells, which service 143 residences or 79 percentof the candidate residences. Nine homeowners (five percent) in the Granby mining subdistrict
12
P:VNEWTONCO\SREM\SREM.RVi DAMES & MOORE
refused to permit sampling. Contact was not made after two attempts at 28 residences (16 percent).The locations of these residences were submitted to EPA at a meeting on March 4,1998.
Groundwater samples were analyzed for total (unfiltered) cadmium, lead, and zinc in accordancewith the methods specified in the approved SAP (Dames & Moore, 1997). Sample results areprovided in Appendix A and discussed in Section 4.0 of this memorandum.
3.2 RESIDENTIAL YARD SOIL INVESTIGATION
Dames & Moore attempted to contact owners of urban or grouped residences in the Diamond andGranby mining subdistricts in person. If a resident was not home when contact was attempted, andthe residence was in close proximity to former mining or smelting activity, a written message wasleft identifying the date, time, and reason for the attempted contact along with a name and localtelephone number to call regarding sampling. If a homeowner or other responsible individual wasnot home when contact was attempted, and there was no evidence of former mining activity in closeproximity to the residence, field personnel attempted to get permission to sample the yard soil atthe adjacent residence. If field personnel were unable to make contact with a property owner aftertwo separate attempts, the residence was dropped from the list of candidate residences.
At each residence where permission was granted, a minimum of two composite soil samples, onefront yard composite and one back yard composite were collected. A minimum of three compositesoil samples were collected at residences with yards in excess of 10,000 square feet in area. Ifgardens and/or play areas were present, separate composite soil samples were collected from each.Permission to sample, along with a diagram depicting the general layout of the property withassociated buildings, yards, play areas, garden, and other Site features were documented on fieldworksheets.
Composite soil samples were collected at 66 (21 percent) of the urban residences in the Diamondmining subdistrict and 198 (26 percent) of the urban residences in the Granby mining subdistrict.Composite soil samples were also collected at all residences where drinking water was sampled.Composite soil samples consisted of four subsamples (0- to 2-inch depth interval) distributed asequally as possible in the area to be sampled.
Each composite soil sample was given a unique alpha-numeric code, identifying the town and thelocation in the yard where the composite sample was collected. Composite soil samples collected
13
P:\NEWTONCO\SREM\SREM.RV1 DAMES & MOORE
at urban residences were assigned a D or G identifier for Diamond or Granby respectively, followedby a second letter indicating location relative to the residence (F denotes front yard, B back yard,S side yard, G garden, P play area). Composite soil samples collected in rural areas were assigneda D or G identifier followed by an S for soil, a unique number, and a letter (N, S, E, W) designatingthe location of the composite sample relative to the residence. Due to space limitations, only thenumeric identifiers are shown on the figures in Section 4.0.
Composite soil samples were analyzed for lead and zinc in the field using a field portable x-rayfluorescence spectrometer (XRF). Results for samples analyzed by XRF are provided in AppendixA. If the XRF lead assay concentration for any composite soil sample exceeded 500 mg/kg,delineation sampling was conducted to define the extent of elevated lead concentrations. Delineationsampling consisted of sampling the residential yards adjacent to and in the 90 degree sectors fromthe residential yards where XRF assays indicated lead concentrations greater than 500 mg/kg.Delineation sampling continued in an area until the soil lead concentration, as measured with theXRF, was less than 500 mg/kg at each residential yard sampled in the four sectors.
Ten percent of the composite samples analyzed by XRF were submitted to Core Laboratories, Inc.for confirmation analysis. Confirmation samples were analyzed for total cadmium, lead, and zincusing SW-846 Method 6010A. Laboratory results are provided in Appendix A. Laboratory resultswere used to calculate the Site-specific cadmium to zinc ratio which was used to estimate thecadmium concentrations in samples analyzed by XRF.
14
P:\bffiWTONCO\SREM\SREM-RVl DAMES & MOORE
4.0 DISCUSSION OF RESULTS
Laboratory data packages were reviewed for submission of all documentation required to assess QCelements contained in the analytical methods and EPA guidelines for Contract Laboratory Programdata review. Laboratory data were assessed relative to the project data quality objectives. Thereview included a review of the following:
• Overview of the data package for inclusion of all appropriate raw data• Calculation of holding times for all analytes• Review of transcriptions, calculations, and raw data documentation• Evaluation of all QC samples for required frequency of analysis and required control limits.
During data review EPA Contract Laboratory Program data review guidelines were used to qualifydata. Although some data points are qualified as estimated, they are still considered usable for thepurposes of this project. The QA/QC assessment report for the laboratory data is provided inAppendix B.
Precision of the XRF data was evaluated by analyzing check samples at a five percent frequency.Check samples are selected from the Site-specific samples used to calibrate the XRF. Theconcentrations of the check samples were approximately midrange on the calibration curves.Precision of the XRF data for this project was acceptable. An evaluation of the check sample resultsis provided in the QA/QC assessment report (Appendix B).
Accuracy of the XRF data was evaluated by comparing the XRF results with confirmation sampleresults. Accuracy of the XRF data for this project was acceptable. Accuracy of the XRJF data isexplained further in the QA/QC assessment report provided in Appendix B.
Analytical results and the significance of these results are discussed in this section.
4.1 RURAL DRINKING WATER RESULTS
The objectives of the rural drinking water investigation were to characterize mining-relatedcadmium, lead, and zinc concentrations in drinking water and to identify domestic wells where AOCtime-critical removal action levels for metals have been exceeded. Mining-related metal
15
P:\NEWTONCO\SREM\SREM.RVI DAMES & MOORE
concentrations in domestic well water can be differentiated from other, nonmining-related sourcesusing the zinc/lead relationship described in the AOC. Based on this relationship, for metalsconcentrations in wells to be attributed to mining-related activities, zinc concentrations must begreater than 500 u.g/L and the ratio of the concentration of zinc to the concentration of lead (zinc/leadratio) must approach or exceed 100.
A total of 208 domestic well samples were collected, 100 from the Diamond mining subdistrict and108 from the Granby mining subdistrict. Twenty-six of the 208 domestic well samples (13 percent)exceed the AOC time-critical removal action level for cadmium, lead, and/or zinc. Analytical resultsand zinc/lead ratios for these samples are shown on Table 4-1. The AOC time-critical removalaction levels for cadmium and lead were exceeded at a 27th well, well DW-39, however, water fromthis well is not used for domestic purposes. Sampling locations are shown on Figures 4-1 and 4-2.
Eleven of the 100 drinking water samples (11 percent) collected in the Diamond mining subdistrictexceed the AOC time-critical removal action level for cadmium and/or lead. The zinc/lead ratiosfor all 11 samples are less than 100 and the zinc concentrations for nine of the 11 samples are lessthan 500 u.g/L. Therefore, based on the zinc/lead relationship criteria specified in the AOC, metalsconcentrations in these wells may not be attributable to mining-related activities.
Fifteen of the 108 drinking water samples (14 percent) collected in the Granby mining subdistrictexceed AOC time-critical removal action levels for cadmium, lead, and/or zinc. However, based onthe zinc/lead relationship criteria specified in the AOC, metals concentrations in only six of thesewells (six percent) may be attributable to mining-related activities.
16
P:\NEWTONCO\SREM\SREM.RV1 DAMES & MOORE
Table 4-1Removal Site Evaluation
Selected Results for Domestic Wells
liilplliliSsSWKJSSraSSJSiSBJSli
DW-16
DW-17
DW-19
DW-43
DW-45
DW-46
DW-57
DW-59
DW-92
DW-93
DW-94
GW-519
GW-520
GW-522
GW-534
GW-575
GW-581
GW-584
GW-585
GW-590
GW-594
GW-601
GW-602
2.5
1.0
<0.6
0.7
5.8
6.2
8.6
<0.6
1.9
0.8
1.1
19.9
4.9
24.5
2.4
7.3
5.1
8.6
15.5
2.1
1
16.7
8
32
23
18
24
16
104
15
88
29
16
31
<2
17
2
36
<2
37
<2
<2
87
26
<2
42
423
141
719
83
216
845
426
142
187
107
191
1,580
1,180
5,880
1,690
342
648
2,470
1,120
4,210
294
4,220
2,320
No
No
Yes
No
No
Yes
No
No
No
No
No
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
Yes
No
Yes
Yes
No
No
No
No
No
No
No
No
No
No
No
Yes
No
Yes
No
Yes
No
Yes
Yes
No
No
Yes
No
17
P:\NE WTONCO\SREM\SREM.RV1 DAMES & MOORE
Table 4-1Removal Site Evaluation
Selected Results for Domestic Wells
l^upMH^mmmmmmmS:GW-604
GW-605
GW-607
20.2
11.1
<0.6
<2
11
23
2,580
381
1,910
Yes
No
Yes
Yes
No
NoThe time-critical removal action levels for cadmium and lead were exceeded for DW-39; however, water from this wellis not used for domestic purposes.Time-critical removal action levels are: cadmium - 5 ug/L; lead - 15 ug/L; zinc - 5,000 ug/L.Time-critical removal action level exceedances are shown in bold-face type.
18
PANE WTONCO\SREM\SREM.RVI DAMES & MOORE
4.2 RESIDENTIAL YARD SOIL RESULTS
Soil samples were analyzed for lead and zinc in the field using an XRF. Ten percent of the soilsamples were submitted to a laboratory for confirmation of the XRF results. The confirmationsamples were analyzed for cadmium, lead, and zinc. All XRF results and laboratory analyticalresults for the confirmation samples are provided in Appendix A.
Because cadmium is found as a replacement for zinc in the lattice structure of sphalerite (zincsulfide), a Site-specific cadmium to zinc ratio can be calculated. The average cadmium and zincconcentrations of 169 XRF confirmation samples were used to calculate the Site-specific ratio. TheSite-specific cadmium to zinc ratio is 1:436. For example, if a given sample yields 5,000 mg/kgzinc, then the cadmium concentration would be expected to be approximately 11 mg/kg. Thiscadmium to zinc ratio was used to estimate the cadmium concentrations of the samples analyzed byXRF. Estimated cadmium concentrations are provided in Appendix A.
The objectives of the residential yard soil investigation were to characterize mining-relatedcadmium, lead, and zinc concentrations in residential yard soils and to identify residences where thetime-critical action level for lead has been exceeded. Care was taken when selecting soil samplelocations to avoid potential influences from nonmining-related sources such as lead-based paint andautomobile emissions.
Over 1,400 composite soil samples were collected at 471 urban and rural residential yards in theDiamond and Granby mining subdistricts. The urban residential yard soil sampling locations areshown on Figures 4-3 through 4-5. The lead concentration of composite soil samples collected atten urban residences exceeded the AOC action level for lead of 2,500 mg/kg.
None of the composite soil samples collected in the Diamond mining subdistrict or rural yards in theGranby mining subdistrict exceeded the AOC time-critical removal action level for lead. The leadconcentration of composite soil samples collected at ten of the 198 urban yards sampled in theGranby mining subdistrict exceed the AOC action level for lead. The lead concentrations for thesesamples range from 2,542 mg/kg to 7,028 mg/kg as shown on Table 4-2.
21
P:\NEWTONCO\SREM\SREM.RV1 DAMES & MOORE
x> c" jr8 §
1
•2.U3
o"sa3a.
OLQ. B
CO O
' D £ °3C Q
OJ *"**>~j co D 13a a • a a ~a D ., ———
Fearl St. Q . f . „
OJ O ^4 O O
" a O B 0 0 a,/, Rave Dr. —————————« n a a B
1 OJ 1 ° Dco o 3 oj/•*- O.. OJ
, . , „ , ODD "a • Dn a no ana
M a n
CDDDDs^ a D a a a a m
jf OJ
r OJB D n D X • •a to n a a n
SCHOOL jj; BSi!
° a4' a* 5-
n y
Bobbie Smith SL -.. ,-°°°D.-DD
• a DD B01 rjoj01 ™
a B S«
a
D XDepot St. - ———————
r*-
CO
Thurman St.
George St. Q Q ^
C
ifn
X a "a
&"a a -"a•a on 8 D
C °*
a a DOJS a
D p Dn y
a n • nD a • nan
yi 01»OJ *•
OJ „J-]l£> Q
aa a a n
a a B n B aOJ OjCj 00 D
D X a aa a a«
a K3 OJa D o « DP
D «OJ D
^.^
a «.D o.a
L KJ MU. g>
' a.en X
naa
OJ _
a
o5"M
?£
o«to
a •""
n 55§' !?
a a
o, •en ojB & tn
00
a
3 DD aDOJ D
• g D
3p I
a "</a rn a
n
D
DDOJ OJen OJ•" • D
cr** /—— Hamilton St. "§_ /
fe/§
/f-1 ———————————
Washington St. / *^v
o j Q a n / a D ° b o j£ a a / n g
D Main St. a B a a o D/ D a a D a B1^ a
Locust a St. M X ,.§ * ^a)gf u SQ-a n ' n » co
D o^n r**• D
0) OJ >
• 0y a a • a a a£ ————————— Cullum Ct.
D B D B a a a^ OJ OJ
^ OJ
OJ
r Iow}
)shkosh St.
Odessa St.
a XD )g( P
X
m
Za5-2-co
fc-J
Panda Rd.
CTQ13
(D
Q
-<
Q.Q
o> S Or-IZO D ^OCQ
- 5T
EXPIANAHON
^° • Urban residential yard soil sampling location and Identification number22 B Bolded number Indicates lead concentration exceeds
3,340!B - back yard; F - front yard; S - side yard'Laboratory result
Several attempts were made to conduct delineation yard soil sampling at residences adjacent to thosewith soil lead concentrations greater than 500 mg/kg. However, approximately 52 property ownerswere not home when contact was attempted. The addresses of residences where delineationsampling was required, but contact with property owners was not made, were provided to EPA ata meeting on March 4,1998 and in a letter dated September 2,1998.
25
PANE WTONCO\SREM\SREM.RV1 DAMES & MOORE
Four of the ten residences (G-22, G-89, G-124, and G-160) are located immediately east and southof the smelter formerly located on McKinley Street. Due to the proximity of the residences to theformer smelter, the soils at these residences are likely impacted by past smelting-related activities.
Sample GF-44 was collected from the front yard at a residence located more than 1200 feet from thenearest known mining area. Lead concentrations for two other soil samples collected at the sameresidence are considerably lower than the lead concentration for GF-44, as are soil leadconcentrations at adjacent residences.
Sample GB-68 was collected at a residence built on a chat mortar foundation. The back porch stepsat the residence where GB-50 was collected were also constructed of a chat mortar and the housepaint was peeling. These residences are located more than 900 feet from the nearest mining area.
The residence at which sample GB-84 was collected is located east of the Fortune Teller tailings pileand north of the Goade Mill tailings pile in southeast Granby. Lead concentrations for three othersoil samples collected at the same residence are considerably lower than the lead concentration forGB-84, as are soil lead concentrations at adjacent residences.
Sample GB-156 was collected from the front yard at a residence located more than 1,300 feet fromthe nearest known mining area. Lead concentrations for two other samples collected at the sameresidence are considerable lower than the lead concentration for GB-156.
Sample GB-174 was from the back yard at a residence with peeling house paint. This residence islocated nearly one-half mile from the nearest known mining area. Lead concentrations for two othersamples collected at the same residence are less than 350 mg/kg, as are soil lead concentrations atadjacent residences.
hi summary, lead concentrations for composite soil samples collected at ten urban residential yardsin the Granby mining subdistrict exceed the AOC time-critical removal action level for lead of 2,500mg/kg. Four of the ten residences are located in close proximity to the former smelter. Only oneof at least three samples collected at each of the other six residences exceed the time-critical removalaction level for lead. Lead concentrations for other samples collected at these six residences areconsiderably lower than the concentrations of the samples that exceeded the time-critical removalaction level.
.26
P:\NEWTONCO\SREM\SREM.RV1 - DAMES & MOORE
5.0 REFERENCES
American Zinc Lead and Smelting Company, 1924. Main Office, Custom Mill and Lead Smeltermap.
Brockie, Douglas C, Edward H. Hare, Jr., and Paul R. Dingess, 1968. The Geology and OreDeposits of the Tri-State District of Missouri, Kansas, and Oklahoma in Ridge, J.D., ed., Oredeposits in the United States, 1933-1967, The Graton-Sales Volume, The American Instituteof Mining, Metallurgical, and Petroleum Engineers, Inc. New York, Volume 1, p. 400-430.
Dames & Moore, 1995. Final Remedial Investigation, Neck/Alba, Snap, Oronogo/Duenweg, Joplin,Thorns, Carl Junction, and Waco Designated Areas, Jasper County Site, Jasper County,Missouri.
Dames & Moore, 1992. Sampling and Analysis Plan, Neck/Alba, Snap, Oronogo/Duenweg, Joplin,Thorns, Carl Junction, and Waco Designated Areas, Jasper County Site, Jasper County,Missouri.
Feder, G.L., John Skelton, E.G. Jeffery, E.J. Harvey, 1969. Water Resources of the Joplin Area,MO. Missouri Geological Survey and Water Resources Report 24, p. 97.
Hagni, R.P., 1986. A summary of the geology of the ore deposits of the Tri-State District, Missouri,Kansas, Oklahoma: in Guidebook to the Geology and Environmental Concerns in the Tri-State Lead-Zinc District, Missouri, Kansas, Oklahoma. Association of Missouri Geologists,p. 30-46.
Jacobs Engineering Group Inc., 1995. Final Expanded Site Inspection Report for Newton CountyMine Tailings Site Newton County, Missouri.
Lamp, Joanne, 1998. Personal communication. Telephone conversation on September 1, 1998 withDan Lincicome of Dames & Moore.
27
P:\NEWTONCCASREM\SREM.RV1 DAMES & MOORE
MacFarlane, P.A. and L.R. Hathaway, 1987. The hydrogeology and chemical quality of groundwater from the lower Paleozoic aquifers in the Tri-State Region of Kansas, Missouri, andOklahoma. Kansas Geological Survey, Ground-water Series 9, 37 p.
Missouri Department of Natural Resources, 1989. Narrative Site Inspection Report Tri-State Mining- Granby Missouri.
Schloss, J.A., 1986. Changes in the potentiometric surface of deep aquifer in the Joplin area,Missouri, 1900-present. Unpublished Master's Thesis, Southwest Missouri State University,73 p.
Shrader, W.D., M.E. Springer, R. Hamby, W.J. Pettijohn, and J.T. Miller, 1954. Soil survey ofJasper County, Missouri. U.S. Department of Agriculture Series 1942, No. 5, 67 p.
Smith, W.S.T. and E.G. Siebenthal, 1907. Joplin district (Missouri/Kansas). U.S. GeologicalSurvey Atlas, Folio 148.
Spruill, T.B., 1987. Assessment of water resources in lead-zinc mined areas in Cherokee County,Kansas and adjacent areas. U.S. Geological survey Water Supply Paper 2258.
Stewart, D.R., 1987. Natural sources of contamination, ground and surface waters, Kansas-Oklahoma portion, Tri-State District. Private report prepared for Gold Fields MiningCorporation, 34 p.
U.S. Department of Commerce Economics and Statistics Administration Bureau of the Census,1991. 1990 Census of Population and Housing, Summary Population and HousingCharacteristics, Missouri.
U.S. Environmental Protection Agency Office of Solid Waste and Emergency Response, 1986. TestMethods for Evaluating Solid Waste, Physical/Chemical Methods.
U.S. Soil Conservation Service, 1989. Soil Survey of Newton County, Missouri.
28
P:\NEWTONCO\SREM\SREM.RV1 DAMES & MOORE
APPENDIX A
ANALYTICAL RESULTS
GROUND-WATER RESULTSNewton County, Missouri Removal Site Evaluation
U - Not detected at the level shown.J - Estimated quantity.UJ - Estimated as not detected at the level shown.
P:\NEWTONCO\DATA\LAB_DAT.XLS (RINSE BLANK)
APPENDIX B
QA/QC ASSESSMENT OF DATA
LABORATORY DATAQUALITY ASSURANCE/QUALITY CONTROL
ASSESSMENT
Removal Site EvaluationDiamond and Granby Mining Subdistricts
Newton County, Missouri
1.0 INTRODUCTION
A review of the analytical results and associated supporting documentation was conducted to assessdata quality and usability for 169 residential yard soil samples and 208 drinking water samplescollected during the Removal Site Evaluation conducted in the Diamond and Granby miningsubdistricts of Missouri. The evaluation criteria used were those specified in the methods and thoseoutlined in the USEPA Contract Laboratory Program National Functional Guidelines for InorganicData Review, February 1994.
The purpose of this report is to present the findings of the quality assurance/ quality controlassessment of the laboratory data and summarize our opinion of the usability of the data. Thisreport is organized into the following sections:
• Inorganic Data Validation• Field Quality Control Review• Data Usability.
2.0 INORGANIC DATA VALIDATION
Data validation is a systematic process for reviewing a body of data against a set of criteria toprovide assurance that the data are adequate for their intended use. For the Removal Site Evaluation,data validation consisted of an evaluation of the following:
• Holding times• Instrument calibration• Laboratory blanks• Interference check samples• Laboratory control samples• Laboratory duplicates• Matrix spikes• Serial dilutions.
Holding Times
The holding times for metals analyses were assessed by comparing the sampling date with the dateof analysis. All samples were analyzed within the established holding time of 180 days.
P:\NEWTONCO\SREM\LABDATA.RPT
Instrument Calibration
Instrument calibration is performed to ensure that the instruments are capable of producingacceptable quantitative data. Initial calibration demonstrates that the instruments are capable ofacceptable performance prior to sample analysis, and continuing calibration sample analyses verifythat the initial calibrations are still valid. Initial and continuing calibration verification sampleresults were within the acceptable control limits of 90-110 percent recovery throughout the analysisof the Newton County Removal Site Evaluation samples.
Laboratory Blanks
The assessment of laboratory blank sample results is conducted to determine the existence andmagnitude of contamination resulting from laboratory activities. The criteria for evaluation of blankresults applies to any blank associated with the samples (i.e., method blanks or calibration blanks).No contaminants should be present in any blank at concentrations greater than the instrumentdetection limit (DDL). Sample results greater than the IDL, but less than five times the laboratoryblank concentrations are qualified.
Calibration blanks were analyzed at a 10 percent frequency of sample throughput. A method blankwas analyzed with each batch of samples. A trace concentration of cadmium was detected in onecalibration blank analyzed with soil samples, potentially impacting one sample result; the cadmiumresult for this sample is estimated as not detected. Trace concentrations of lead and zinc weredetected in selected continuing calibration blanks analyzed with equipment rinsate samples. Leadresults for 12 equipment rinsate samples and zinc results for 41 equipment rinsate samples areestimated as not detected. Trace concentrations of cadmium and zinc were detected in two methodblanks analyzed with groundwater samples. Cadmium results for two groundwater samples and zincresults for three groundwater samples are estimated as not detected. Lead was detected in one initialcalibration sample analyzed with groundwater samples, potentially impacting six sample results.Lead results for these six samples are estimated as not detected.
Interference Check Samples (ICS)
The ICS analyses verify the laboratory's interelement and background correction factors for theinductively coupled plasma (ICP) instrument. The ICS results must be within + 20 percent of thetrue value for all analytes in the ICS solution. An ICS was analyzed at the beginning and end of eachanalytical run with acceptable results thus verifying the acceptability of the correction factors.
P:\NEWTONCO\SREM\LABDATA.RPT
Laboratory Control Samples (TLCS)
The LCSs serve as monitors of the overall performance of all steps in the preparation and analysisprocess. LCSs were analyzed with each batch of samples. LCS recoveries were within the controllimits of 80 to 120 percent for water LCSs and within the control limits provided by themanufacturer for soil LCSs.
Laboratory Duplicates
Laboratory duplicate analyses are indicators of laboratory precision based on each sample matrix.For soil samples, a control limit of ± 35 percent was used to evaluate the relative percent difference(RPD) between duplicate results greater than five times the Contract Laboratory Program (CLP)contract required detection limits (CRDL). A control limit of ± two times the CRDL was used toevaluate soil duplicate results less than five times the CRDL. For water samples, a control limit of±20 percent was used to evaluate the RPD between duplicate results greater than five times the CLPCRDL. A control limit of ± the CLP CRDL was used to evaluate duplicate water results less thanfive times the CLP CRDL.
The RPDs for two soil duplicates for cadmium and one soil duplicate for zinc did not meet criteria;therefore, 48 cadmium results and 10 zinc results are qualified as estimates. The RPD for one waterduplicate for zinc did not meet criteria; therefore, 20 zinc results are qualified as estimates.
Matrix Spikes
Matrix spike sample results provide information about the effect of each sample matrix on thepreparation and measurement methodology. Control limits for matrix spike recovery are 75 to 125percent. Matrix spikes were analyzed at a frequency often percent.
The matrix spike recovery for lead was low for one soil matrix spike. Lead results for 10 soilsamples are qualified as estimates.
Serial Dilution
Serial dilution determines whether significant physical or chemical interferences exist due to samplematrix. If the concentration of an analyte is minimally a factor of 50 above the IDL, an analysis ofa 5-fold dilution should agree within 10 percent of the original results. Serial dilutions wereanalyzed at an 10 percent frequency.
P:\NEWTONCO\SREM\LABDATA.RPT 3
Serial dilution criteria were not met for cadmium and zinc for one serial dilution performed on a soilsample. Serial dilution criteria were not met for lead for two serial dilutions on soil samples.Cadmium results for 15 soil samples, lead results for 40 soil samples, and zinc results for 25 soilsamples are qualified as estimates. Serial dilution criteria were not met for one lead serial dilutionand two zinc serial dilutions performed on groundwater samples. Lead results for six groundwatersamples and zinc results for 58 groundwater samples are qualified as estimates.
3.0 FIELD QUALITY CONTROL REVIEW
Seventy-two equipment rinsate blanks and ten groundwater blind field duplicates were submittedto the laboratory for an assessment of field precision. Cadmium was detected in equipment rinsateblanks, but at concentrations below the CLP CRDL. Lead and zinc were also detected in selectedequipment rinsate blanks, but at concentrations slightly above the CLP CRDL. The samplesassociated with these equipment rinsate blanks were evaluated and it appears that the potentialimpact to the sample results is negligible. The RPDs for the ground-water blind duplicate resultsmet criteria for 87 percent of the data points indicating acceptable precision.
4.0 DATA USABILITY
Completeness of this data set was assessed by calculating the percentage of valid data points to thetotal data set. The completeness of this data set is 100 percent. Although some data points arequalified, as estimated, they still are considered usable for the purposes of this project.
This data set represents QA/QC Objective 3 data as defined in "Quality Assurance/Quality ControlGuidance for Removal Activities," (EPA, 1990). QA/QC Objective 3 data can be used for thefollowing purposes:
Site characterizationEvaluation of alternativesEngineering design.
P:\NEWTONCO\SREM\LABDATA.RPT
XRFDATAQUALITY ASSURANCE/QUALITY CONTROL
ASSESSMENT
Removal Site EvaluationDiamond and Granby Mining Subdistricts
Newton County, Missouri
1.0 X-RAY FLUORESCENCE THEORY AND OPERATION
X-ray fluorescence (XRF) spectroscopy is an analytical technique that provides a non-destructivequantitative analysis of the elemental composition of a sample. XRF theory is based on x-rayfluorescence spectroscopy whereby a sample is exposed to an energy source from either aradioactive or x-ray emitting source. The x-ray energy impinges on the electron cloud of theelemental atom causing vacancies in the inner shell(s) of the atom. The vacancies createinstability within the structure of the atom. The atom regains stability by filling the inner valanceswith electrons from the outer shell of the atom. When outer shell electrons fill the vacancy in theinner valance, the atom emits energy (fluorescence) because the electron is in a lower energystate. The fluorescence has a specific energy level that is characteristic of the element in whichit was produced. The total fluorescence emitted from a sample is proportional to the concentrationof the originating element in the sample.
The main advantage of using XRF for environmental applications is that the analytical equipmentis portable and is readily adapted for use in the field. Preparing samples for XRF analysisrequires minimal processing and analytical results can be obtained within minutes of samplecollection. The XRF used for this study was the Metorex X-Met 920 energy dispersive XRFspectrometer (X-Met). The X-met is battery operated and employs a surface analysis probe whichis specifically designed for field use.
The X-Met has sufficient memory to store thirty-two models and can be calibrated to analyze upto six different elements. Depending on the radioactive source used in the probe, the X-Met canbe calibrated to measure many different elements. The analysis probe used for this studycontained one radioactive source, Curium 244, well suited for lead studies.
2.0 XRF CALIBRATION
The X-Met was calibrated using site-specific samples (SSSs) collected from the study area priorto the Removal Site Evaluation. SSSs were used to calibrate the X-Met because they likelyrepresent the expected concentration range of the target metal(s) and relative backgroundcomposition of other metals in the soil.
All of the SSSs were homogenized and split into two equal parts. One part, the laboratory split,
was submitted for chemical analysis using SW-846 Method 3050 for the digestion and SW-846Method 6010A for the analysis. The other part, the calibration split, was used to calibrate the X-Met. To calibrate the X-Met, each calibration split was measured on the X-Met followed bymanually entering the result of the laboratory split into the X-Met. The calibration was completedwhen all the calibration splits and their respective results were entered into the X-Met.
The X-Met computes a linear regression based on the measured x-ray intensity of the calibrationsplits versus the actual chemical concentration of the laboratory splits. The linear regression,called a model, calculates chemical concentrations in unknown environmental samples of unknownconcentration based on the x-ray intensities of those samples as compared to the SSSs. As longas the elemental concentration in the unknown sample is within the range of the regression resultsare comparable to the digestive chemistry analyses of the SSSs. In addition to similar elementalconcentrations, the background chemical composition of the unknown sample should be similarto the background chemical composition of the SSSs.
Once the instrument is calibrated, periodic recalibration is not needed as long as soil conditionsdo not change significantly. The Instrument contains software which corrects for calibration shiftsdue to changes in ambient temperature. Additionally, known calibration check samples areperiodically measured to verify proper instrument function.
Three site-specific calibration models were developed using the results from samples previouslycollected and from samples collected during the first week of field activities. These samples wereused to construct the "low lead," "high lead," and "zinc" models. The SSSs were selected tospan the expected range of lead concentrations encountered during the study, as shown in Table1, Model Calibration Information.
Table 1, Model Calibration Information
Low Lead 20 44-1,470 ppm 457 ppm Residential Yards
High Lead 34 100-4,030 ppm 457 ppm Soils with Pb greater than600ppm on the low lead
model
Zinc 20 152-26,000 ppm 7360 ppm Residential Yards
Samples which read higher than 600 ppm on the XRF were reanalyzed on the "high lead" modelbecause the expected higher concentration of the sample would be closer to the midpoint of thehigher-range calibration curve. Any sample which was near the limits of the calibration curvewere reanalyzed on another model which closer approximated the expected concentration value.
During the calibration of the XRF, only SSSs with detectable levels of lead and ziac were usedto calibrate the instrument. Non-detect SSSs were never used to calibrate the instrumentregression.
3.0 XRF ANALYSIS
Each composite soil sample was sieved through a U.S. Standard Sieve No. 10 and placed in aSPEX® cup that was sealed with polypropylene film. The soil in each SPEX® cup was analyzedby placing the SPEX® cup on the XRF instrument probe. The samples were exposed to theradioactive source for one, 60-second interval. At the conclusion of analysis, the XRF displaysthe sample concentration in parts per million and the x-ray intensities of the target sample andbackground metals. Sample results are provided in Appendix A of the Site Removal EvaluationMemorandum.
The majority of the samples were analyzed using the zinc model and low lead model. Sampleswith a lead concentration greater than 600 ppm were reanalyzed using the high lead model. The
accepted result was based on comparing x-ray intensities of the sample to the x-ray intensities ofthe model mid-range calibration standard. If the x-ray intensities for lead, zinc, and arsenic(arsenic commonly interferes with lead readings) were within 15 percent of the x-ray intensitiesof the mid-range calibration standard for the given model, the result was considered valid.
4.0 QUALITY ASSURANCE
Quality assurance (QA) is a critical aspect of an XRF sampling program. The precision of XRPanalyses is determined by periodic measurements of the mid-range calibration sample. Theaccuracy of XRF analyses is assessed by comparing XRF data with corresponding samplessubmitted for laboratory confirmation. The precision and accuracy of the Removal SiteEvaluation XRF data is discussed in the following subsections
4.1 PRECISION
Precision of XRF data can be affected in several ways, including instrument variability (changesin ambient temperature and gain control), matrix effects, and sample variability. Instrumentvariability results from changes in the radioactive source. Because radioactivity is random, thex-rays which impinge on the atom vary. This causes the instrument to detect x-rays at slightlydifferent intensities (over the course of a sample measurement and between measurementintervals). Because the x-ray intensity is directly related to determining sample concentration, theprecision of the instrument and resulting sample result is affected. The electronics of the XRFare also sensitive to rapid temperature changes. Precision degrades if too many samples aremeasured consecutively, due to a phenomena called instrument drift. The instrument periodicallyneeds to "rest" after every 6 to 10 measurements.
The matrix of the sample will also affect precision. Matrix effects can be caused by the elementalcomposition of the sample, the particle size and shape, and homogeneity. Precision will bediminished if the matrix of the sample is different than the matrix of the SSSs.
Intrasample variability will cause degradation of precision. Intrasample variability is site specificand relates to varying soil types of the study area. For this investigation matrix effects andintrasample variability were minimized by consistent sample preparation and constructingcalibration models that accounted for variable soil conditions.
Precision of the XRF is determined by periodic measurements of the mid-range calibrationstandard for a given model. During XRF operations, the precision of the instrument is routinelychecked by analyzing a sample of known concentration, called the check sample. The checksample is selected from the SSSs used to calibrate the XRF. The sample used to check precisionhas a concentration which is approximately mid-range on the SSS calibration curve. A checksample is selected for each model and is measured at a five percent (1 for every 20 XRF analyses)frequency. After each XRF measurement, the microprocessor computes the standard deviationfor that reading. According to standard operating procedures for the instrument, if the measuredconcentration was greater than ±35 % of the known concentration of the standard, the instrumentmust stabilize (gain control) for five minutes. During the study, the mid-range check standardsdid not exceed the 35% target range, and thus, none of the previous 20 samples requiredreanalysis. A plot of the check sample results shown in Figure 1, Lead Check Sample Results,and Figure 2, Zinc Check Sample Results.
4.2 ACCURACY
Accuracy of XRF data is directly related to errors in construction of the model, and thusdiscrepancies are systematic. For example, if the SSSs used to construct the model do not havean even range, (i.e. a disproportionate number of samples with high metal concentration ratherthan an even gradation from low to high), then the model may be biased toward detecting metalswithin the high range. In this case, soils with lower metal concentrations may consistently readhigher than the actual concentration.
To check for accuracy, XRF samples are split and submitted for laboratory confirmation. For theRemoval Site Evaluation, 10 percent of the samples analyzed by XRF were submitted forlaboratory confirmation. The confirmatory samples serve to insure accuracy of the instrument andsampling program. Accuracy is measured by relative percent deviation (RPD) and is calculatedas follows:
XRF assays and laboratory confirmation results for all models were compared on scatter plots tographically depict systematic trends in XRF performance. The scatter plots were produced byplotting XRF analyses versus the laboratory confirmation results, as shown on Figures 3 through5. XRF results are depicted as diamonds on the plots. The 45 degree line, or correlation line,is the ideal one-to-one correlation between XRF and lab results. Deviations from the 45 degreeline can be used to assess both the accuracy and precision of the XRF analyses over allconcentration ranges of the model. Each model is evaluated individually and discussed below.
5.1 LOW LEAD MODEL
Overall, the low lead model functioned consistently, even below the computed instrumentdetection limit. Figure 4, Low Lead Scatterplot, reveals an even distribution of data points aboveand below the correlation line. The XRF had a tendency to overestimate lead values, and thusXRF results tended to be more conservative.
The main reason for overestimation is related to the soil chemistry within the study area. Soilchemistry in Newton County has been variously altered by the introduction of lead and othermetals from historic mining- and smelting-related activities, lead-based paint, non-native fill,automobile emissions, and industrial activities. The cause of overestimation is related to thedifference between the target and background composition of the SSS and the field samples. Fieldsamples with significantly different, metal chemistry than the SSS had different x-ray intensitiesand the lead in these samples were masked by the other interfering metals in the sample.Overestimate is pronounced in samples with very high zinc concentrations, relative to lead.
5.2 HIGH LEAD MODEL
As shown in Figure 4, High Lead Scatterplot, the XRF performed well within the calibrationlimits of this model (100 ppm to 4030 ppm lead). Elevated zinc levels are known to mask leadintensities. If zinc concentrations in soil are high, then the zinc atoms can absorb the energy fromemitting lead atoms resulting in the XRF detecting less lead energy than if the zinc concentrationswere lower; however, review of the data indicates that lead was not significantly masked whenthe x-ray intensities for all pure element standards used to construct the model were relative to the
8
600 H
Figure 3Low Lead Scatterplot
XRF vs. Laboratory Data
100 200 300 400XRF Assay [Pb] ppm
500 600
7000 -,
Figure 4High Lead Scatterplot
XRF vs. Laboratory Data
2000 3000 4000
XRF Assay [Pb] ppm5000 6000 7000
Figure 5Zinc Scatterplot
XRF vs. Laboratory Data
35000
5000 10000 15000 20000
XRF Assay \Zn] ppm
intensities in the sample which was analyzed.
Any samples analyzed hi the field which had a laboratory confirmation of over 4030 ppm wouldbe out of the range of the model calibration and tend to be underestimated.
5.3 ZINC MODEL
Figure 5, Zinc Scatterplot, indicates that the XRF had a tendency to overestimate zinc valueswhen the zinc concentration exceeded 2000 ppm. The reason for the overestimation in zinc value sis because the SSSs used to construct the model, did not span the expected range of zincconcentration found in the field. Any samples analyzed in the field which had a laboratoryconfirmation of over 26,000 ppm would be out of the range of the model calibration and tend tobe underestimated.
U.S. Environmental Protection Agency726 Minnesota AvenueKansas City, KS 66101
BVSPC Project 46515.257BVSPC File E.8
XTowgmVoM- 1 g IQQg.,
Subject: Comments for Revised Draft SREM
Attention: Mr. Don Bahnke, WAM
Gentlemen:
_
<LT
Black & Veatch Special Projects Corp. (BVSPC) has reviewed the September 9, 1998,submittal of the Revised Draft Site Removal Evaluation Memorandum (SREM) for theDiamond and Granby Subdistricts, Newton County Mine Tailings site prepared by Dames &Moore on behalf of AS ARCO, Inc. and Blue Tee Corp.
Overall, the comments submitted by USEPA Region VII on May 15, 1998, were sufficientlyaddressed by the Respondents. However, the following comments were not addressed in theSREM and have been slightly revised to clarify the requirements for the SREM:
General Comment3. As stated in Paragraph 11 of the Administrative Order on Consent, Docket No. VII-96-f-
0022, dated June 18, 1997, the EP A has established ajremoval action level of3^00(3' }micrograms per litei; (ug/L) for zinc. Any removal-action approved^yJERAr'will have-to"''meet this establishedNstendard.
Specific Comments9. Section-j.l. Page 12, Second Paragrap'h'rBast Sentence: Provide a brief-summary of the"- •,
information obtained on thejaeld worksheets (such as a summary-of how many residents/use water treatment unitscT'bottled water, well depth, etcT)~r~This information may be
L/ considered fofihe-erigineering evaluation.
28.
30..V
$w
Appendix A: Upon review of Figures-4=4 through 4-4 in-eomparison with the soils data inAppendix A, soil,.d-aTa-sfor GS-5201ias nonbeen inchided. cWfirmJwh[gttertlais--data^is''"''''ayailablexand/OT usable.NQisjcuss the deleti'Qn^pf'tnis data from the memorandum, ifnecessary.Appendix B. XRF OA/OC. Page 3: Provide tlieJCRF4ntensity_d.ata_asjpart of the QA/QCdiscussioH^oTThis. SREM. Indicate whether any modeling informationneeded"Tcrbe~
/' "x^^ ——^—modified to confirnrthe-resuitsT
Page 2
USEPARACVII BVSPC Project 46515.257Newton County Site November 18, 1998
r&f.v31. Appendix B, XRF QA/QC. Section 4.2. Page 5: Provide the relative percent differencesr t '' l'-\s. ^ (RPDs) for all.XRF vs. Laboratory data tq ensure that the^.35-percent c"fitefia~was met for\Jfi'' £//;/ I" the project. Verify the equation for calculating me.RPD. it appears that the "+" and'"-"-are
32. Appendix B. XRF OA/QC. Page 11. Section 5.2: This section indicates that elevated XRFzinc levels are known to mask XRFiead intensities. The following samples were noted tohave zinc levels greater than 24",000 ppm with no confirmation s'ample collected: GS-4-07
^5TGB^4rQFB?4res=^^Indicate the measures taken (e|g., model re-calibration or
sample re-analysis) to ensure the accjuracy of the lead concentrations at these samplelocations. Sieveral of these sample locations are near a tailings pile and other/residenceswith notable! hish lead levels in the soil. Consider that these locations may/be minirig-/ i *~ / * > \ /related lead Revels. In/addition, this paragraph now contradicts itself by first indicatingelevated zindlevels mask lead levels 'and then^indicating, for this sit-e^ne lead levels were
/not significantly-masked. Clarify how^we-snould interpret the findings of the High Lead/ Model and why the results of the XRF should be interpreted differently for this site.
/.The following comment is presented for additional consideration.
Appendix A, Soil Results: Upon review of the soils results, Diamond sample numbers DB-100, DF-100, and DS-100 were sampled on December 6, 1997, and samples JDS-1Q0G, DS-100P, and.DS-lOOS were samplecLon July 8,-1.998. Clarify vsfhetherthis is'thTsamelocation or re-assign a different-'s'ample number tcrone'oT the samples to avoid confusion if
_... these 'samples are teferencedlat a later date.\ •-' v - ^^
If you have any questions or need further information, please do not hesitate to call me at 458-6538.
Sincerely,Black & Veatch Special Projects Corp.
Robin D. Wankum, P.E.Site Manager
Comments on September 9, 1998 Revised Site Removal Evaluation Memorandum (SREM):
General Comments
Generally I believe the document is in a condition to be approved as far as meeting the goals forthe removal evaluation outlined in the Order. I have a few comments which are somewhatsemantical in nature. We may want to approve the document conditionally on some editorialrevisions.
1. Section 3.1, Rural Drinking Water Investigation, page 12, third full paragraph. Thisparagraph and the.,next one state that j_certain number of homes-from-Diamond and
2. Section 3.2, Residential Yard Soil Investigation, page 14, first full paragraph. Thisparagraph,seems to indicate that-all, delineation samplingjyas completed. This is not true.Same issuers in 1 above, whether We were able togef access for all of thej.ddttionalarea^c where delineation/sampling wasxneeded^rbelieve this qould beoompleted as part
( of Ae work under th^EE/CA. Also there is no indication in Hits-section if all licensedcare facilities-were sampled. I assume they were and if so we should state this fact.
3. Section 4.1, Rural Drinki-rrf'Water Results, page 1^,-first'paragraph. This discussion-identifies the zinc/lead'ratio whicji EPA does nof agree with. "First line of gage"! 6,.change "can be differentiated" to '%nay beusble to be differentiated".—Iff"tfie next sentencechange "Based^on this relationship," to "If this relationship were true,".
^Section 4.2 ResiHential Yard Soil Results. This paragraph indicates tihaLnet-all______/ dehneation^sampli^ig was dojier"S?ex;5nTnient2.-^B'"couldMoJhis^s part of EE/CA.
Additionally, this seGtion-sfiould specifically discuss the results at daycare facilities.,,,--•' "-x ^,____^
Ground water dat 'esutts, Appendix A^JMryls there nVpW-7^or'DW---3-7-Tesults? IT""xth'ere\were data/problems,xe,gJJnyal-id'(Iata, this should be indicated.
Soil data results, Appendix A. There are several results that appear to be duplicated. Forexample GF-166 is listed twice with the same results. This is true of GF-167, GF-168,GG-170, GF:i171,_GF1172JiFJ^^
sbe_others./{FYI, by my count there are 66 yards which exceed 800 ppm Pb either by~XRT~~or lab results, 65 in Granby, 1 in Diamond. There are 115 yards which exceed 500 ppmPb either by XRF or lab results, 112 in Granby, 3 in Diamond.)
^ . ^ — i i / ^ x i i ' . . t V T T ^ T ' McK^niflbr,, Owernor • Steal w.M,TAT Ri
OF NATURAL RESOURCESDIVISION Oi: ENVIRONMENTAL QUALITY
P.O. Box 176 Jefferson dry, MO 65102-0176
October 19,1998
Mr. Mark DoolanU.S. EPA, Region ViiSuperfund Division726 Minnesota AvenueKansas City, KS 66101
Dear Mr. Doolan:
I have reviewed the Revised Draft Site Removal Evaluation Memorandum for the Diamond andGranby Subdistricts, Newton County Mine Tailings site. ! have a few substantive comments thatmay impact the characterization of the site. However, my comments should not require aresubmitta! of this document. My specific comments are included below:
1. Composite soi! sampling described in section 3.2, pages 13 and 14, and in section 4.2 willnot detect isolated high !0ad concentration areas of the yard .--This wiif likely underestimate4he\number,pf homes tjrat exceed an acjierrf Sevelxaridjs-'inconsistf nt wjttiuaclions taken stthe Jasper County sifc^. Action ieVeisJfi Jasper Counfy were based' offbne discreetsample sxceeding-800 ppm or 2,500 ppm.
2. On page 14, it is assumed thatdelineationexeeeding SOQxppmJeadajB'also composite-sampies. However, this ts not explicitly
/ - f l ipK v ' 1 recommend further deiineation,sampling be included in the EE/CA, which identifies
.discreet areas in the yards that are over action levels. This sampling should be -conducted(at alkysrds that exceed 500 ppm and, any other additional sampling that is done.
3. In section 4.1, page 16, and in Table 4-1, the\data from Newton Countyfyeils does notsupport the use,of the Zn/Pb ratio andlZn concentration over 500 ppb criteria adopted fromJasper County, |n Jasper County, a vast majority of the wells/that exceeded the Pb or Cd•action level aiso [exceeded the Zn/Pb ratio of J1QQ and Zn concentrations fever 500 ppb. InNewton County, {this trend is reversed. The majority of wells exceeding the Pb and Cdactiori levels do not meet the PRP's removal action levels' based on theif proposed Znconcentration criteria. Their model, therefore! is not valid for Newton County and theyshould fund the removal action for all wells inijthe Diamond and Granby subdistricts thatexceed an action level. / i / I
0ne possible explanation for the lower Zn concentrations in groundwater compared to/Jasper County may be based on the time that-has elapsed since the areas Were mined.Exposed sphalerite in .mine voids may have long since oxidized, dissoSvep!,/and beenflushed downgradient; Gaiena has a lower soiubifity and the dissolution process may be
Mr. Mark DoolanOctober 19,1998Page Two
slower. Mining in Newton County may be old enough that the majority of Zn dissolutionfrom mine voids may already have occurred and concentrations are declining. Obviously,much more data would be required to test this theory or any theory the PRPs may offer.
^ff£ 4, jn figures 4-1 and 4-2h some of the action level exceedences/forwells in both DiamondY v/ ,. jv and Granby have-beea found at the-boundaries of the subdtstricts. Sampling should
V , f{/\(j* continue outside of thes-e districts/until thelimits of contamination are\determinedj3r—.,^__r> tr', / ,\// y /hydrologic/dtvides or boUFidafies are encounferedr'This sampling shoaW-be-incorporated"
intothVE"E/CA.
5. In figures 4-4 and 4-5rsevera! homes adjacent to known tailings impc .in tDlS'orclfiatpiles were not sampled. I recommend all houses withih-SOO-feet-ofThine waste besample^ as part of further action'.' this would be consistent with actions at the JasperCounty sfiteT
/v section 4<2\page 26, mosfofthe discussion .of'eaoh of the residejttial-yardsjhat exceed'action levels isVrelevantand should be deleted if a redraft of this^ocument is riqukf d.Most exceedences irufesplin where-ihe ERA"has conducted-a-temoval or remedial actiondemo/fstrated the same heterogeneity-in'concentrations as described on this page.