EPA Superfund Record of Decision: Murray Smelter EPA ID: UTD980951420 OU 00 Murray City UT


                                    EPA/ROD/R08-98/078
                                    1998
EPA Superfund
     Record of Decision:
     MURRAY SMELTER
     EPA ID: UTD980951420
     OUOO
     MURRAY CITY, UT
     04/01/1998

-------
EPA 541-R98-078

MURRAY SMELTER
PROPOSED NATIONAL PRIORITIES LIST SITE
MURRAY, UTAH

RECORD OF DECISION

CERCLIS ID UTD980951420

1. Site Name and Location

The Murray Smelter Site ("the Site") is located in the city of Murray, Utah, in Salt Lake County as
illustrated on Figure 1. The Site includes the former operational areas of the Murray Smelter and
adjacent Germania Smelter which are referred to as the "on-facility" area, as well as surrounding
residential and commercial areas where airborne emissions from the smelters impacted the environment or
where contamination in shallow ground water may be transported in the future. These surrounding areas are
referred to as the "off-facility" area.

The on-facility area is approximatelv 142 acres. Its boundaries are 5300 South Street to the south, State
Street to the east, Little Cottonwood Creek to the north, and the west set of Union Pacific railroad
tracks to the west. The off-facility area is approximately 30 acres to the west of the on-facility area,
approximately 106 acres south and southeast of the on-facility area, and a small area between 5200 South
Street and Little Cottonwood Creek to the east of the on-facility area. The west portion of the
off-facility area is bounded by Little Cottonwood Creek to the north, 300 West Street to the west, 5300
South Street to the south, and the on-facility boundary to the east. The south/southwest portion is
bounded by 5300 South Street to the north and Wilson Avenue to the south. The off-facility boundaries
were determined by EPA based on the results of air dispersion modeling performed in November, 1994. The
purpose of the modeling was to identify the area that potentially would have received the greatest amount
of deposition resulting from lead and arsenic emissions from the Murray Smelter during its operating
period.

For environmental sampling, risk assessment, and risk management purposes, the Site was divided into
smaller areas to represent realistic areas of human and ecological exposure. The 142 acre on-facility
area was divided into eleven "exposure units" (EUs) and the 136 acre off-facility area was divided into
eight "initial study zones" (ISZ's). The riparian area along Little Cottonwood Creek was delineated as
the ecological study area. The Site boundaries, EU's, and ISZ's are shown on Figure 2.

2. Operational History

The Germania Smelter was built in 1872 on the north west corner of the on-facility area adjacent to
Little Cottonwood Creek. The Germania Smelter processed lead and silver ores. Asarco bought the Germania
Smelter in 1899 and operated it until 1902. At the time, Asarco was also constructing the Murray Smelter
on property to the south and adjacent to the Germania Smelter. In 1902, operations at Germania stopped
and the Murray Smelter began operating and continued processing lead and silver ores until 1949. Smelting
operations produced a variety of by products including arsenic (as sulfates/oxides in flue dust or as
arsenic trioxide), matte (an iron sulfide matrix with high lead and copper content), arsenical speiss (an
iron-arsenic-sulfide matrix), and slag (a vitrified iron silicate).

The on-facility portion of the Site includes both the former Germania Smelter and Murray Smelter facility
areas. Minimal specific information is available on the smelter operations at the Germania facility.
After operations ceased, the area was regraded with Germania slag and, later, with slag from the Murray
Smelter. Subseguently, no significant historical features of the Germania Smelter remain and the
description of smelter operations provided below is based solely on descriptions of the Murray Smelter.

At the time of its construction, the Murray Smelter was reportedly the largest primary lead smelter in
the world. In addition to lead, several byproducts were also generated including gold, silver, copper,
antimony, bismuth, arsenic, and cadmium. The main byproducts by volume were slag, arsenic and cadmium.

Figure 3 is a layout of the Murray Smelter facilities. The Murray Smelter included an extensive rail
network, two stacks (330 feet and 455 feet high), eight blast furnaces, roasters, arsenic kitchens,
sinter plants, mills and power houses. The facility also included a baghouse for emissions control. Most
of the Murray Smelter facilities have been demolished, except for the smelter stacks, some building
foundations, and the original office/engine room building.

-------
A flow sheet for Murray Smelter operations for 1920 is shown in Figure 4. Although modifications occurred
during the period of operation, the fundamental processes remained the same. The raw material, lead ore,
was shipped from various locations and was classified either as sulfide ore or oxide ore. Oxide ore was
capable of being smelted directly, whereas sulfide ore reguired a preliminary, roasting step to reduce
the sulfur content. The primary manufacturing process was therefore characterized by two major
operations: (1) roasting operations to lower the sulfur content of sulfide ores and to produce sintered
material suitable for final smelting; and (2) smelting operations to produce lead bullion (shipped away
for final refining), matte (sent to the roasters to be treated again by oxidation of its sulfur), and
slag. The secondary manufacturing process was the re-processing of flue dust and baghouse dust to produce
arsenic trioxide.

2.1 Roasting Operations

Prior to 1920, roasting operations involved three furnace types: (1) four Wedge roasters, (2)Dwight-Lloyd
roasters: and (3) five Godfrey Roasters, operated in conjunction with twenty-seven Huntington and
Heberlein ("H&H") pots.

The Wedge roasters received charge consisting of sulfide ore, matte from the blast furnaces, lead
concentrates from various points, and silica. These furnaces produced roasted ore which was then loaded
into tram cars and conveyed to cooling bins where it was combined with low sulfur ores and charged to the
Dwight-Lloyd roasters. Air emissions from the Wedge furnaces passed directly into a dust chamber that ran
along the north side of the Wedge roaster building and connected the main roaster flue to the Cottrell
Plant. The Dweight-Lloyd roasters, or sintering machines, produced material which was transferred
directly into rail cars and sent to the roast bins where the blast furnace charge was made up. Air
emissions from the Dwight-Lloyd roasters were also sent to the Cottrell Plant.

The Cottrell Plant was an electrostatic precipitator. Precipitated materials fell or were shoveled
directly into rail cars. These materials were either returned to the roasters or sent to the briguetting
plant to be briguetted for charging to the blast furnace. Gases from the Cottrell Plant were sent to the
455-foot stack, which began operating in May 1918. During repairs or other activities on the baghouse,
the roaster flue and treatment process received blast furnace gases.

The Godfrey Roasters were used to process flue dust from the baghouse and Cottrell Plant. Flue dust was
roasted in the Godfrey Roasters and the resulting arsenic trioxide vapor was conveyed to the arsenic
kitchens where it was collected as relatively pure arsenic trioxide. Exit gases from the kitchens were
sent to the western portion of the baghouse and collected dust was recycled to the Godfrey Roasters.
Arsenic trioxide was stored in one of two concrete storage bins before transportation offsite for sale as
a product. In 1942, additions were made to the arsenic kitchens to increase their production capacity
(additional kitchens were added) and to provide additional storage (new storage bins for arsenic product
were installed) and conveyance capacity (a system to convey baghouse dust to the kitchens was installed).

2.2 Smelting Operations

Smelting was achieved by eight blast furnaces. The charge to the blast furnaces included oxide ore, flux
material, and roasting products. Air emissions were sent to an enlarged flue, along the west side of the
building. From this chamber, the gases passed to a rectangular brick flue, 18 feet wide by 17 feet high,
which led to the baghouse. Exit gases from the baghouse were usually sent to the 330 foot stack, although
gases from the baghouse or blast furnace were occasionally routed to the 455 foot stack. The baghouse,
installed in 1907, was constructed of brick 216 feet long and 90 feet wide, and contained approximately
4,000 woolen bags, each 30 feet in length and 18 inches in diameter. In 1920, the baghouse was divided
into four compartments, three of which were operated while the fourth was cleaned out. Dust from the
baghouse was either loaded into rail cars for transport to temporary storage areas near the thaw house
where it was kept prior to off-site transport or conveyed to the Godfrey Roasters and arsenic kitchens by
narrow gauge railway for production of high-grade arsenic trioxide. The material from the baghouse was
low-grade arsenic oxide, which contained lower amounts of arsenic than the arsenic kitchen product, with
arsenic present in oxide and sulfate forms. Prior to off-site shipment, arsenic kitchen product was
stored in a wooden arsenic storage bin to the south east of the thaw house.

-------
2.3 Materials Used/Generated by the Smelter pperaton

The contaminants of concern to human health at the Site are lead and arsenic 1. Based on the data
generated at the Site and information on historic smelter operations, elevated levels of arsenic and lead
at the Site can reasonably be attributed to the following materials:

• Lead Ore: No analytical data are available to describe the range of arsenic and lead
concentrations in ore matenials processed at the smelters. Lead contents for ore from Utah
were reported between 4.4 and 32 percent by weight. Ore mineralogy was variable, but may
have included: galena, pyrite, arsenopyrite, sphalerite, anglesite, cerussite, and lead
oxide (massicot).

• Blast Furnace Products/By-products. Four materials were typically generated during blast
furnace operation: metallic lead, speiss, matte, and slag. The materials would separate due
to their varing densities. Metallic lead was the primary product of the operation, and it is
not expected that any guantity is currently present at the Site.

• Matte/Speiss: In smelting of ores at the Murray Smelter, the amount of speiss
produced was too small to separate it from the matte. Matte/speiss generated in the
blast furnaces was comprised of metal sulfides, with iron being the dominant metal.
Analysis of speiss for various smelters in the western U.S. show lead contents
between 0.5 and 2 percent and arsenic contents between 31 and 32 percent. Analysis of
matte at the same smelters show lead contents between 8.5 and 18 percent and arsenic
contents below detection limits. Since speiss contents were probably small at Murray,
it is believed that any material present at the Site will contain higher levels of
lead than arsenic. Lead matte/speiss concentrate was stored out in the open in the
northern plant area.

• Slag: Slag is an amorphous, vitrified furnace product and the primary byproduct of
the smelting process. Air-guenched slag was the material generated in the highest
volume by the smelter process and significant guantities are still present at the
Site. Lead concentrations of 8,200 to 16,000 milligrams per kilogram (mg/kg) and
arsenic concentrations of less than 5 to 1,500 mg/kg have been measured in slag from
the Site (both Germania and Murray slag piles). Metals are not typically released
from slag under normal environmental conditions. A series of leaching tests was
performed on a sample of slag material collected from the Site. The details of the
leaching tests and the results are summarized in the final Feasibility Study. The
tests indicate that a minimal proportion of the metals present is released from slag
when precipitation and ground water are the leaching solutions. However, the release
of arsenic appears significantly enhanced at both extreme high and low pH.

• Flue Dust: Roasting and furnace operations had a tendency to volatilize arsenic. These gases
were collected and transported in flues to treatment units, the Cottrell Plant or the
baghouse. Exit gases from these units were sent to the stacks. Flue dust is present in areas
where operations were located (flues, the arsenic kitchens, the Cottrell Plant, and the
baghouse) and in areas where flue dust was managed (next to the thaw house). Similar
materials are also present at the ground surface over a wider area. This is due to
dispersion resulting from spillage during material handling, and stack emissions. Arsernic
levels in flue dust have been measured at 25,000 mg/kg.

• Arsenic Trioxide: Arsenic trioxide was produced primarily during the processing of flue
vapors from the Godfrey Roasters in the arsenic kitchens. The material was probably in a
relatively pure form, with arsenic primarily present in oxide forms and some sulfate
present. Pure arsenic trioxide has been measured at 760,000 mg/kg arsenic. Approximately
2000 cubic yards of arsenic trioxide have been found in the on-facility area of the Site.

• Stack Emissions: Exit gases from the baghouse and Cottrell Plants were routed to the stacks.
Stack emissions resulted in the deposition of lead and arsenic onto surface soils in the
off-facility area. These emissions occurred during the entire period of smelter operation.
Lead levels in off-facility soils impacted by stack emissions have been measured as high as
1800 mg/kg. Arsenic levels in these soils have been measured as high as 610 mg/kg.
1 As will be discussed in subseguent sections of this ROD, contaminants of concern to ecological
receptors within the ecological study area include other metals in addition to lead and arsenic.
However, the majority of the Site is sufficiently characterized by focusing on lead and arsenic.

-------
2.4 Smelter Demolition

Records indicate that as part of the shut down of the Murray Smelter, existing raw material feed stock
was processed and the resulting products and by-products were collected and sold. Due to this seguenced
shut down, the amount of residual raw materials, products, and by-products left at the Site is limited.
The exception is slag, the primary by-product of the smelting process, which was initially present over a
large area. The initial guantity has been significantly reduced by mining in the period since the smelter
shut down.

The majority of smelter structures were demolished in the period immediately after operations ceased in
1949. Based on environmental sampling and historical photographs, it appears that demolition of the main
smelter structures was conducted in an organized manner. Salvageable materials (e.g., metal from the
processing units and rail lines, and other process eguipment) were taken off-site, and building
structures were subseguently demolished with the brick and concrete debris typically spread in the
immediate area. Slag was then brought in from the slag pile area to cover the debris and to provide a
suitable surface for subseguent development of commercial/manufacturing operations. Today, smelter
materials are typically present within the upper three feet below the current ground surface, primarily
in the form of slag brought in for fill, residual materials such as flue dust within footprints of former
operations and mixed structural debris from smelter demolition in the immediate vicinity of former
structure locations. At a limited number of locations, relatively high levels of arsenic such as that
associated with flue dust are present as deep as 10 feet. This is thought to be the result of dissolution
and transport by surface water infiltration.

Several smelter structures remained after the initial demolition activities. Some of the structures were
used as storage buildings until around 1980 when they were demolished as part of Site development. A few
structures, including the engine house and the stacks, are still present today.

3. Site Description

3.1 Land Use

3.1.1 Current Land Use

The on-facility area is currently zoned Manufacturing General Conditional, M-G-C. This zoning designation
allows light industrial processes to be conducted with heavier industrial uses allowed after a
conditional use permit has been approved by Murray City. The majority of the on-facility area is owned by
the Buehner family and leased by a concrete manufacturing company; the unrelated Buehner Corporation. The
company makes pre-cast and pre-stressed concrete building and transportation products as well as
architectural concrete products. Other uses within the on-facuity area include a pipe warehouse and
distribution facility, the W.R. White Company, a telecommunications eguipment company, Skaggs
Telecommunication Services; a Federal Express outlet; the Murray City Police Training Facility; a
Portland cement transfer and supply facility, Ashgrove Cement; other warehouses; and an abandoned asphalt
plant owned by Monroe, Inc. There are two residential trailer parks within the on-facility area. The "Doc
and Dell's" trailer park is located on State Street. The "Grandview" trailer park is on the southwest
corner of the on-facility area on 5300 South Street. The locations of these trailer parks are noted on
Figure 2.

Land use in the off-facility area is mixed residential/commercial. The western portion of the
off-facility area is currently zoned M-G-C and Commercial Development Conditional, C-D-C. C-D-C Zoning
provides areas where a combination of businesses, commercial, entertainment and related activities may be
established and maintained. The southern portion of the off-facility area is currently zoned M-G-C and
low density single family residential, R-l-8. The Murray Junior High School and the Murray High School
are located in the south portion of the off-facility area.

3.1.2 Future Land Use

In 1997, the Murray City council adopted a land use plan for future development of the on-facility
portion of the Site and amended its General Plan accordingly. The land use plan for the on-facility area
includes construction of a north-south roadway corridor from Vine Street to 5300 South Street through the
central portion of the on-facility area. Murray City council has appropriated the funding for the road,
which extends north and south of the Site along the alignment shown in Figure 5. This alignment takes
into account the City's desire for traffic volume and the owners' desire for sufficient access. Largely
due to the construction of this access road, a significant portion of the on-facility area is highly
likely to be redeveloped in the near future. Current land owners are discussing options with the City and
potential developers to optimize future use of the area. Much or all of the outdoor industrial activity
is expected to end, to be replaced with light industrial/ commercial activities. The City will rezone the

-------
area to C-D-C use by passing an ordinance establishing an "overlay district" which restricts certain uses
and reguires city review of development plans within the on-facility area boundaries.

Also, all residential occupation within the on-facility area will soon end. A Site developer has acguired
an option to purchase the Doc and Dell's trailer park with the intention of converting the trailer park
to commercial uses. Grandview Trailer Park has been purchased by the Utah Transit Authority (UTA) and
residential leases are not being renewed. UTA intends to swap the Grandview parcel for a parcel of land
owned by the Buehner family near Ashgrove. Within two years, UTA will construct a light rail station
platform adjacent to the existing railroad tracks along with associated off-street parking, and
landscaping. If the land swap with Buehner occurs, then residential occupation of Grandview will be
terminated more rapidly as the site is developed. In either case, residential occupation of Grandview
will likely end within two years.

The Amendment to the General Plan for Murray City also includes three other potential public use
projects:

1) Murray City Court/Police Administrative Office. There is interest in locating a court/police complex
somewhere south of Little Cottonwood Creek, and south of Vine Street. The City will be establishing
its own court system within a few years and will ultimately need facilities to be constructed for
this purpose. There is an urgent need to provide adeguate police facilities as well as additional
space in City Hall. It is anticipated that three to five acres will be needed for this facility.

2) Little Cottonwood Creek Parkway Improvements. The Murray Parks & Recreation Department is interested
in obtaining property to enhance the south side of Little Cottonwood Creek with landscaping, a
walking and bicycle trail, urban plaza, pavilion and restroom facilities contained within
approximately 5 acres. This would allow the extension of the City's existing trail system with a
target of connecting to the Jordan River trail system.

3) Smelter Site Interpretive Park. There is also interest in developing a small interpretive park at the
base of the smelter stacks that would be no larger than approximately two acres. The small park could
contain a plaza, seating, fountain and landscaped areas. Historical information relating to the
smelter Site history would be integrated into the park development.

This type of development provides the opportunity to integrate implementation of remedial actions into
development activities, a key objective of EPA's Brownfields Program. Given the interest in developing
the on-facility area and the high level of involvement and commitment by the City of Murray and the
current land owners, there is sufficient certainty concerning, future land use to identify the reasonably
anticipated future land use scenario as recommended in the EPA OSWER directive "Land Use in the CERCLA
Remedy Selection Process". The reasonably anticipated future land use for the on-facility area is light
industrial/commercial use.

In the off-facility area, areas to the west of the on-facility area (ISZ-1 and ISZ-8) are zoned M-G-C and
C-D-C but do have some residential occupation. This zoning prevents the construction of new homes, and
therefore, residential occupation is expected to end in the future. To the south of the on-facility are
(ISZ-6 and ISZ-7) a portion of the land is zoned for residential use and a portion is zoned M-G-C.
Similar to the western off-facility area, although there are some existing non-conforming residences,
residential occupation is expected to end sometime in the future due to the prohibition of new home
construction and the redevelopment of the on-facility area. The reasonably anticipated future land use
for the off-facility area is a combination of commercial/light industrial and residential.

3.2 Topography

The Site is mainly flat in the southern portions. Near Little Cottonwood Creek on the
north, the terrain slopes steeply. This area used to be filled with slag from the Murray Smelter but
over the years since the smelter shut down, the slag has been excavated and used throughout the
Salt Lake Valley. A steep wall of concrete debris from recent Site uses and residual slag remains
in the northern area where slag used to exist.

3.3 Geologic Units and Soils

The geologic units at the Site consist primarily of lake sediments from Pleistocene Lake Bonneville,
however, younger alluvial floodplain deposits are found along Little Cottonwood Creek. The lake sediments
consist of clays, silts, and fine sands and underlie the more recent alluvial stream deposits which
generally consist of silt, sand, and gravel. Surface soils within the on-facility portion of the Site
have been disturbed, affected by the construction and operation of smelting, ore handling, and refining
facilities over a period of 77 years. In more recent times, construction and operation of concrete,

-------
asphalt, and other commercial or manufacturing facilities have further disturbed the area's soils. In
particular, construction of the facilities and the deposition of slag from smelting operations and other
fill materials have covered the majority of the original surface soils.

In the off-facility area, surface soils have been significantly affected by extensive general urban
development.

3.4 Hydrogeology

The Site lies on an area covered by thick valley-fill (alluvial) deposits that comprise several distinct
aguifers within the aguifer system. Specific components of the aguifer system are as follows:

• Shallow Aguifer: a shallow, unconfined aguifer comprised of interbedded sandy clays and
clayey sands occurring above the Bonneville Blue Clay;

• Bonneville Blue Clay: approximately 30-foot-thick continuous layer of clay separating the
shallow and intermediate aguifers;

• Intermediate Aguifer: a confined aguifer immediately underlying the Bonneville Blue Clay
comprising approximately 10 to 20 feet of relatively coarse-grained deposits; and

• Deep Aguifer: an artesian aguifer, several hundred feet below the intermediate aguifer,
comprising various coarse-grained valley-fill deposits

The shallow aguifer is unconfined with a saturated thickness that ranges from 2.5 to 25 feet within the
on-facility area. The average depth to water is approximately 10 feet. The aguifer materials have a
geometric mean hydraulic conductivity of 5 feet per day (based on estimates from different locations in
the study area ranging from 1 to 112 feet/day) . Groundwater in the shallow aguifer flows along the top of
the Bonneville Blue Clay, generally north-northeast, toward Little Cottonwood Creek as shown in Figure 6.
Water levels measured adjacent to the creek indicate that the shallow aguifer is hydraulically connected
to Little Cottonwood Creek and that groundwater discharge to the creek occurs during certain times of the
year.

The second component of the aguifer system is the Bonneville Blue Clay. Available hydrogeologic
information indicates that the Bonneville Blue Clay is continuous across the facility and the surrounding
area. This lithologic unit forms an effective barrier for vertical groundwater movement from the shallow
aguifer to the intermediate and deep aguifers. Analyses presented in the Feasibility Study support this
conclusion.

Beneath the Bonneville Blue Clay, the intermediate and deep aguifers are separated by more than 200 feet
of interbedded fine- and coarse-grained valley-fill and alluvial deposits. Both receive recharge
primarily up gradient of the Site. Groundwater in the intermediate aguifer flows north-northwest across
the Site as shown in Figure 7, and the aguifer is not hydraulically connected to surface water bodies in
the vicinity of the Site. The deep aguifer is the main source of drinking water for most residents in the
Salt Lake Valley. Municipal water-supply wells located in the vicinity of the Site are screened more than
500 feet below the ground surface in the deep aguifer.

3.4.1 Potential for Use of Ground Water as a Drinking Water Supply

It is unlikely that the shallow aguifer will ever be used as a potable water supply due to several
conditions. Primarily, the water is of poor guality for drinking water. Background total dissolved solids
(TDS) concentrations range from 606 to 3,236 mg/L and exceed EPA's secondary drinking water guality
standard of 500 mg/L. Additionally, this water supply is only available in limited guantity due to the
aguifer thickness coupled with low hydraulic conductivities which do not produce sufficient water for
typical water supply needs. The intermediate and deep aguifers provide lower TDS and higher yielding
water supplies. However, within EPA's ground water classification system, two factors are considered in
designating ground water as a potential drinking water source; water guality and yield. In EPA's
regulatory scheme, water is considered to be suitable for drinking if it has a TDS concentration of less
than 10,000 mg/L and either can be used without first being treated or can be rendered drinkable after
being treated by methods reasonably employed in a public water supply system and can sustain a yield of
150 gallons per day. The characteristics of both the shallow aguifer and the intermediate aguifer at the
Site meet EPA's criteria for designation as a potential drinking water source, Class lib and Utah's
criteria for designation as a Class II drinking water under Utah's Ground Water Quality Protection Rule.
The deep aguifer meets both EPA's and Utah's criteria for designation as a Class I aguifer, a current
drinking water source.

-------
3.5 Surface Water

Little Cottonwood Creek is a perennial stream flowing along the north/northeast boundary of the
on-facility area and into the Jordan River approximately one mile downstream. The stream has been altered
by urban and agricultural development both upstream and downstream of the Site. In the northern portion
of the on-facility area, the course of the stream was altered during smelter operation. Facility drawings
and aerial photographs indicate that the creek originally flowed through the northern portion of the
on-facility area, but during smelter operation the creek was diverted to the north with the former
channel incorporated into the slag pile. Today, the upstream reaches of the creek are bordered by
residential areas or parks, while the Site and downstream reaches are mainly bordered by
commercial/industrial areas.

Historically, Little Cottonwood Creek has been stocked with rainbow trout and German brown trout;
however, reproductive success of these fish is thought to be poor due to the steep gradient and a below
average availability of good guality pools in the creek. In the vicinity of the Site, Little Cottonwood
Creek is designated by the State of Utah for secondary contact recreation use such as boating and wading
(classification 2B), for cold water game fish use (classification 3A) and agricultural use
(classification 4). A survey of Little Cottonwood Creek conducted in 1997 found no diversions of surface
water for agricultural use downgradient of the Site. Although no formal petition has been brought forward
to the Utah Water Quality Board to change the agricultural use designation, existing evidence documented
in the survey report suggests that such use is not likely in the future.

In addition to the use designations assigned by the State of Utah, fisheries habitat in Utah is
inventoried and classified on a statewide basis by the Utah Division of Wildlife Resources. The section
of stream near the Murray Smelter has been designated as a Class 5 stream based on esthetics,
availability, and productivity as determined in a physical habitat survey conducted in 1974. According to
the classification system, Class 5 streams are now practically valueless to the fishery resource, however
many waters in this class could provide valuable fisheries if sufficient guantity of water could be
provided.

On the northern area of the Site, shallow ground water within the floodplain of Little Cottonwood Creek
surfaces at three distinct locations to form wetlands. An area of 0.75 acres of wetlands were identified
in a delineation study done in June, 1997 by Hydrometrics titled "Report of Wetland Determination, Little
Cottonwood Creek Riparian Area, Former Murray Smelter Site, Murray, Utah".

3.6 Climate

The Salt Lake area has a semi-arid climate. Average precipitation is approximately 16 inches per year and
the average air temperature is approximately 64 degrees Fahrenheit. The Site elevation is approximately
4280 - 4315 feet above sea level.

3.7 Floodplain

The most recent flood insurance study which includes Little Cottonwood Creek was done by HUD in 1994.
Several differences have been observed between existing floodplain topography and the floodplain cross
section data utilized for development of the most recent floodplain map. Existing conditions compared
with conditions from which previous floodplain delineations were based, show more floodplain area in the
southbank (within the on-facility area) and less flood plain was in the northbank (north of the Site
boundary). The large existing southbank floodplain area probably resulted from excavation of slag from
this area, or it may have been excluded from previous studies because it may not be part of the effective
flow conveyance. Most of the site is outside of the 100 year floodplain as shown on Figure 8 from the HUD
study.

3.8 Nearby Populations and Demographics

Based on data from the 1990 census, approximately 20,000 people live within a mile radius of the Site.
The majority of this population is non-minority. Of the 20,000, there are approximately 2,100 children 5
years old or younger, 2,700 adults over the age of 60, and 4,200 women of child-bearing age (18-45 years
old). Figure 9 summarizes this demographic information.

4 Site History and Enforcement Activities

4.1 Administrative Order on Consent for an Engineering Evaluation/Cost Analysis

In January, 1994, EPA proposed that the Site be listed on the National Priorities List. On August 5,
1994, EPA issued a letter of "Notice of Potential Liability and Demand for Payment to Asarco.

-------
Negotiations between EPA and Asarco commenced shortly thereafter culminating in September, 1995 when EPA,
Asarco, and Murray City entered into an Administrative Order on Consent (AOC) for the performance of an
Engineering Evaluation/Cost Analysis (EE/CA) for the Site. EPA retained responsibility for performing a
baseline human health and ecological risk assessment for the Site. The EE/CA was intended to support a
Non-Time-Critical removal action.

4.2 AOC for Time Critical Removal

On September 13, 1995, EPA and Asarco entered into a separate AOC for conducting a time critical removal
at the playground area of the Grandview Trailer Park. The scope of this time critical removal was
excavation of soils within and adjacent to the playground which contained unacceptable levels of lead and
arsenic and backfill of those areas with clean fill. This removal action was completed by Asarco in
November, 1995. The removed soils have been temporarily stored in a waste pile on-Site and will be
consolidated on-Site as part of the remedial action selected in this ROD.

4.3 Memorandum of Understanding with Murray City

In April, 1996, EPA and Murray City entered into a Memorandum of Understanding which established that
Murray City would assist EPA in identifying current and potential future land use at the Site, in
developing response action alternatives, and in implementing any institutional controls reguired by EPA's
chosen response action.

4.4 EE/CA

Data needs were identified in the EE/CA Work plan, an attachment to the EE/CA AOC. Environmental sampling
to support the EE/CA and risk assessments began in April 1995. Asarco completed a Site Characterization
Report in August, 1996. Shortly thereafter, EPA decided to redirect what had been a Non-Time-Critical
Removal activity into the remedial action framework. Accordingly, the reguirement for an EE/CA was
changed to a Feasibility Study. Table 1 shows the completion dates for the major documents which support
this Record of Decision (ROD).

Table 1: Completion Dates for Major Documents Supporting the ROD

DOCUMENT RESPONSIBILITY COMPLETION DATE

Site Characterization Report Asarco August, 1996
Baseline Human Health Risk EPA May, 1997
Assessment
Feasibility Study Report Asarco August, 1997
Baseline Ecological Risk EPA September, 1997
Assessment
Proposed Plan EPA September, 1997

4.5 Information Requests

EPA sent CERCLA, 104(e) reguests to Asarco and on-facility property owners by letter dated April 25, 1996
seeking information on operations at the Site and material handling and storage details. Responses to the
information reguests were provided by all recipients.

5. Scope of Response Action

The remedial action which is the subject of this ROD is the second of the three response actions EPA
considers to be necessary at the Site. The first response action was a time critical removal of soils
located in and adjacent to the playground area at the Grandview Trailer Park. These soils were
contaminated with lead and arsenic at levels considered by EPA to be unacceptably high. The area was
backfilled with clean fill. The decision to undertake the time critical removal action is documented in
an Action Memorandum signed by EPA Region 8 on November 7, 1995. Asarco completed the Grandview Trailer
Park time critical removal action in November, 1995.

The remedial action described in this ROD addresses contaminated ground water, the subsurface soil which
is the source of the ground water contamination, contaminated surface soils, and the surface water of
Little Cottonwood Creek as follows:

-------
1. Contaminated ground water. Source control will be implemented by excavation and off-site disposal of
the principal threat wastes at the Site, approximately 2000 cubic yards of residual undiluted arsenic
trioxide. This material is considered a principal threat due to its high mobility and its demonstrated
ability to act as a source of Ground water contamination. In addition, direct contact with this
material may result in acute human health risks. Further source control will be implemented by
excavation of approximately 68,000 cubic yards of low level threat waste, diluted arsenic trioxide or
flue dust mixed with soil, fill, or debris from former smelter structures. This material will be
consolidated within a repository system constructed within the Site boundaries. The repository will be
designed as the base for a new access road through the Site which was planned by Murray City. The
access road is expected to be the catalyst for Site development. Monitored natural attenuation will
address the residual ground water contamination within and down gradient of these source areas.
Institutional controls in the form of a Murray City ordinance establishing an "overlay district" and
restrictive easements that run with the land both will prohibit the construction of new wells or use
of existing wells (except EPA approved monitoring wells) within the on-facility area and the western
and eastern portions of the off-facility area.

2. Contaminated surface soils. On-facility surface soil containing levels of lead and arsenic which
exceed remediation levels will be covered. The barriers will provide protection by breaking the
exposure pathways associated with long term direct contact with these soils. Site development itself
is expected to result in additional protection of human health since land uses associated with
unacceptable human health risks will end. Also, the development will result in the construction of
additional barriers (new buildings, roads, sidewalks parking lots, and landscaping) over remaining
surface soil and slag. Although no unacceptable risks associated with exposure to slag were identified
by EPA, the development of the Site will ensure no exposure to slag in the future. Institutional
controls in the from of a Murray City ordinance will establish an "overlay district" which includes
zoning to prevent residential and contact intensive industrial uses within the former smelter
operational areas and will reguire maintenance of the barriers and controls or excavated subsurface
material within this same area. Restrictive easements that run with the land will be established in
addition to the overlay district to prevent residential or contact intensive industrial uses.

Off-facility surface soils containing levels of lead exceeding remediation levels will be removed and
replaced with clean fill. The removed soil will be used on-facility as subgrade material in
construction of the repository system.

3. Surface water. Little Cottonwood Creek, which forms the northern boundary of the Site and to which
shallow ground water discharges will be monitored to ensure continued protection during the ground
water natural attenuation process. Additional monitoring of the ecological study area of the Site will
be used to reduce the uncertainties identified in EPA's predictions of ecological risk.

The remedial action protects ground water and Little Cottonwood Creek and incorporates the construction
of a new north-south access road through the site which will encourage future development of the Site and
help achieve Murray City's goal of more appropriate land use through site development. Institutional
controls will prevent exposure of people to ground water with arsenic concentrations that represent an
unacceptable risk and will also ensure that future uses of the land will be protective and that the
remediation will be maintained.

EPA expects that an additional response action will be reguired at the Site. A structural analysis of the
existing stacks at the Site was completed in January, 1998. The study concludes that both stacks as they
exist today are not able to withstand seismic events which are specified in the current Uniform Building
Code. Based on information collected as part of Site characterization efforts on the nature and extent of
contamination on interior bricks of the stacks, EPA expects that an additional time critical removal
action will be reguired to address the potential for release of hazardous substances and resulting health
risks associated with the potential structural failure of the stacks.

6. Highlights of Community Participation

EPA's community involvement activities at the Site began in March, 1995 with the establishment of the
information repository at the Murray City Library. In August, 1995, when the EE/CA work plan was in final
preparation, EPA and UDEQ released a fact sheet describing the scope and objectives of the site
investigation. With the assistance of Murray City officials, two public meetings were conducted on August
9, 1995 and August 10, 1995 to inform the affected citizens of Murray about the up-coming investigation
activities on or near their property.

In September, 1996, EPA released another fact sheet describing the preliminary results of the baseline
human health and ecological risk assessments. Since the results were specific to separate populations,
EPA conducted six separate public meetings and two availability sessions to explain the results of

-------
environmental sampling and risk assessments.

In October, 1996, EPA initiated the formation of the Murray Smelter Working Group consisting of
representatives of UDEQ, Asarco, owners of property and businesses on the Site, Murray City, and EPA. The
purpose of the Working Group was to inform EPA about pending Site development plans and to provide a
forum for discussing alternative cleanup strategies for the on-facility area of the Site. In a series of
open meetings conducted during October, 1996 through February, 1997, implications of remedial
alternatives were discussed by the working group. EPA provided information on the nature and extent of
contamination and the clean up reguirements.

The following commitments were made as a result of the Working Group sessions:

1. Current property owners, Murray City, and Asarco are committed to accomplishing the necessary tasks to
ensure that a new road will be constructed on the Site between Vine Street and 5300 South Street.
These tasks include dedication of the land for the road right of way and agreement on the
establishment of a "Special Improvement District" to fund utility construction.

2. Current property owners and Murray City are willing to work together to establish appropriate public
and private institutional controls as reguired by EPA's selected remedy.

3. Asarco is willing to use its best efforts to design a remedial action that is consistent with the
Murray City General Land Use Plan.

The agreements among the members of the Murray Smelter Working Group are memorialized in an Agreement in
Principle signed in May, 1997.

In September, 1997, EPA released the Proposed Plan for the Site and made available all supporting
documents in the information repository established at the Murray City Library and the EPA Superfund
Records Center at the EPA Region 8 offices in Denver, Colorado. The notice of availability of these
documents was published in the Salt Lake City Tribune and the Deseret News on September 23, 1997. A
public comment period was held from September 22, 1997 until October 22, 1997. In addition, a public
meeting was held on October 1, 1997. Responses to the comments received during the public comment period
are included in the Responsiveness Summary which is part of this ROD. A summary of the highlights of
community participation is presented in Table 2.

This decision document presents the selected remedial action for the Murray Smelter Site in Murray, Utah,
chosen in accordance with CERCLA and the National Contingency Plan. The decision for this Site is based
or the administrative record.

Table 2: Highlights of Community Participation Activities

ACTIVITY SUBJECT DATE

Fact Sheet summary of site investigation August, 1995
activities

Public Meeting explanation of sampling August 9-10, 1995
activities

Fact Sheet draft risk assessment release September, 1996

Public Meetings/Availability draft risk assessment and September, 1996
Sessions sampling results

Murray Smelter Working future site use plans and October, 1996 - February,
Group Sessions remediation alternatives 1997

Fact Sheet Proposed Plan of Action September, 1997

Public Meeting comments on the Proposed October, 1997
Plan

Public Comment Period Proposed Plan of Action September 22 - October 22,
1997

-------
7 Summary of Site Characteristics

7.1 Scope of Site Investigation Activities

Using data available from Preliminary Assessment/Site Investigation activities. EPA performed screening
level calculations to identify the chemicals of concern which would be the focus of site
characterization, risk assessment, and remedial activities at the Site. This analysis is documented in
the "Preliminary Scoping Report" prepared by EPA in December, 1994. The analysis concludes that lead and
arsenic are the chemicals likely to be of substantial concern to humans. Based on these results, the
EE/CA Work Plan specified lead and arsenic chemical analysis of soil and ground water samples collected
to support site characterization and the baseline human health risk assessment. Recognizing that
chemicals of concern to ecological receptors, especially aguatic organisms, often are different from
those of concern to humans, EPA selected the ecological chemicals of concern by evaluating historical
data collected from surface water, sediment, and soil in the Little Cottonwood Creek riparian zone. This
evaluation was done by the EPA Region 9 Ecological Technical Assistance Group (ETAG) at a meeting on
January 31, 1995. In addition to arsenic and lead, the ETAG identified aluminum, cadmium, copper,
mercury, nickel, selenium, silver, thallium, and zinc as ecological chemicals of concern to be
investigated in the ecological study area.

7.2 Soil and Dust Investigation

The site investigation for surface soil, subsurface soil, and dust is detailed in the Final EE/CA Work
Plan completed in September, 1995. Prior to sampling, the on-facility area was divided into eleven ISZ
based on current property boundaries and land use. Similarly, the off-facility area was divided into
eight EUs based on consideration of the predicted pattern of historic air deposition from the site along
with current street and land use features. A total of 10-20 surface soil samples (defined as 0"-2" deep)
were collected from each on-facility EU. More samples were collected from the larger exposure units. In
addition, test pits were excavated in several exposure units, using existing and historical features to
select the location of the pits. Special emphasis was placed on areas where potential sources of
contamination such as historical locations of the smelter flues, the bag house, waste transfer
facilities, the roasting areas, the arsenic kitchen, and the smelting areas were located. At each test
pit, subsurface samples were collected in one foot intervals to a depth of 5 feet.

In the off-facility area, surface soil samples were collected from 10 to 16 distinct residential yards
(depending on the size of the ISZ) within each ISZ. Each sample was a composite of surface soil from 4 to
6 sub-locations within the yard. In addition, 16 soil borings were collected (two different locations in
each ISZ) and subsurface soil samples were collected from the 0"-2", 2"-6", 6"-12" and 12"-18" intervals.
These subsurface samples were collected to characterize the vertical extent of contamination in each
off-facility ISZ. Indoor dust samples were collected from 22 different homes or buildings in the
off-facility areas. Samples were collected using a hand held vacuum. Typically, each sample was a
composite of dust collected from three areas, each about 2 feet by 7 feet. Summaries of sampling results
for soil and dust can be found in Tables 3-5.

After the Baseline Human Health Risk Assessment was completed, supplemental soil sampling was conducted
in every residential yard within those ISZs which were predicted to have unacceptable risk. A summary of
this supplemental sampling effort can be found on Figures 10-12.

Table 3: Summary Statistics for Indoor Dust Samples

CHEMICAL # OF SAMPLES AVERAGE RANGE

Arsenic 22 27 mg/kg 5 mg/kg - 94 mg/kg

Lead 21 303 mg/kg 83 mg/kg - 757 mg/kg
In order to gain information on the physical and chemical nature of the lead and arsenic present in
surface soil, EPA collected 10 samples from locations on the Site. These samples were dried and sieved to
yield the fine fraction (<250 um) and submitted for geochemical characterization. The lead in soil at the
Site occurs in a variety of different forms, most commonly as lead phosphates, lead silicates, lead
oxides, iron-lead oxides, lead arsenic oxide, and lead sulfide. In contrast, arsenic occurs mainly as
ferric-lead-arsenic oxide and lead-arsenic oxide with only small amounts of other arsenic species. The
lead and arsenic bearing particles were mainly smaller than 20 um with about 80% of all the lead or
arsenic bearing grains existing in a liberated or cemented state, with only about 20% existing within a

-------
rock or glass matrix.

7.3 Slag Investigation

EPA collected a single composite sample of slag from nine different locations at the Site. Two of the
subsamples were from the Germania smelter slag pile, six were from the face of the slag monolith located
adjacent to EU-2, and one was from the slag at the base of the slag pile adjacent to Doc and Dell's
trailer park. The composite slag sample was analyzed in duplicate using Contract Laboratory Program
methods. The mean values of the duplicate analyses are 695 mg/kg arsenic and 11,500 mg/kg lead.

In addition to chemical analysis, the slag sample was submitted for geochemical characterization. As
expected, the principle form of lead-bearing particle in the slag sample is slag (i.e., particles of
glassy matrix with lead dissolved in the glassy phase). However, this type of particle contains a
relatively low concentration of lead and so does not account for most of the lead mass in the sample.
Rather, the majority of the relative lead mass exists in the form of lead oxide with smaller
contributions from galena (9%), lead arsenic oxide (6%) and other metal lead oxides (4%). About 87% of
all lead bearing particles in the slag sample are liberated, accounting for about 77% of the relative
lead mass.

Similarly, the most freguent type of arsenic bearing particle in the slag sample is slag, accounting for
62% of the relative arsenic mass. The majority of these particles are liberated, existing partially or
entirely outside the confines of glassy slag particles.

7.4 Ground Water Investigation

The ground water investigation was conducted in two phases which included installation of 13 monitoring
wells in the shallow aguifer, 7 monitoring wells in the intermediate aguifer (Phase I), and a hydropunch
investigation (Phase II). Several other on-facility wells that had been installed in earlier
investigations were also redeveloped and sampled. A presentation of the results of all the ground water
sampling performed between October, 1995 and April, 1996 is contained in the final Site Characterization
Report. Shallow alluvial and intermediate around water continues to be monitored guarterly. Summaries of
the sampling results for key analytes in shallow ground water can be found in Table 6. A full summary of
a ground water sampling results can be found in the October, 1997 Ground Water and Surface Water
Monitoring Report. The most severe Site-related impact to shallow ground water was found to be arsenic
contamination. Figure 6 illustrates the arsenic levels detected in shallow ground water in January, 1996.
District plumes of contamination can be seen in areas underlying the former locations of smelter
operations.
7.5 Surface Water, Sediment, and Riparian Soil Investigation
Samples of surface water, sediment, benthic macroinvertebrates, and riparian soil were collected in the
ecological study area and analyzed for ecological chemicals of concern as part of site characterization
efforts. Figure 13 shows the locations of these samples. Summaries of the results of this sampling can be
found in Tables 7-10.

Subseguent to site characterization efforts, additional guarterly surface water sampling was conducted
beginning in July, 1996. Additional locations were established to characterize areas of Little Cottonwood
Creek which receive ground water discharge from the shallow aguifer and to characterize the effects of
ground water and point source discharges on the water guality of Little Cottonwood Creek. Figure 13a
shows these additional locations. This supplemental sampling was limited to arsenic analysis. Summaries
of the surface water results can be found in Table 11.

The results of the point source discharge sampling are particularly significant because they indicate
that the increase in dissolved arsenic concentrations in Little Cottonwood Creek occurs in the vicinity
of the discharge from a storm sewer culvert running north along State Street. Loading calculations
presented in the April, 1997 guarterly monitoring report demonstrate that nearly all of the dissolved
arsenic loading (88%-100%; accounting for flow measurement accuracy) observed in the creek appears to
originate from the culvert point source discharge. Ground water discharge from the shallow aguifer in the
on-facility area to the south of the creek was not shown to have a measurable effect on arsenic load in
the creek.

-------
8. Summary of Site Risks and Remedial Action Objectives

8.1 Human Health Risks

EPA completed a baseline risk assessment for the Site in May, 1997. Human health risks were calculated
separately for four groups of people to characterize risks for the current and reasonably anticipated
future land use; on- and off-facility residents; on-facility workers who spend most of the day indoors
(non contact intensive (NCI) workers); on-facility workers who spend most the day outdoors and are
engaged in activities that result in significant exposure to soil and dust (contact intensive (CI)
workers) ; and teenagers who have been observed congregating in areas along Little Cottonwood Creek. The
exposure pathways evaluated for each group were ingestion of soil and dust, ingestion of slag (only
evaluated for current and future teenagers) , and ingestion of ground water. Other exposure pathways to
site-related wastes are judged to be sufficiently minor that guantitative evaluation was not warranted.
The current land use for the site is a combination of commercial (best represented by NCI Workers),
industrial (best represented by CI workers), and residential. As discussed in Section 3, the reasonably
anticipated future land use for the on-facility area is commercial/light industrial (NCI) and for the
off-facility area is a combination of commercial/light industrial (NCI) and residential. The exposure
assumptions used in the risk assessment were also used to develop preliminary remediation goals for soil.
These assumptions can be found in Appendix B.

The risk assessment, was performed using two distinct approaches for the on-facility and off-facility
portions of the Site. The majority of the on-facility was divided into seven EUs, sized to approximate
the area over which a typical office or industrial worker would come into contact with surface soils
during a working lifetime. The residential trailer parks within the on-facility area were divided into
four smaller EUs sized to approximate the area over which a child or adult might come in contact with
soil during the period of residence. Soil samples were collected within each exposure unit and averaged
according to EPA guidance. This average, the "exposure point concentration", was the basis for the risk
calculation. EPA will manage risks for the on-facility area by EU.

In contrast, the off-facility was divided into eight ISZs sized to represent neighborhoods, not
individual residences. This was because historical data indicated little variability in concentrations of
lead and arsenic within neighborhoods. Concentrations in general tended to decrease with distance from
the smelter site. The term ISZ was chosen deliberately to reflect that the risk assessment for the
off-facility area is an "initial" or screening level assessment. The exposure point concentrations for
the off-facility risk assessment were the average concentrations for each ISZ or neighborhood. EPA
established the following decision rule for the off-facility: If the screening level risk assessment
predicts unacceptable risks in a given ISZ, the assessment will be refined (i.e., additional samples will
be collected to characterize each residence, exposure point concentrations will be established based on
these samples and will be compared to the remediation goal), if the screening level risk assessment
predicts acceptable risks in a given ISZ, that ISZ is considered to reguire no further action. Based on
this decision rule, additional soil samples were collected from each residence within ISZ 1, 6, and 7.
The refinement of the screening level assessment was completed after this supplemental soil sampling was
performed in ISZs 1, 6 and 7 in January, 1997. A comparison of these sampling results with the
residential remediation goals comprises the final risk assessment for the off-facility area. EPA will
manage risks for the off-facility area by individual yard.

8.1.1 Arsenic Risks

The risks associated with exposure to arsenic in soil are summarized in Table 12 excerpted from the final
Human Health Baseline Risk-Assessment. Current EPA policy, summarized in OSWER Directive 9355 0-30,
states that where the cumulative carcinogenic site risk to an individual based on the reasonable maximum
exposure for both current and future land use is less than 10 -4, and the non-carcinogenic hazard
guotient is less than 1, action is generally not warranted. Using this criteria, the cancer and
non-cancer risks associated with the reasonable maximum exposure to arsenic in soil by NCI workers are
predicted to be unacceptable in (i.e., warranting remedial action) in EU-3 and EU-4 only. The cancer and
non-cancer risks associated with the reasonable maximum exposure to arsenic in soil by CI workers are
predicted to be unacceptable in all exposure units. The cancer and non-cancer risks associated with the
reasonable maximum exposure to arsenic by residents within the on-facility area are unacceptable in one
exposure unit, EU-8. As can be seen in Figure 2, EU-8 is adjacent to areas where people are currently
living. However, no trailers are present and no people currently reside within this EU.

In the off-facility area, risks to residents are unacceptable in ISZ-8. Close inspection of ISZ-8 reveals
that the unacceptable risk is attributable to one property. The risk assessment broadly assumed that all
off-facility properties were used as residences. This particular property is used for a commercial
business (it is a lumber yard) and is expected to remain in commercial use in the future. Comparison of
soil concentrations to those considered to be acceptable for NCI workers demonstrates that risks are

-------
acceptable for commercial use of this property.

The risks associated with the reasonable maximum exposure to arsenic in ground water are summarized in
Table 13 excerpted from the final Human Health Baseline Risk Assessment. As can be seen in the table, the
non-cancer and cancer risks associated with exposure to arsenic in ground water are unacceptable for both
workers and residents.

The risk assessment also evaluated the potential risks associated with exposure of teenagers to slag
while visiting the Site. The cancer and non-cancer risks associated with the reasonable maximum exposure
to arsenic in slag are below a level of concern. The hazard quotient is 0.2 and the cancer risk is
1 x 10 -5.

8.1.2 Lead Risks

The health risks associated with exposure to lead are evaluated in a different manner than those
associated with exposure to arsenic. The health effect of most concern associated with lead exposure is
the impairment of the nervous system, especially in young children and unborn children. Analyses
conducted by the Centers for Disease Control and EPA associate levels of lead in the blood of 10
micrograms per deciliter (ug/dL) and higher with health effects in children. EPA's risk management goal
for lead is to achieve a level of protectiveness such that a typical child or group of similarly exposed
children would have an estimated risk of no more than 5% of exceeding the 10 ug/dL blood lead level. The
risk assessment results for lead exposure at the Site are reported as the probability of an individual
child or the fetus of an individual pregnant worker having a blood level above the 10 ug/dL goal. EPA's
Integrated Exposure/Uptake Biokinetic Model was used to assess risks to residential children. A
biokinetic slope factor approach was used to assess risks to adults and teenagers. The risk assessment
considered the exposed population within the on-facility EUs 1-7 to be adults.

The health risks associated with exposure to lead in soils at the Site are summarized in Tables 14 and 15
excerpted from the final Human Health Baseline Risk Assessment. Risks to NCI workers are predicted to
exceed EPA's health goals in EU-3 only. However, the health risks associated with exposure to lead in
soils by CI workers exceed EPA's health goals in all exposure units, with probabilities of 25%-99% of
exceeding the target blood lead level. The risks from exposure to lead within the on-facility residential
areas of EU 8, 9 and 11 are predicted to exceed EPA's health goals. In the residential areas south and
west of the site, risks from exposure to lead exceed EPA's health goals in ISZ-1, ISZ-3, ISZ-6, ISZ-7,
and ISZ-8. Close inspection of these results showed that ISZ-3 was occupied by the Murray High School and
commercial businesses and further, the elevated lead levels in ISZ-8 were associated with commercial
properties. Considering these land uses, the lead risks in ISZ-3 and ISZ-8 were determined by EPA to be
acceptable. Supplemental sampling and refinement of the risk assessment was limited to ISZ-1, ISZ-6, and
ISZ-7.

The risk assessment also evaluated the potential effect of the exposure of teenagers to slag while
visiting the Site. The assessment concluded that there is a less than 0.02% probability of exceeding
EPA's health based goal as a result of this exposure.

8.2. Ecological Risks

The ecological risk assessment evaluated potential exposures of fish, birds, mallard ducks, frogs, and
pocket gophers to smelter related chemicals of concern within likely habitat areas. Potential risks to
ecological receptors were estimated by calculating Hazard Quotients (HQs) and Hazard Indices (His). The
HQ is the ratio of environmental concentration or dose to a safe level or dose. If the HQ for a chemical
is equal to or less than 1, it is assumed that there is no appreciable risk that adverse health effects
will occur. If an HQ exceeds 1, there is some possibility that adverse effects may occur, although an HQ
above 1 does not indicate an effect will definitely occur. However, the larger the HQ value, the more
likely it is that an adverse effect may occur.

Hazard quotients for each contaminant at each location and by each pathway were summed to obtain a Hazard
Index (HI) for each receptor. Figures 14 to 17 summarize the His for the belted kingfisher, killdeer,
valley gopher, and the mallard. The assessment considered exposure via ingestion of water, sediment,
soil, and food within the ecolocical study area of the Site. The Hi's are calculated for both the No
Observed Adverse Effect Level (NOAEL) and the Lowest Observed Adverse Effect Level (LOAEL). The NOAEL HI
is appropriate to consider when determining risks to individual ecological receptors. The LOAEL HI best
characterizes risks to populations. Figures for the kingfisher and mallard also illustrate an adjustment
with "area use factors" as their home ranges are larger than the actual Site areas. All figures
illustrate risk up gradient of the Site, on-Site, down gradient of the Site, and in the depressions
(wetlands). Lead concentrations in soils and sediments as well as selenium concentrations in plants are
the largest contributors to risk to ecological receptors at the Site.

-------
Hazard quotients for trout and frogs were calculated by comparing exposure point concentrations for
surface water with toxicity reference values. The evaluation, documented in the ecological risk
assessment, shows essentially no risks to brown trout or frogs in Little Cottonwood Creek.

8.2.1 Discussion of Results

The estimate of relative risk is the risk estimate on-Site divided by the risk estimate up gradient. It
is a useful measure of how much higher the risk is due to the Site relative to inherent risks. The
estimate of absolute risk is the HQ or the HI for each location. As can be seen in Figures 14-17, in
general, the relative risks to terrestrial receptors on-Site are two or more times higher than the risks
observed up gradient. Both relative risk estimates and absolute risk estimates are considered by EPA when
determining if remedial action is warranted. There are essentially no risks to aguatic life in Little
Cottonwood Creek considering both relative and absolute risk estimates. The greatest areas of concern at
the Site are the wetlands, where both absolute and relative risk estimates are high.

Interpretation of these risk estimates must take into account the following sources of uncertainty in the
calculations:

1. Where measured concentration data were not available, literature based bioaccumulation factors were
applied to estimate concentrations. This use of predicted rather than measured data adds to the
uncertainty in the assessment. This uncertainty may be significant for the risks predicted for the
mallard and the pocket gopher, since predicted excess risk is associated with ingestion of
contaminants in vegetation. These plant concentrations driving the risk were predicted using
literature based bioaccumulation factors. Without true site measurements, it is difficult to ascertain
if this risk is representative.

2. Sample preparation may also lead to some degree of uncertainty. Benthic macro invertebrates which were
connected at this Site were not rinsed prior to analysis. This could lead to a carry over of sediments
thereby influencing contaminant levels in this media. Sediments were ground and acid-digested. This
method of treatment could possibly lead to a release of contaminants from the sediment which might not
typically be available to a receptor. Therefore, EPA believes preparation of samples collected from
this Site to support the ecological risk assessment may have contributed to artificially high metal
concentrations, thereby elevating risk estimates.

3. The risks were calculated on the assumption that the receptor spent 100% of its time within a
location. Depending on the home range and actual use of each location, the actual risks could be
lower.

Observations of the ecological receptors at the Site in the form of qualitative surveys documented in the
ecological risk assessment suggest that the predicted effects are not occurring. EPA believes that
further biomonitoring is needed to validate this assumption. Attempts to reduce the risks through active
measures such as removing and replacing sediments in the wetlands will likely result in loss of the
habitat. In EPA's judgement, the wetlands are of great ecological interest and loss of this habitat may
have a more negative impact on the local ecosystem than the highly uncertain predicted risks.

Also relevant to the discussion of ecological risks is the fact that current Site development plans
include extensive regrading which will likely result in filling of the wetlands. The Corps of Engineers
has jurisdiction over the wetlands if affected by development actions and may or may not allow the
filling of these wetlands. If it were to occur, the filling of the wetlands would be an ecological impact
in itself but would essentially break the exposure pathways of concern for ecological receptors.

8.3 Remedial Action Objectives

The baseline risk assessment provides the basis for EPA's decision that actual or threatened releases of
hazardous substances at the Site may present an imminent and substantial endangerment to public health,
welfare, or the environment. Specifically, unacceptable risks were identified for the following exposed
populations via the ingestion of arsenic and lead in dust and soil and the ingestion of arsenic in ground
water.
Current and Future NCI Workers
Current CI Workers
Current and Future Residents

EPA has determined that remedial action is warranted at this Site. Remedial Action Objectives (RAOs) were
developed by EPA for the exposure pathways and contaminants of concern associated with unacceptable risks
under the current and reasonably anticipated future land use. These RAOs are presented in this section.

-------
8.3.1 Overarching RAO

Development of the on-facility portion of the Site is a key assumption on which this remedy decision is
based. Integration of development and Site remediation is a goal of EPA's Brownfields program. EPA's
Brownfields Initiative is an organized commitment to help communities revitalize properties where
expansion or redevelopment is complicated by real or perceived environmental contamination, to mitigate
potential health risks, and to restore economic vitality. Based on consideration of Brownfields goals,
the key overarching RAO is:

       Develop a comprehensive remedy that protects human health and the environment, is
       consistent with the current and reasonably anticipated future land use,  and removes
       obstacles to Site development associated with real or perceived environmental contamination.

EPA developed media-specific RAOs using the basic assumption that the reasonably anticipated future land
use will be commercial/light industrial use of the on-facility area and residential use of the
off-facility areas where homes are currently located. EPA based this assumption on the information
gathered during the Site Characterization and subseguent Murray Smelter Working Group sessions all of
which is summarized in Section 3.1. This information supports EPA's conclusion that the current
industrial and residential use of the on-facility property will end in the very near future.

8.3.2 Chemical Specific Applicable or Relevant and Appropriate Requirements (ARARs)

In accordance with the National Contingency Plan  (NCP) ,  remediation levels are a subset of the RAOs and
consist of medium-specific chemical concentrations that are protective of human health and the
environment. These remediation levels are based on risk assessment or ARARs. Table 16 presents the
chemical specific ARARs for the Site which are incorporated into the RAOs as remediation levels to
address specific contaminants and exposure pathways. Appendix B presents the derivation of the risk based
remediation levels for soil which are also incorporated into the RAOs. Appendix C presents the technical
support for EPA's selection of the remediation level for arsenic in shallow ground water.

8.3.3 On-Facilitv Soil/Smelter Materials

RAOs:              Prevent unacceptable risks to current and future workers or to ecological
                   receptors due to the ingestion of soil/smelter materials containing arsenic
                   or lead.

                   Reduce the uncertainties in the predicted risks to ecological receptors

Remediation Levels:

The remediation levels for soils/smelter materials are risk-based.

                   For workers, prevent exposure to soils/smelter materials containing levels
                   of arsenic or lead which would pose a potential excess cancer risk greater
                   than 1E-4; a potential chronic health risk defined by a hazard guotient of
                   one; or result in a greater than 5% chance that the fetus of a pregnant
                   worker would have a blood lead level greater than 10 micrograms per
                   deciliter (Ig/dL). Based or the findings of the Baseline Human Health
                   Risk Assessment and a reasonably anticipated future land use that is
                   commercial/light industrial, these levels correspond to:

                        Surface soils shall not exceed 1,200 milligrams per kilogram
                        (mg/kg) arsenic as the 95% upper confidence limit on the arithmetic
                        mean within any given exposure unit.

                        Surface soils shall not exceed 5,600 mg/kg lead as the arithmetic
                        mean within any given exposure unit.

8.3.4 On-Facilitv Groundwater

RAOs:              Minimize future transport of arsenic from source materials to the shallow aguifer.

                   Prevent exposure of human and ecological receptors to ground water with
                   arsenic concentrations that represent an unacceptable risk.

                   Prevent unacceptable increases in the arsenic concentrations of the
                   intermediate aguifer resulting from arsenic migration from the shallow aguifer.

-------
Remediation levels:

The remediation levels for ground water are based on ground water ARARs.

Meet the MCL (0.05 milligrams per Liter (mg/L)) for dissolved arsenic in
shallow groundwater at the east and west Site boundaries.

Meet the MCL (0.05 mg/L) for dissolved arsenic in the intermediate aguifer.

Meet the Alternate Concentration Limit (ACL) of 5.0 mg/L for dissolved
arsenic, within the unconfined shallow aguifer within the Site boundaries.
The compliance points for the ACL in shallow ground water are in the
vicinity of ground water discharge locations south of Little Cottonwood Creek.

8.3.5 Little Cottonwood Creek Surface Water

RAOs Protect Little Cottonwood Creek water guality by preventing unacceptable
increases of arsenic concentrations in surface water resulting from ground
water discharges or surface water run-off from the Site.

Remediation Levels:

The remediation levels for surface water are based on surface water ARARs.

Meet the Utah Standards of Quality for Waters of the State for trivalent
arsenic of 190 micrograms per liter (ug/L) as a 4 day average and 360 ug/L
as a 1 hour average in Little Cottonwood Creek.

Meet the Utah Standard of Quality for Waters of the State for dissolved
arsenic of 100 ug/L in Little Cottonwood Creek.

8.3.6 Off-Facility Soils

RAOs: Prevent unacceptable risks to current and future residents due to the
ingestion of soil containing lead.

Prevent unacceptable risks to current and future NCI workers due to the
ingestion of soil containing lead.

Remediation Levels:

The remediation levels for off-facility soils are risk based.

The concentration of lead in surface soils within residential areas of the Site
shall not exceed 1200 mg/kg as an arithmetic mean within any given
residential yard. EPA developed a range of 630 mg/kg-1260 mg/kg for the
remediation level for soils in residential areas. Appendix B provides the
details of the development of this range. The April 23, 1997 risk
management strategy prepared by EPA provides the rationale for EPA's
selection of 1200 mg/kg as the appropriate remediation level for the
residential areas of this Site. The specific factors considered in making this
determination for each property were the current land use, the reasonably
anticipated land use, the likelihood of exposure to soil (measured
gualitatively by ground cover), and empirical evidence of exposure to lead.

The concentration of lead in surface soils within commercial areas of the
Site shall not exceed 5600 mg/kg as an arithmetic mean within any given
commercial property.

8.3.7 On-Facilitv Ecological Study Areas

RAO: Reduce uncertainties in predicted risks to ecological receptors.

-------
9.0
COMPARATIVE ANALYSIS
This section presents, a summary of the comparative analysis of the remedial alternatives developed for
the Site to achieve the RAOs. This two-stage analysis reviews the remedial alternatives in relation to
the threshold criteria and primary balancing criteria specified in the National Contingency Plan (NCP).
Modifying criteria are then discussed in Section 9.2. The findings of the comparative analysis are
summarized in Section 9.3, including selection of a comprehensive remedy for the entire Site.
9.1
Identification of Alternatives
A range of comprehensive remedial alternatives was developed to address human health risks and
environmental protection for the Site. For the purpose of organizing the various Site materials and their
associated environmental effects, smelter materials present in the on-facility area of the Site were put
into one of four categories based on information from the Site Characterization Report and the Baseline
Risk Assessment:
Category I and II:
Category I:
Category I and II materials are the sources of arsenic concentrations in
ground water above the ACL. Both relatively high arsenic concentrations
and large material volumes are necessary for material to be a potential
threat to ground water and be classified as Category I or II. Alternatives
were developed for Category I and II ground water source material to
achieve the RAO of minimizing future transport of arsenic from source
materials to the shallow ground water. Alternatives for Category I and II
material must achieve the remediation levels established for ground water.

Category I materials are distinct in that they are considered by EPA to be
principal threat wastes characterized as large volumes of material
containing relatively undiluted arsenic trioxide. There is an estimated
guantity of 2000 cubic yards of Category I material within the on-facility
area. The identification of Category I materials considers:

A. Associated with distinctly elevated arsenic concentrations in underlying
shallow ground water (greater than or egual to 15 mg/L);

B. High arsenic concentrations compared to other categories of materials on Site;

C. Visual characteristics (e.g., color, particle size) which indicate arsenic
trioxide;
Category II:
D. Direct contact risks which are considered to be a principal threat if this
material were ever brought to the surface at the Site; and

E. Located where former smelter structures which processed or stored
arsenic trioxide were historically located. Category I materials are located
in the areas of the arsenic kitchens, the western compartment of the
baghouse, and the arsenic storage bin(s). The exact limits of Category I
material will be defined in remedial design considering the results of
sampling material deeper or adjacent to this material.

Low level threat around water source material characterized as large volumes of
diluted arsenic trioxide or flue dust often mixed with soil, new fill, or debris from
former smelter flues. These materials have lower arsenic concentrations than
Category I materials and are potentially a significant source of ground water
contamination. There is an estimated guantity of 68,000 cubic yards of Category,
II material within the on-facility area. The identification of Category II materials
considers:
A. Located near or within the footprint of former smelter structures such
as the concrete flues, the roasting plant, the baghouse, storage areas,
transport areas, and the blast furnace area. The exact limits of Category II
material will be defined in remedial design considering the results of
sampling material deeper or adjacent to this material;

B. Visual characteristics (e.g., color, particle size) which indicate flue dust
or diluted arsenic trioxide; and

-------
C. Potential current or future threat to ground water quality. Category II
material is associated with arsenic in shallow ground water above the ACL.

Category III: Category III materials are surface soils which are predicted to pose an
unacceptable risk to NCI workers within the on-facility area. Alternatives for
Category III materials must achieve the remediation levels for on-facility,
soils/smelter materials. Material in this category will not pose a threat to ground
water. The identification of Category III materials considers;

A. Located within on-facility EUs identified as causing unacceptable health
risks to NCI workers (EU-3 and EU-4).

B. Lead concentrations greater than 5600 mg/kg as the arithmetic mean
within the EU; and

C. Arsenic concentrations greater than 1200 mg/kg as the 95% upper
confidence limit on the arithmetic mean within the EU.

Category IV: Slag

Remedial alternatives were developed to address all four categories of smelter materials. The key
components of each alternative considered in the comparative analysis are summarized below.

Alternative 1 - No Action

• The Murray Smelter Site would be left in its current condition.

Alternative 2 - Excavation & Onsite Consolidation/Barrier Placement/Monitored Natural
Attenuation/Institutional Controls Removal and Disposal of Off-Facility Soils

• Source control via excavation of Category I and II materials and consolidation in separate
repositories in the on-facility area.

• Monitored natural attenuation of shallow ground water within and down gradient of source
areas to achieve the ACL. The mechanism of attenuation of arsenic in shallow ground water is
adsorption to the iron oxides in the subsurface soil.

• Surface water monitoring in Little Cottonwood Creek and monitoring of the on-Site ecological
study area. Monitoring of wetlands will include surface water, sediment and benthic macro
invertebrates. Monitoring of terrestrial areas will include plants and soil.

• Institutional controls in the form of a Murray City ordinance establishing an "overlay
district" which includes zoning to prevent residential and contact intensive industrial uses
within the former smelter operational areas, prohibitions on the development or use of any
ground water wells within Site boundaries for EPA approved monitoring wells, maintenance of
the barriers, and controls on excavated subsurface material within the former smelter
operational areas. Other institutional controls include restrictive easements that run with
the land which contain the same land use and ground water well construction restrictions.

• Covering of Category III materials in place with barriers sufficient to prevent direct
contact. Such barriers may be pavement, landscaping, soil caps, or sidewalks.

• Soil removal/replacement with clean soil, or other fill in off-facility residential or
commercial properties with lead concentrations in soils that may represent an unacceptable
risk. Excavated soil will be used in the on-facility area of the Site as subgrade material
during development or road construction.

Alternative 3 - Excavation/Onsite Consolidations Offsite Disposal/Monitored Natural
Attenuation/Barrier Placement/Institutional Controls/Removal and Disposal of Off-Facility Soils

• The same actions as Alternative 2, except Category I materials are excavated and disposed
offsite.

Alternative 4 - Excavation/Onsite Consolidations Offsite Disposal/Barrier Placement/Institutional
Controls/Ground Water Extraction/Removal and Disposal of Off-Facility Soils

• All Alternative 3 components.

-------
       •      Ground water extraction in areas of richest arsenic concentrations,  treatment of extracted
              ground water,  and discharge to the sanitary sewer system.

Alternative 5 - Excavation/Onsite Consolidations Offsite Disposal/Barrier Placement/Institutional
Controls/In-Situ Ground Water Treatment/Removal and Disposal of Off-Facility Soils

       •      All Alternative 3 components.

       •      Option A - Constructed wetlands to treat shallow ground water prior  to discharge to Little
              Cottonwood Creek.

       •      Option B - Permeable barrier treatment wall to treat shallow ground  water prior to
              discharge to Little Cottonwood Creek.

Alternative 6 - Excavation/Onsite Consolidation & Off Site Disposal/Monitored Natural Attenuation/Barrier
Placement/Institutional Controls/Off-Facilitv Community Health Education, Monitoring and Intervention

       •      All Alternative 3 components for the on-facility area.

       •      Community health education and monitoring for residents and workers  in off-facility areas  of
              concern. This alternative also includes intervention actions such as surface control,
              barrier placement or soil removal,  if the potential for unacceptable risk is indicated by
              the monitoring program.

Alternative 7 - Excavation/Onsite Consolidation & Offsite Disposal/Monitored Natural
Attenuation/Barrier Placement/Institutional Controls/Soil Tilling in Off-facility Areas

       •      All Alternative 3 components for the on-facility area

       •      Deep tilling in off-facility residential or commercial  properties with lead concentrations
              in soils that may represent an unacceptable risk.  Institutional controls to protect the
              integrity of soil barriers and to place reguirements on the handling and disposal of any
              excavated material from beneath the tilled zone if the  concentrations in this material are
              above a level of concern.

9.1.1 Threshold Criteria Analysis

9.1.1.1       Overall Protection of Human Health and the Environment

As demonstrated in the Baseline Risk Assessment, Alternative 1, No Action does not meet the threshold
criteria of overall protection of human health and the environment except that no action is appropriate
for slag since no unacceptable risks associated with exposure to slag were identified by EPA in the
Baseline Risk Assessment. With the exception of Alternative 1, all alternatives considered in the
comparative analysis meet the reguirements of the RAOs and provide overall protection of human health and
the environment. Differences in overall protection are related to the level of certainty with regard to
actions for Category I materials and relative effectiveness of actions on ground water and the
off-facility soils. There are also differences with respect to the key overarching RAO reguiring that
remedial actions be consistent with the current and proposed land use.

Source control via excavation and consolidation of Category I and II materials in separate repositories
(Alterative 2) would prevent future infiltration of surface water, thus protecting ground water from
further impact due to transport of arsenic from this source material. Excavation/onsite consolidation is
an effective method of source control at this Site primarily due to the ease in locating the source
material. The material is generally within the locations of historical smelter structures. For example,
the results of sampling subsurface soils to a depth of 5 feet in the vicinity of the baghouse show that
excavation of the upper 2 feet of material from within the footprint of the former baghouse would remove
approximately 97 percent of the arsenic present in this source area.   (This calculation was done by
dividing the mass of arsenic in 2 feet by the total mass of arsenic measured in 5  feet of subsurface soil
at the location of the highest arsenic levels.)

Barrier placement over Category III materials is a component of all alternatives except Alternative I and
would be effective in preventing direct exposure as long as barriers are maintained. The institutional
controls which include public and private land use restrictions and a ban on construction of ground water
wells  (with the exception of EPA approved monitoring wells) within the on-facility area will prevent Site
uses which could result in unacceptable risks due to residential or contact intensive use or ground water
ingestion. In the off-facility area, soils containing lead exceeding remediation levels would be
excavated to at least 18 inches and the excavated soil brought onto the on-facility area for

-------
incorporation into remedial actions or development. The off-facility excavated areas would be replaced
with soil or other clean fill. Removal of soil with lead concentrations above remediation levels provides
protection of human health and the environment by breaking the exposure pathway of direct contact with
contaminated source material.

The source control action for Category I materials in Alternatives 3, 4, and 5 is off-site disposal.
Although both on-site disposal (Alternative 2) and off-site disposal (Alternatives 3, 4, and 5) actions
provide essentially the same level of overall protection, removal of Category I materials from the Site
would eliminate completely any long-term concerns regarding the potential for direct exposure (the levels
of arsenic in Category I materials may cause acute health effects) and the potential for the materials to
act as sources of arsenic to ground water in the future (arsenic in Category I materials is predominantly
the soluble oxide and sulfate forms) in the event that the repository was damaged resulting in a release
of these materials into the environment. Although not likely to occur, the possibility of its occurrence
illustrates the difference between the two alternatives.

Alternative 4 contains the same components as Alternative 3 and adds a ground water extraction system.
Site specific hydrologic and chemical factors limit arsenic transport rates to the extraction wells and
thus limit the rate at which arsenic may be removed from the aguifer. Long term pumping rates are limited
by the flux or supply of ground water introduced to the aguifer. Section 4 of Appendix A of the
Feasibility Study contains a conceptual design for a ground water extraction system and approximate time
frames are predicted for arsenic extraction rates. The analysis demonstrates that the flux of water
through targeted portions of the aguifer will not change as a result of installing a pumping system.
Therefore, addition of an extraction system within the source areas will not accelerate the rates of
decline in arsenic concentrations in ground water relative to the rates achieved through source control
and natural attenuation. The time frame reguired to meet remediation levels in ground water within the
source areas is predicted to be between 100-125 years with the installation of a ground water extraction
system. Monitored natural attenuation is predicted to reguire approximately 100-150 years to achieve
remediation levels throughout the Site. For both source control with monitored natural attenuation and
source control with ground water extraction, the same set of site specific factors limits the rate at
which arsenic concentrations will decline. In addition, operation of an extraction system may not be
compatible with the desired future land use, because of the large area and numerous wells necessary.

Alternative 5 contains the same components of Alternative 3 and adds in-situ treatment of shallow ground
water (either by constructed wetlands or by a permeable barrier treatment wall) near Little Cottonwood
Creek. Currently ground water discharges to Little Cottonwood Creek. However, the principal areas of
elevated arsenic concentrations in ground water are distant from the creek and are not predicted to
intercept the creek for over 100 years. Due to source control and attenuation within the aguifer, arsenic
concentrations are expected to be significantly lower by the time arsenic from these areas intercepts the
creek. The types of treatment systems included in Alternative 5 are not expected to be effective for
periods greater than 10 years without extensive routine maintenance. Implementation of either treatment
option will have limited short-term effectiveness due to the diffuse source areas which may include
ground water from both sides of the creek and surface water runoff and complex ground water flow patterns
near the creek and may provide no benefit for long-term effectiveness in reducing arsenic transport to
Little Cottonwood Creek. Therefore, implementation of in-situ ground water treatment systems is not
expected to provide additional performance over the source control and monitoring actions included in
Alternative 3.

Alternatives 6 and 7 include two different options for addressing the off-facility soils containing
unacceptable concentrations of lead. Alternative 6 includes community education to inform residents on
methods to prevent unacceptable exposures and a voluntary blood-lead monitoring program. If the
monitoring program indicates the potential for unacceptable risk intervention actions would be
implemented. These actions would be designed on a case-by-case basis and could include surface control
such as vegetation of bare areas, barrier placement or soil removals. This alternative is expected to be
protective of human health if participation in the program is sufficiently high. Alternative 7, soil
tilling, is also expected to be protective of human health and the environment. In the majority of
off-facility areas of concern, lead concentrations are elevated at the surface. The source of this lead
is likely due to deposition of emissions from the smelter during its period of operation. In these cases,
deep tilling will reduce lead concentrations to below levels of concern. Site characterization data
indicate that at some locations lead concentrations are above a level of concern over the entire tilling
zone, possibly due to the placement of slag. In these areas, lead concentrations in surface soils would
not be reduced below a level of concern by tilling and community health education and monitoring would be
implemented to provide long term protection.

-------
9.1.2 Compliance with Applicable or Relevant and Appropriate Requirements

9.1.2.1 Ground Water ARARs

Chemical specific ARARs are identified in Table 16. Section 121(d)(2)(B)(ii) of the Comprehensive
Environmental Response Compensation and Liability Act (CERCLA) allows EPA to establish alternate
concentration limits (ACLs) to those otherwise applicable under the following conditions stated in 55
Federal Register 8732:

The ground water must have a known or projected point of entry to surface water with no
statistically significant increase in contaminant concentration in the surface water from
ground water at the point of entry, or at any point where there is reason to believe
accumulation of constituents may occur downstream. In addition, the remedial action
must include enforceable measures that will preclude human exposure to the contaminated
ground water at any point between the facility boundary and all known and projected
points of entry of such ground water into surface water.

Quarterly monitoring of surface water and ground water at the Site has demonstrated that ground water
from the shallow aguifer discharges to Little Cottonwood Creek at locations along the northern Site
boundary. The contaminant of concern in ground water is arsenic. Information collected since April, 1997
and documented in guarterly monitoring reports indicates the primary source of arsenic to Little
Cottonwood Creek exists at a point discharge at the eastern facility boundary. Loading calculations
indicate that 88%-100% of the arsenic loading to Little Cottonwood Creek is due to this point discharge,
not to the ground water discharge from the Site. EPA has determined that the conditions at Murray Smelter
satisfy the reguirements of CERCLA 122(d)(2)(B)(ii) which allow the establishment of an ACL for
groundwater. EPA has established 5.0 mg/L as the ACL for dissolved arsenic in ground water. Appendix C
provides a summary of the calculations used by EPA to determine a range of acceptable ACLs. In making its
determination, EPA considered the zone of potential shallow ground water discharge from the Site and
conservatively assumed all discharge is from the Site or south side of the creek. EPA also based its
determination on low flow conditions in Little Cottonwood Creek and Site specific hydraulic conductivity
and hydraulic gradient measurements. The ACL of 5.0 mg/L for dissolved arsenic is a ground water
concentration which will assure that Little Cottonwood Creek is protected at its beneficial use,
agricultural use, given the discharge of shallow ground water to the creek.

In accordance with the NCP, the situation at Murray Smelter fulfills the CERCLA statutory criteria for
ACLs, including the analysis in the Feasibility Study which demonstrates that active restoration of the
groundwater to MCLs is not practicable. The existing documentation of these conditions precludes the need
for an ARAR waiver. The remediation level for dissolved arsenic in shallow ground water within the Site
boundaries is the ACL of 5.0 mg/L. Achieving this level will constitute compliance with the groundwater
ARARs. The MCL is currently met at the on-facility area boundaries (the north boundary is north of the
ground water-surface water mixing zone north of Little Cottonwood Creek).

Source control actions contained in Alternatives 2 through 5 are expected to minimize transport of
arsenic from smelter materials and result in improvement of ground water guality such that the ACL will
be met within the entire on-facility area in a time frame of 100-150 years. This approach is reasonable
given the unlikelihood that the shallow aguifer will ever be in demand as a drinking water source.
Improvement in ground water guality would also reduce arsenic discharge to the creek. The additional
action of ground water extraction contained in Alternative 4 would not result in a significant decrease
in the time reguired to meet the ACL or a reduction in arsenic loading to the creek. Time frames for
achieving the ACL in Alternative 4 are estimated to be 100-125 years. There are fundamental technical
limitations for the effective performance of an extraction system related to the low aguifer yield and
high partitioning of arsenic to aguifer solids. The additional action of an in-situ treatment contained
in Alternative 5 would not contribute to reduction of current arsenic concentrations in the shallow
aguifer and would have a minimal effect on near- and long-term loading of arsenic to Little Cottonwood
Creek. The areas of highest arsenic concentration are currently distant from the creek and are not
predicted to intercept the creek for at least 100 years. Attenuation by adsorption is expected to
significantly reduce the arsenic concentrations from these areas by the time they reach the creek.

The ACL is currently achieved at monitoring well MW-112, the well location closest to compliance points
near Little Cottonwood Creek which will be established as part of the remedy. Within 30-40 years, the
effects of the source control actions of Alternative 3 along with the monitoring activities are expected
to demonstrate that the rate of natural attenuation of arsenic in shallow ground water is sufficient to
predict that the ACL will never be exceeded at the established compliance points. EPA expects the
remaining areas of the shallow aguifer to achieve the ACL within a time f:rame of 100-150 years.

Although not identified as a contaminant of concern, selenium has been detected in the shallow ground
water within the Site boundaries at levels exceeding the MCL of 0.05 mg/L. These detections are at 8 well

-------
locations within the on-facility area. However, the MCL for selenium has consistently been met at well
locations just south of Little Cottonwood Creek and the east and west on-facility boundaries (well
locations MW-112, MW-109, MW-102, and MW- 104 on Figure 6). The preamble to the NCP states at 55 Federal
Register 8753 :

"...there may be certain circumstances where a plume of ground water contamination is
caused by releases from several distinct sources that are in close geographical proximity."

In cases such as these, the NCP preamble suggests that

"...the most feasible and effective ground water clean up strategy may be to address the
problem as a whole, rather than source by source, and to draw the point of compliance to
encompass the sources of release."

EPA considered this discussion, the proximity of the sources of arsenic and selenium (both within the
former smelter operational area) , as well as the reliability of the restrictions on ground water use
within the Site boundaries in establishing the points of compliance for the selenium MCL at the well
locations just south of Little Cottonwood Creek. The ground water ARAR for selenium is currently met at
the points of compliance. Selenium will be included as part of the ground water monitoring component of
the remedy.

9.1.2.2 Surface Water ARARs

State of Utah Water Quality Standards are identified as applicable in Table 16. The data gathered during
the site characterization effort and subseguent sampling events indicate that Utah's aguatic life
standard for arsenic (0.19 mg/L arsenic as As [III]) is consistently being met, but that the arsenic
standard for agricultural use (dissolved arsenic of 0.1 mg/L) is not being met during low-flow conditions
within the on-facility boundaries. The standards for both uses were met at location SW-6, in Little
Cottonwood Creek downstream of the Site during the site characterization sampling events.

The source control actions will address the Murray Smelter-related source of the arsenic in the point
discharge from the culvert along State Street which discharges to Little Cottonwood Creek. This source
has been located near the storm drain along State Street near the Doc and Dell's trailer court. The
control of this discharge and the natural attenuation of shallow ground water to the level of the ACL is
expected to result in compliance with the surface water ARARs in Little Cottonwood Creek within a period
of 3 years. The improvement of ground water guality as a result of source control, natural attenuation
and surface water management will protect Little Cottonwood Creek in the future.

Little Cottonwood Creek does not currently meet the beneficial use for agriculture due to high levels of
TDS from urban runoff and high phosphorus. Neither TDS nor phosphorus are related to the Site. An
investigation of the actual use of Little Cottonwood Creek was conducted in April, 1997. Two diversions
of surface water were observed up gradient of the Site, neither of which was for agricultural use
purposes. No diversions were observed down gradient of the Site. This information suggests that the
current uses of Little Cottonwood Creek are not consistent with the beneficial use. EPA believes that a 3
year period for achieving the agricultural use standard for dissolved arsenic in Little Cottonwood Creek
is reasonable in this case.

9.1.2.3 Action- and Location- Specific ARARs

Tables 17 and 18 present the action specific and location specific ARARs for the Site. All alternatives
will meet these ARARs. On-facility alternatives which include consolidation of source materials within
the Site boundaries do not trigger the land disposal restrictions, therefore these reguirements are not
applicable. The Site boundaries are considered by EPA to be an "Area of Contamination" as defined in the
NCP. Movement of waste within an Area of Contamination does not constitute placement.

9.1.3 Primary Balancing Criteria

9.1.3.1 Short-Term Effectiveness

As discussed above, all alternatives with the exception of Alternative 1, No Action, meet the
reguirements of the RAOs and provide overall protection of human health and the environment. There are no
substantial differences between alternatives 2,3,4,and 5 in terms of short-term effectiveness. Each
alternative entails excavation and handling of Category I and II materials. However, dust control
measures are easy to implement and the potential for risks to the community or workers will be minimized.
Short-term risks from the presence of heavy construction eguipment on the Site would be similar with
respect to each alternative as well as to the potential risks posed by current industrial uses. Response
objectives would be met at the same time for all alternatives once excavated materials are disposed and

-------
barriers installed.

Alternative 6 contains community health education and monitoring for the off-facility area. This
alternative provides a high level of short term effectiveness. Although there are potential uncertainties
associated with the willingness of residents to participate, the high level of involvement by Murray City
and the high level of community awareness concerning the Site suggest that the program will be effective
in the short term. Alternative 7, tilling in the off-facility areas, may not be as effective as soil
removal in breaking the exposure pathway due to the presence of lead below the tilling zone.

9.1.3.2 Long-term Effectiveness and Permanence

A primary consideration in the evaluation of long-term effectiveness is that a major portion of the
on-facility area is expected to be redeveloped in the near future. The expected land use of office/light
commercial will reduce the potential for unacceptable risks of ("contact intensive" activities would
end) , and integration of remedial actions with redevelopment, the key overarching RAO, would allow for
optimizing the management of smelter materials remaining at the Site such that confidence would be
increased that the remedy and subseguent institutional control/monitoring will be effective over the long
term.

Alternatives 2 and 3 differ in terms of actions on Category I materials. Under Alternative 2, Category I
materials would be excavated and consolidated in a repository in the on-facility area. Under Alternative
3, Category I materials would be excavated and disposed of off-site. Given the current and reasonably
anticipated future land use and the opportunity to install a repository in a suitable location under the
control of Murray City, both actions would provide long-term protection of human health and the
environment: Removal of Category I materials from the Site would completely eliminate any future concerns
regarding the potential for direct exposure or contact of Category I materials with infiltrating ground
water and therefore provide a higher level of performance in terms of long-term effectiveness. For
Category II materials, consolidation into a repository would provide long term protection of human health
and the environment. Category II materials may be low-level sources of arsenic to ground water under
ambient infiltration conditions. Minimizing the potential for infiltration of surface water through these
materials by consolidation beneath a low-permeability barrier with surface control is expected to be
effective in preventing migration of arsenic to ground water. This same action is included in
Alternatives 2, 3, 4 and 5. Control on the use of land and ground water, the second component of the
institutional controls, will be effective in preventing direct contact with unacceptably high levels of
arsenic and lead in soil and ground water and will prevent the migration of arsenic from the shallow
aguifer to the intermediate aguifer. These controls will be implemented through city zoning and
restrictive easements which run with the land. Thus they will be effective in the long term and are
considered permanent restrictions.

Alternatives 3, 4 and 5 contain the same actions on smelter materials and provide the same basic level of
long-term effectiveness. Alternatives 4 and 5 include additional actions to contain the extent of arsenic
transport. Alternative 4 contains a ground water extraction system in the areas of highest arsenic
concentrations in the shallow ground water. The additional action of ground water extraction would
eventually provide for reductions in arsenic concentrations in shallow ground water and would be
effective for long-term containment of arsenic already present in the shallow aguifer near the former
baghouse and thaw house areas. However, modeling indicates that an extensive ground water extraction
system would not substantially reduce the time reguired to achieve the RAOs for the shallow aguifer and
Little Cottonwood Creek. Overall, Alternative 4 provides lower performance than Alternative 3 with
respect to long-term effectiveness because it would not provide a significant improvement in
environmental conditions relative to Alternative 3 and would entail a high level of operation and
maintenance.

Alternative 5 includes in-situ treatment of shallow ground water in the vicinity of Little Cottonwood
Creek with the purpose of limiting arsenic transport and discharge to the creek. Groundwater monitoring
indicates that the two principal areas of ground water contamination do not currently extend to Little
Cottonwood Creek and are not predicted to do so for more than 100 years. Source control actions and
natural attenuation of arsenic in the aguifer are expected to significantly reduce the arsenic
concentration by this time. The long-term performance of systems such as constructed wetlands and
permeable barrier treatment walls to treat arsenic is limited. Effective removal is only expected for a
period of approximately 10 years due to the mildly-oxidizing groundwater chemistry. Therefore, if these
types of systems were installed in the near future, they would not be effective at the time when arsenic
from the principal source areas reaches them.

In the off-facility area, lead concentrations in residential soils range up to 1,800 mg/Kg. The
remediation level lead in soil in the off-facility area is 1,200 ppm. Alternative 6, which includes
community education to provide information on methods to prevent unacceptable exposure, is expected to
provide long-term protection of human health through the education/monitoring components with additional

-------
assurance due to the option for intervention measures in the future if the potential for unacceptable
exposures is indicated.

For Alternative 7, because lead concentrations are above levels of concern throughout the tilling layer
at some locations, tilling may not always be effective in reducing concentrations to below the level of
concern. In this case, Alternative 7 would rely on similar community education measures described under
Alternative 6. Therefore, Alternatives 6 and 7 essentially provide the same level of long-term
effectiveness.

The off-facility component of Alternatives 2,3,4, and 5 would provide a high level of long-term
protection because surface soils with lead concentrations above a level of concern would be excavated and
replaced with clean soil or other fill. If complete removals are achieved, this action would provide the
highest level of long-term effectiveness because all soils of concern would be removed.

9.1.3.3 Reduction of Toxicity, Mobility and Volume Through Treatment

With the exception of the no action alternative, the alternatives considered by EPA do not provide
significantly different performances in terms of reduction of toxicity, mobility and volume of arsenic or
lead through treatment.

Alternatives 2 and 3 do not contain any treatment components except the possible treatment of Category I
material before disposal at an off-site facility. For Alternative 2, a reduction in the mobility of
arsenic in subsurface soils would be expected due to the minimization of infiltration through Category I
and II materials. For Alternatives 3, 4 and 5 a similar reduction would be expected due to removal of
Category I materials and minimization of infiltration through Category II materials.

Alternative 4 contains a treatment component; treatment of extracted ground water to remove arsenic prior
to discharge to the sanitary sewer. This treatment component would provide little if any reduction in
toxicity, mobility or volume of arsenic at the Site in comparison to Alternatives 2 and 3. However, the
ground water extraction system would provide some additional reduction in mobility of arsenic in the
shallow aguifer relative to Alternatives 2 and 3 due to physical containment of arsenic related to
sources in the former thaw house and baghouse areas. The aguifer characteristics which result in low-flow
rates and high arsenic attenuation currently limit the mobility of arsenic and an extraction system would
have minimal additional benefit.

The in-situ treatment of shallow ground water in the vicinity of Little Cottonwood Creek contained in
Alternative 5 would not provide any reduction in toxicity or volume of arsenic at the Site. It would
provide a minor reduction in the mobility of arsenic in shallow ground water near Little Cottonwood
Creek. As discussed above, the principal areas of ground water contamination are distant from the creek
and arsenic from these areas is not predicted to intercept the creek for over 100 years. At this time,
the arsenic concentrations are predicted to be significantly lower due to the high attenuation of arsenic
in the aguifer. Passive constructed wetlands or a treatment wall would be expected to operate efficiently
for only 10 years without continued routine maintenance and would, therefore, not be effective for the
time frame of principal interest.

Overall, therefore there are no substantial differences in performance of the alternatives against this
criterion. Alternatives 2, 3 and 5 perform at essentially the same level, whereas Alternative 4 performs
at a slightly higher level due to physical containment of arsenic in shallow ground water.

For the off-facility area, lead is immobile in Site soils and lead concentrations in the off-facility
area are well below levels which would warrant treatment. Treatment is therefore not applicable to
off-facility soils.

9.1.3.4 Implementability

The source control activities contained in Alternatives 2, 3, 4 and 5 are implementable either for
current land use or for the expected future land use. Excavation of Category I and II materials would be
implementable with some minor disruptions to current industrial activities. Physically suitable
repository locations for Category I and II materials are also available for current or future land use.
Off-site disposal of Category I materials (a component of Alternatives 3, 4 and 5) would also be readily
implementable. In addition, barrier placement over Category III materials would be implementable with
minor disruption to current industrial/commercial activities, or could be implemented during
redevelopment of the area. Institutional controls to protect barriers are implementable given the high
degree of involvement of the current land owners and Murray City.

Alternatives 4 and 5 contain the same source control actions as Alternative 3 with the addition of two
types of remedial action alternatives on ground water. The extraction system contained in Alternative 4

-------
would be difficult to implement due to the low yield of the aquifer and high partitioning of arsenic to
the aguifer solids. A large number of wells would be necessary, each pumping at a low rate over an
extended period of time. Operation and maintenance of this type of system, including a treatment plant
would be difficult and would not be compatible with future land use. Alternative 4 therefore has a lower
performance than Alternatives 2 and 3 in terms of implementability. Either of the options evaluated for
in-situ ground water treatment under Alternative 5 (wetlands or treatment wall) would have numerous
technical difficulties associated with effective implementation and operation. Considerations include the
limited area available (for wetlands), depth and complex flow patterns of ground water in the vicinity of
the creek, the presence of the units in the flood plain, and uncertainties associated with the
effectiveness of the technologies in removing arsenic. In addition, the technologies would reguire a high
level of long-term maintenance. For the ground water conditions found at the site, effective performance
of the types of technologies under consideration is approximately 10 years without on-going maintenance.
Replacement of substrate in a wetlands or of ferric sulfate in a treatment wall may be required at
approximately 10-year intervals. This action would not be compatible with the future land use and
Alternative 5 has a lower performance than Alternative 3 in terms of implementability.

In the off-facility area, community health education and monitoring programs contained in Alternative 6
would be readily implemented because only non-engineering controls are considered. Excavation and soil
replacement evaluated under Alternatives 2-5 are also expected to be readily implemented. Residents in
the areas of concern have participated in the site characterization study, and there is a high level of
awareness concerning the Site in the general community. These types of actions have been performed at
several sites around the country. Alternative 7, which requires soil tilling rather than excavation at
the same locations, would be more difficult to implement than the other alternatives. This is primarily
due to technical difficulties of tilling in small spaces such as residential yards, where structures and
plants would make some areas difficult to access.

9.1.3.5 Cost Analysis

Details of the cost analysis are contained in the final Feasibility Study. The costs estimated for the
on-facility area are shown in Table 19.

Table 19

Estimated Costs - On-Facility Area (Millions)

Alternative Alternative Alternative Alternative Alternative
Item 2 3 4 5a 5b

Capital Cost $8.7 $8.9 $10.8 $10.6 $21.9

Annual O&M $0.14 $0.14 $0.27 $0.21 $0.23

Present Net $10.1 $10.3 $14.3 $13.4 $40.2
Worth

O&M costs are estimated for 30 years. The extraction component of Alternative 4 and in Situ treatment
components of Alternative 5 would require O&M for over 100 years and so would entail substantially higher
costs than shown above.

The cost to implement Alternatives 2 and 3 is considered to be low; the costs to implement Alternative 4
and 5a are considered to be moderate; and the cost to implement alternative 5b is considered to be high.

The costs estimated for off-facility alternatives are shown in Table 20.

Table 20
Summary of Estimated Costs for Off-Facility Remedial Alternatives (Millions)

Alternative Alternative Alternative
Item 6 7 2-5

Capital Cost $0.57 $0.64 $1.1
Annual O&M $0.05 $0.015 $0.013
Present Net $1.34 $0.93 $1.33
Worth

-------
9. 2 Modifying Criteria

9.2.1 State Acceptance

The Utah Department of Environmental Quality (UDEQ) was provided the opportunity to review and comment on
all documents generated in support of this remedial action decision. UDEQ also participated in all
meetings of the Murray Smelter Working Group and the technical task group meetings. In comments on the
Proposed Plan. UDEQ indicates agreement that Alternative 3 is the most reasonable choice for the Site.
However, UDEQ indicated that this agreement was not based on the length of time or the current levels of
contamination. Given the "extremely long" time frames and uncertainty involved in ground water
restoration under any alternative, UDEQ has determined that it is technically impracticable within a
reasonable time frame to meet ARARs at this Site, and has agreed on that basis and for other reasons
given in this ROD (e.g., protection of human health and the environment) that the remedy described in
this ROD is appropriate. While EPA characterizes the situation differently, both parties are in agreement
about the ultimate approach. UDEQ believes that the long time frame for achieving ground water
remediation levels is acceptable only in the context of the technical impracticability
of any alternatives.

EPA's responses to UDEQ's comments an the Proposed Plan are provided in the Responsiveness Summary of
this ROD.

9.2.2 Community Acceptance

Few comments were received from the community on the Proposed Plan. Based on these comments and EPA's
extensive work with the community through the Murray Smelter Working Group sessions, it appears that the
community accepts EPA's selected remedy presented in Section 9.4. EPA's responses to verbal and written
comments on the proposed plan are provided in the Responsiveness Summary of the ROD.

9.3 SUMMARY

Seven remedial alternatives were evaluated for the Murray Smelter Site. Through an analysis using the
nine criteria of the NCP, EPA has selected Alternative 3 as the Site remedy. The remedy consists of the
following components:

• Ground water in the shallow aguifer contaminated with arsenic at levels above the ACL of 5.0
mg/L dissolved concentration will be addressed via source control and monitored natural
attenuation as follows:

1. Source control will be implemented by excavation and off site disposal
of the principal threat wastes at the Site, an estimated guantity of 2000
cubic yards of Category I material defined in Section 9.1 of this ROD.
This material is considered a principal threat due to its high mobility and its
demonstrated ability to act as a source of ground water contamination. In
addition, direct contact with this material may result in acute human health
risks. Off site disposal will be conducted in accordance with EPA's Off Site
Rule, 40 CFR 300.440 and the generator reguirements identified in Table 17.

2. Further source control will be implemented by excavation of
approximately 68,000 cubic yards of low level threat waste, Category II
material defined in Section 9.1 of this ROD. This material will be
consolidated within a repository system constructed within the Site
boundaries in accordance with the ARARs identified in Table 17. The
repository will be designed as the base for a new access road through the
Site which was planned by Murray City. The access road is expected to be
the catalyst for Site development to commercial/retail uses.

3. Monitored natural attenuation will address the residual ground water
contamination within and down gradient of these source areas. Monitored
natural attenuation will continue until shallow ground water achieves the
level of the ACL for dissolved arsenic of 5.0 mg/L. The intermediate
aguifer will also be monitored to demonstrate continued compliance with
the MCL of 0.05 mg/L dissolved arsenic.

4. The shallow aguifer will be monitored to evaluate the concentrations of
selenium at the established compliance points south of Little Cottonwood
Creek. The selenium monitoring is not for evaluation of the remedy, it is to

-------
                      ensure continued compliance with the selenium MCL.

                      5. Institutional controls in the form of a Murray City ordinance
                      establishing an "overlay district" and restrictive easements that run with the
                      land which both will prohibit the construction of new wells or use of
                      existing wells within the on-facility area and the western and eastern
                      portions of the off-facility area except for EPA approved monitoring wells.

       •       Surface soils (0"-2")  within the on-facility area contaminated with lead and arsenic
              exceeding remediation levels of 1200 mg/kg arsenic as  the 95% upper confidence limit on the
              arithmetic mean within an EU or 5600 mg/kg lead as the arithmetic mean within an EU  will be
              addressed as  follows:

                      1. Soils will be covered in place with barriers sufficient to prevent direct
                      contact. Such barriers may be pavement, landscaping,  soil caps, or
                      sidewalks. Site development itself is expected to result in additional
                      protection of human health since land uses associated with unacceptable
                      human health risks will end. Also, development will result in the
                      construction of additional barriers  (new buildings, roads, sidewalks parking
                      lots, and landscaping) over remaining surface soil and slag. Although no
                      unacceptable risks associated with exposure to slag were identified by
                      EPA,  the development of the Site will ensure no exposure to slag in the future.

                      2. Institutional controls in the form of a Murray City ordinance will
                      establish an "overlay district" which includes zoning to prevent residential
                      and contact intensive industrial uses within the former smelter operational
                      areas and will reguire maintenance of the barriers and controls on
                      excavated subsurface material within this same area.  Restrictive easements
                      than run with the land will be established in addition to the overlay district
                      to prevent residential or contact intensive industrial uses.

       •       Off-facility  surface soils (0"-2")  containing levels of lead  exceeding 1200 mg/kg as the
              arithmetic mean in individual residential yards or 5600 mg/kg as the arithmetic mean in
              commercial areas will be removed to a depth of 18 inches and  replaced with clean fill.  Any
              landscaping disturbed in this action will be replaced.  The removed soil will be used
              on-facility as subgrade material in construction of the repository system.

       •       Surface water of Little Cottonwood Creek will be monitored to ensure continued protection
              during the ground water natural attenuation process at the level of 190 ug/L as a 4  day
              average for trivalent arsenic and 360 ug/L as a 1 hour average for trivalent arsenic and 100
              ug/L for dissolved arsenic.

              The established ecological study area will be monitored and the resulting information will
              be used to reduce the uncertainties identified in the final Ecological Risk Assessment for
              the Site. Monitoring of wetlands will include surface water,  sediment and benthic macro
              invertebrates. Monitoring of terrestrial areas will include plants and soil.

The goals of the selected remedy are to protect the intermediate and deep principal aguifer at the level
of the MCL for dissolved arsenic, to restore the shallow ground water to the level of the ACL of 5.0 mg/L
for dissolved arsenic established to protect Little Cottonwood Creek at its beneficial use, and to
remediate surface soils to levels protective of the reasonably anticipated future land use. The remedy
incorporates the construction of a new north-south access road through the Site which will encourage
future development of the Site and achieve Murray City's goal of more appropriate land use through Site
development.

Based on information obtained during the Site investigation and on a careful analysis of all remedial
alternatives, EPA believes  that the selected remedy will achieve these goals. It may become apparent
during the monitored natural attenuation process for ground water that dissolved arsenic levels have
ceased to decline and are remaining constant at levels higher than the ACL over some portion of the plume
within the shallow aguifer. If it is determined on the basis of system performance data that certain
portions of the aguifer cannot be restored to the ACL, EPA will prepare a justification for a waiver of
the ground water ARAR based on technical impracticability of achieving further contaminant reduction.

-------
10. Statutory Determinations

Under its legal authorities, EPA's primary responsibility at Superfund sites is to undertake remedial
actions that achieve adequate protection of human health and the environment. In addition, section 121 of
CERCLA establishes several other statutory requirements and preferences. These specify that when
complete, the selected remedial action for this Site must comply with applicable or relevant and
appropriate environmental standards established under Federal and State environmental laws unless a
statutory waiver is justified. The selected remedy also must be cost-effective and utilize permanent
solutions and alternative treatment technoloqies or resource recovery technoloqies to the maximum extent
practicable. Finally, the statute includes a preference for remedies that employ treatment that
permanently and significantly reduce the volume, toxicity, or mobility of hazardous wastes as their
principal element. The following sections discuss how the selected remedy meets these statutory
requirements.

10.1 Protection of Human Health and the Environment

The Baseline Human Health Risk Assessment identified unacceptable risks over the entire on-facility area
associated with potential direct contact with lead- and arsenic-contaminated soil and smelter debris by
workers engaged in outdoor industrial activities. The assessment identified substantially less risk
(although still unacceptable in limited on-facility areas) associated with exposure to the same materials
under a scenario of commercial uses wherein workers would be primarily indoors. The assessment also
identified unacceptable risks associated with direct exposure to lead contaminated soil by residents and
commercial workers in the off-facility area. Potential ingestion of ground water from the shallow aquifer
within the Site boundaries was also predicted to result in unacceptable risk.

There is a large portion of the on-facility area where slag is exposed at the surface. It is not likely
that commercial or industrial workers or other adults will spend much time in areas of exposed slag.
Therefore, direct contact with slag by workers or residents is likely to be minimal. However, area
teenagers have been observed to visit the site in areas where slag is exposed. The Baseline Human Health
Risk Assessment characterized risks to teenagers who congregate in areas along Little Cottonwood Creek
and are potentially exposed to slag. The assessment concluded that risks associated with exposure to slag
are within the range that EPA considers to be acceptable.

The selected remedy employs ground water source control via excavation and off-site disposal of the
principal threat at the site, undiluted arsenic trioxide, and will effectively address the identified
risk associated with potential migration of this material into shallow ground water and potential future
direct contact with this material.

The second component of the selected remedy is ground water source control by excavation and
consolidation of ground water source material within an on-Site repository system. The system will be
designed with surface water management features. This action will effectively control the infiltration of
surface water into arsenic contaminated soil and prevent further migration of arsenic into shallow ground
water. The on-Site repository system will be designed to perform as an adequate base for a new access
road from Vine Street to 5300 South Street. The repository thus will serve three functions in the
protection of human health at the Site:

(1) Reduction of mobility of arsenic to ground water by off-Site disposal of and containment of ground
water source material to address risks associated with exposure to contaminated ground water;

(2) Containment of contaminated material which presents unacceptable risks due to direct contact thereby
eliminating this exposure pathway; and

(3) Catalyst for development of the Site by providing the base for a roadway which is expected to provide
the necessary access to promote commercial uses. The Site development will address the unacceptable
risks associated with high contact industrial outdoor activities.

The third component of the selected remedy is a comprehensive public and private institutional controls
package which will restrict the use of ground water within the Site boundaries (with the exception of EPA
approved monitoring wells) and restrict land uses other than general commercial uses as defined by the
Murray City land use code. The institutional controls package will also require that Site features such
as roads, parking lots, and landscaping, which are functioning as barriers to human exposure be
maintained. The institutional controls will provide human health protection into the future. The Site
development itself is expected to result in protection of human health through the construction of
barriers over remaining low level surface contamination and slag. Although no unacceptable risks
associated with exposure to slag were identified, the development of the site will ensure no exposure to
slag in the future.

-------
The fourth component of the selected remedy is monitored natural attenuation of ground water down
gradient of source areas. Analyses performed during Site Characterization and summarized in the final
Site Characterization Report demonstrate that arsenic is being attenuated on the aguifer materials and
that iron oxide is the primary mineral phase responsible for the attenuation of arsenic. Through the
adsorption mechanism, the unacceptably high levels of arsenic in the shallow aguifer will decrease over
time at a rate that depends on the net flux of water moving through the affected portions of the shallow
aguifer. The process of adsorption will effectively reduce the dissolved arsenic concentrations in
shallow ground water. Performance monitoring will be implemented to evaluate the effectiveness of the
attenuation and to ensure protection of human health and the environment. Performance monitoring will
include both ground water and surface water monitoring. The effects of the source control actions of
Alternative 3 along with the monitoring activities are expected to demonstrate within 30-40 years
that the rate of natural attenuation of arsenic in shallow ground water is sufficient to predict that
the ACL will never be exceeded at the established compliance points near Little Cottonwood Creek. EPA
expects the remaining areas of the shallow aguifer to achieve the ACL within a time frame of 100-150
years.

The last component of the selected remedy is soil removal and replacement with clean fill in off-facility
residential or commercial properties with soil lead concentrations that may present an unacceptable
health risk. This action will break the exposure pathway of direct contact with soils.

10.2 Compliance with Applicable or Relevant and Appropriate Requirements

The selected remedy will comply with all applicable or relevant and appropriate chemical reguirements
presented in Tables 16-18.

10.3 Cost Effectiveness

The selected remedy is cost effective because it has been determined to provide overall
effectiveness proportional to its costs, the net present worth value being $11.6 million. The estimated
costs of other alternatives are presented in Tables 19 and 20.

The costs of Alternatives 2,3, 6, and 7 are very similar. Comparing Alternatives 2 and 3, the additional
effectiveness and protectiveness associated with off-site disposal of principal threat wastes
(Alternative 3) was judged to warrant the additional $200,000 cost. The difference between Alternatives
3, 6, and 7 is the option for remediating the off-facility soils. The cost of a community monitoring and
health education program is greater than the excavation of contaminated soils and provides an
approximately egual level of protectiveness. Alternative 7 includes tilling of soils. This Alternative is
less costly than full soil removal but provides slightly less effectiveness in some areas of the Site.

The costs of Alternatives 3, 4, and 5 are guite different reflecting different approaches to ground water
remediation. EPA hydrogeologists carefully considered the potential benefits of extracting and treating
ground water as described in Alternative 4. The effectiveness of this option is limited by the
characteristics of the aguifer which allow very little water to be extracted. The addition of an
extraction system will not increase the rate of improvement in ground water guality over natural
attenuation processes despite the additional cost. Also considered was the amount of land which would be
reguired for dedication of numerous ground water extraction wells. This land would then be unavailable
for Site development. The additional cost of Alternative 4 does not result in effectiveness or benefit
for the Site. Alternative 4 also has greater problems with long term implementability, and greater
incompatibility with Site development. Alternative 5 includes in-situ ground water treatment in
additional to source controls. This alternative reguires high operation and maintenance costs without
appreciable increase in effectiveness or protectiveness.

Balancing costs with effectiveness, protectiveness, and Site development considerations, Alternative 3 is
judged by EPA to be the most cost effective.

10.4 Utilization of Permanent Solutions and Alternative Treatment Technologies (or Resource Recovery
Technologies) to the Maximum Extent Practicable

The selected remedy represents the maximum extent to which permanent solutions and treatment technologies
can be utilized in a cost effective manner for the Site. Neither extraction and treatment nor in situ
treatment of ground water were found to be more effective than natural attenuation at reducing arsenic
concentration in ground water. Yet both technologies are more costly. The institutional controls of the
selected remedy, while not permanent, will provide the reguired level of protection during the period of
natural attenuation of the ground water. The source control measures will provide a permanent solution by
consolidating the material in a engineered repository system preventing contact by water, and people.

-------
Of the alternatives that are protective of human health and the environment and comply with ARARs, EPA
believes that the selected remedy provides the best balance in terms of long term effectiveness and
permanence; reduction in toxicity, mobility, or volume achieved through treatment; short term
effectiveness; implementability; and cost. Overall protection of human health and the environment, long
term effectiveness, and cost were the most decisive criteria in selecting Alternative 3 as the remedy.

10.5 Preference for Treatment as a Principal Element

The selected remedy prescribes excavation and off-site disposal for the principal threat waste. On-site
treatment as a principal element was found not to be cost effective. However, the principal threat wastes
will be treated off-site before disposal. Therefore, the selected remedy satisfies the statutory
preference for treatment as a principal element to some degree.

Because the selected remedy will result in hazardous substances remaining on site, a review will be
conducted every five years after commencement of remedial action to ensure that the remedy continues to
provide adeguate protection of human health and the environment.

10.6 Conclusion

EPA's choice of Alternative 3 for remediation of the Site is protective of human health and the
environment and is in accordance with CERCLA and the National Contingency Plan.

-------
RECORD OF DECISION
DECLARATION STATEMENT

SITE NAME AND LOCATION

The Murray Smelter proposed National Priorities List Site is located in the city of Murray, Utah in Salt
Lake County.

STATEMENT OF BASIS AND PURPOSE

This decision document presents the selected remedial action for the Murray Smelter Site chosen in
accordance with the Comprehensive Environmental Response, Compensation, and Liability Act, as amended by
the Superfund Amendments and Reauthorization Act and the National Contingency Plan. This decision is
based on the administrative record file for the Site.

The State of Utah does not concur on the selected remedy.

ASSESSMENT OF THE SITE

A period of 77 years of lead smelting operations at this Site (1872-1949) resulted in impacts to the
soil, ground water, surface water and sediment. Lead and arsenic have been identified as the contaminants
of concern to human health. In addition to lead and arsenic, aluminum, cadmium, copper, mercury, nickel,
selenium, silver, thallium, and zinc have been identified as the contaminants of concern to ecological
receptors. Risk assessment performed at the Site in 1997 identified elevated risks to ecological
receptors as a result of exposure to lead in soils and sediments and selenium in plants. The risk
assessment also identified unacceptable risks to humans from ingestion of lead and arsenic in surface
soils and the potential ingestion of arsenic in shallow ground water. Although not currently used as a
drinking water source, the shallow aquifer at the Murray Smelter site meets EPA's and the State of Utah's
criteria for classification as a potential drinking water source, Class lib. An alternative drinking
water source is readily available in the deep principal aguifer and there is no near term future need for
the shallow ground water resource. Therefore, EPA believes that a relatively longer time frame for
achieving groundwater clean up levels is appropriate at this Site.

Actual or threatened release of hazardous substances from this Site, if not addressed by implementing the
response action selected in this Record of Decision, may present an imminent and substantial endangerment
to public health, welfare, or the environment.

DESCRIPTION OF THE REMEDY

The remedial action selected by this ROD is the second of three response actions EPA considers to be
necessary at the Murray Smelter Site. EPA expects that an additional time critical removal action will be
required to address the potential for release of hazardous substances and resulting health risks
associated with the potential structural failure of the two smelter stacks located on the Site.

The remedy selected for the Murray Smelter Site in this ROD consists of the following:

1. Contaminated ground water. Source control will be implemented by excavation and off site disposal of
the principal threat wastes at the Site, approximately 2000 cubic yards of residual undiluted arsenic
trioxide. This material is considered a principal threat due to its high mobility and its demonstrated
ability to act as a source of ground water contamination. In addition, direct contact with this
material may result in acute human health risks. Further source control will be implemented by
excavation of approximately 68,000 cubic yards of low level threat waste, diluted arsenic trioxide or
flue dust mixed with soil, fill, or debris from former smelter structures. This material will be
consolidated within a repository system constructed within the Site boundaries. The repository will be
designed as the base for a new access road through the Site which was planned by Murray City. The
access road is expected to be the catalyst for Site development. Monitored natural attenuation will
address the residual ground water contamination within and down gradient of these source areas.
Institutional controls in the form of a Murray City ordinance establishing an "overlay district" and
restrictive easements that run with the land both will prohibit the construction of new wells or use
of existing wells (except EP approved monitoring wells) within the on-facility area and the western
and eastern portions of the off-facility area.

2. Contaminated surface soils. On-facility surface soil containing levels of lead and arsenic exceeding
remediation levels will be covered. The barriers will provide protection by breaking the exposure
pathways associated with long term direct contact with these soils. Site development itself is
expected to result in additional protection of human health since land uses associated with

-------
unacceptable human health risks will end. Also, the development will result in the construction of
additional barriers (new buildings, roads, sidewalks parking lots, and landscaping) over remaining
surface soil and slag. Although no unacceptable risks associated with exposure to slag were identified
by EPA, the development of the Site will ensure no exposure to slag in the future. Institutional
controls in the form of a Murray City ordinance will establish an "overlay district" which includes
zoning to prevent residential and contact intensive industrial uses within the former smelter
operational areas and will reguire maintenance of the barriers and controls on excavated subsurface
material within this same area. Restrictive easements that run with the land will be established in
addition to the overlay district to prevent residential or contact intensive industrial uses.

3. Surface water. Little Cottonwood Creek which forms the northern boundary of the Site and to which
shallow around water discharges will be monitored to ensure continued protection during the ground
water natural attenuation process. Additional monitoring of the ecological study area of the Site will
be used to reduce the uncertainties identified in EPA's predictions of ecological risk.

The goals of the selected remedy are to restore ground water to the level of the ACL of 5.0 mg/L for
dissolved arsenic established to protect Little Cottonwood Creek at its beneficial use and to remediate
surface soils to levels protective of the reasonably anticipated future land use. The remedy incorporates
the construction of a new north-south access road through the Site which will encourage future
development of the Site and achieve Murray City's goal of more appropriate land use through Site
development. Based on information obtained during the site investigation and on a careful analysis of all
remedial alternatives, EPA believes that the selected remedy will achieve these goals. It may become
apparent during the monitored natural attenuation process for ground water that dissolved arsenic levels
have ceased to decline and are remaining constant at levels higher than the ACL over some portion of the
plume. If it is determined on the basis of system performance data that certain portions of the aguifer
cannot be restored to the alternate concentration limit, EPA will prepare a justification for a waiver of
the ground water ARAR based on technical impracticability of achieving further contaminant reduction.

STATUTORY DETERMINATIONS

The selected remedy is protective of human health and the environment, complies with Federal and State
reguirements that are legally applicable or relevant and appropriate to the remedial action and is cost
effective. This remedy utilizes permanent solutions however, the use of alternative treatment
technologies was found not to be practicable for this Site. The remedy will achieve significant reduction
in the mobility of the Site wastes through containment. The principal threat will be addressed by
excavation and off site disposal.

Because this remedy will result in hazardous substances remaining on site above health based levels, a
review will be conducted within five years after commencement of remedial action to ensure that the
remedy continues to provide adeguate protection of human health and the environment.

-------
                                FIGURES

-------
                                 TABLES


              TABIiE 5: LEAD AND ARSENIC IN SUBSURFACE SOIL

                                                  Arsenic
                                                                     Lead
Location

On-
f acility









Off-
facility








Area Number Depth
of stations Intervals

EU-1
EU-2
EU-4
EU-5

EU-6
EU-7
EU-8

EU-9
EU10
ISZ-1
ISZ-2
ISZ-3
ISZ-4

ISZ-5

ISZ-6
ISZ-7
ISZ-8

2
1
1
1

19
4


2
2
2
2
2
2

2

2
2
2



0-1
2-3
3-4
4-5

0-2
2-6
6-12
12-18



0-2
2-6
6-12
12-18






ft
ft
ft
ft

in
in
in
in



in
in
in
in



Average
(ppm)
448
272
158
25

1224
3005
2851

1240
107
69
73
214
68

81

47
185
132
Range Average
(ppm)
BDL-1500
130-340
BDL-620
BDL-56

BDL-48000
BDL-34000
64-7200

13-7500
45-140
17-230
27-170
53-610
6-150

44-120

BDL-70
86-480
BDL-450
(ppm)
8243
9480
1656
222

2259
3793
2751

6858
634
334
1089
520
496

443

599
2659
165
Range
(ppm)
50-16000
8200-10000
66-4800
61-600

57-22000
63-14000
520-9000

75-40000
430-1200
240-420
150-3200
87-1600
290-710

230-560

120-1000
550-7300
140-190
            All data from Hydrometrics 1995a.
            BDL = Below detection limit  (about 5 ppm).

Baseline Human Health Risk Assessment                                                     May 1997
Document Control Number 4500-090-AOAC                                                    Page 2-10
THIS DOCUMENT WAS PREPARED BY ROY F. WESTON, INC. EXPRESSLY FOR EPA.   IT  SHALL  NOT  BE  RELEASED OR
DISCLOSED IN WHOLE OR IN PART WITHOUT THE EXPRESS WRITTEN PERMISSION OF EPA.

-------
                   TABIiE 6
SUMMARY OF CHEMICAL ANALYSIS FOR KEY ANALYTES
        SHALLOW AQUIFER GROUND WATER
               MURRAY SMELTER
WELL # OF TDS
SAMPLES RANGE
TOTAL ARSENIC
detects
JMM-01
JMM-02
JMM-06
JMM-
07B
JMM-08
MS-GW-
1
MS-GW-
2
MW-100
MW-101
MW-102
MW-103
MW-104
MW-105
4
4
4
4

9
9

9

9
8
9
9
9
9
787-1108
777-890
1325-1489
121-1367

549-957
868-1126

981-1270

852-976
484-651
623-3409
1032-1110
605-1439
726-941
4
4
0
4

9
9

9

1
8
9
9
8
9
0
0

0

0
0

2


0
0


0
range
.366-0.746
.452-1.008

.013-0.019

.016-0.078
.487-30.14

.87-6.539

BDL-0.002
.006-0.047
.013-0.021
0.098-0.27
BDL-0.012
.013-0.042
mean
0.502
0.652

0.015

0.039
10.98

4.10

0.002
0.014
0.017
0.21
0.009
0.022
TOTAL LEAD
detects range
2
2
2
0

7
4

4

6
6
1
4
2
6
0.064-0
0.003-0
0.002-0


0.002-0
BDL-0.

BDL-0.

BDL-0.
BDL-0.
BDL-0.
BDL-0.
BDL-0.
BDL-0.
.093
.013
.008
0

.007
003

005

035
301
001
003
01
079
mean
0.079
0.008
0.005


0.004
0.005

0.002

0.01
0.062
0.001
0.002
0.002
0.02
d<
0
0
0


0
8

8

0
6
3
0
6
8
                                                                         TOTAL SELENIUM

                                                                        detects   range
                                                                                0.015-0.192  0.065

                                                                                0.036-0.056  0.046


                                                                                 BDL-0.016    0.01

                                                                                 BDL-0.007   0.004

                                                                                 BDL-0.018   0.012
                                                                                0.016-0.053  0.037

-------
                                TABIiE 6

             SUMMARY OF CHEMICAL ANALYSIS FOR KEY ANALYTES
                     SHALLOW AQUIFER GROUND WATER
                            MURRAY SMELTER
MW-106
MW-107
MW-108
MW-109
MM-110
MW-111
MW-112
MW-113
MW-114
UTBN-1
WELL 1
WELL 2
WELL 3
9
9
9
7
9
9
9
2
2
10
8
9
9
1491-1895
2126-2784
995-1264
1082-1345
1329-1530
658-1578
602-1124
1524-1544
490-506
759-1265
535-801
1434-1782
843-1309
9
8
6
7
9
9
9
2
2
10
8
9
9
23.85-31.06



1
2
0
0
0
0

1
0
BDL-0.
BDL-0.
0014-
.689-2
.903-4
.052-0
.015-0
.015-0
.116-0
0.14-0
.439-1
.134-0
019
02
0022
.388
.535
.134
.021
.021
.27
.316
.974
.236
26.74
0.014
0.006
0.018
2.10
3.60
0.104
0.019
0.018
0.176
0.245
1.68
0.173
6
1
3
4
0
7
7
0
0
8
5
3
7
BDL-0. 079
BDL-0. 001
BDL-0. 026
BDL-0. 012

0013-0.212
0.027-0.084
0
0
0.05-0.101
BDL-0. 086
BDL-0, 008
0.081-0.214
0.02
0.001
0.006
0.003

0.107
0.039


0.069
0.024
0.006
0.139
8
8
8
0
8
8
4


9
0
1
8
                                                                                             0.07-0.137   0.104
                                                                                            0.026-0.186    0.12
                                                                                            0.041-0.095   0.076

                                                                                            0.104-0.141  0.139
                                                                                            0075-0.166     0.115
                                                                                             BDL-0.059    0.016
                                                                                             0.011-0.063  0.036

                                                                                              BDL-0.006   0.003
                                                                                             0.011-0.079  0.028
NOTES:
All values are reported in units of mg/L
Values of one half  (the detection limit were substituted  for below  detection limit data in calculation of mean values.

-------
              TABIiE 7 Exposure Point Concentrations for Surface Water
                                 Part A: Low Flow
Chemical
Upgradient
  (n=2)
            Total a
                      Dissolved b
Onsite
 (n=2)
                                      Total
Downgrading
   (n=2)
                              Dissolved   Total
                         Dissolved
Depression
   (n=2)

   Total
Aluminum
Arsenic
Cadmium
Copper
Lead
Selenium
Zinc
0.193
[0.0025]
[0.00025]
[0.005]
0.008
[0.0015]
0.021
[0.05]
[0.0025]
[0.00025]
[0.005]
0.003
[0.0015]
[0.01]
0.209
0.048
[0.00025]
[0.005]
0.004
[0.0015]
0.035
[0.05]
0.044
0.0012
[0.005]
[0.001]
[0.0015]
[0.01]
0.534
0.054
0.0009
0.012
0.009
[0.0015]
0.079
[0.05]
0.065
0.0012
0.01
[0.001]
[0.0015]
0.028
0.110
0.048
0.0041
0.017
0.045
0.01
0.149
Chemical
                            Part B: High Flow
Upgradient
  (n=2)
            Total a
                      Dissolved b
Onsite
 (n=2)
                                      Total
Downgrading
   (n=2)
                              Dissolved   Total
                         Dissolved
 Depression
   (n=2)

   Total
Aluminum
Arsenic
Cadmium
Copper
Lead
Selenium
Zinc
0.644
[0.0025]
0.0007
0.017
0.013
[0.025]
0.113
[0.05]
[0.0025]
[0.0005]
[0.0025]
[0.001]
[0.0025]
0.049
0.748
0.010
0.001
0.017
0.021
[0.0025]
0.117
[0.05]
0.01
0.001
0.006
[0.001]
[0.0025]
0.121
1.053
0.011
0.001
0.03
0.032
[0.0025]
0.135
[0.05]
0.01
[0.00025]
[0.0025]
[0.001]
[0.0025]
0.072
0.185
0.672
0.003
0.03
0.087
0.049
0.492
   All values are expressed in units of mg/L and represent maximum values due to limited samples as
   described in the text.
   [ ]  Values in brackets represent ^> guantitation  (reporting) limit.
   a Total concentrations were used to evaluate risk to avian receptors.
   b Dissolved concentrations were used to evaluate risk to fish.
   See Appendix C for data and summary statistics.

Draft Final Ecological Risk Assessment                                 September 1997
Document Control Num 4500-090-AOKP                                       Page 3-2
THIS DOCUMENT WAS PREPARED BY ROY F. WESTON, INC. EXPRESSLY FOR EPA.  IT SHALL NOT BE RELEASED OR
DISCLOSED IN WHOLE OR IN PART WITHOUT THE EXPRESS WRITTEN PERMISSION OF EPA.

-------
     TABIiE 8 Exposure Point Concentrations for Sediment
Chemical
Upgradient
  (n=10)
Aluminum
Arsenic
Cadmium
Copper
Lead
Mercury
Nickel
Selenium
Silver
Thallium
Zinc
5523
29
0.63
62
302
0.1
37
0.55
2.3
[0.5]
526
Onsite
(n=10)

 6465
  70
  3.1
  188
 1699
 0.33
  63
 0.78
  5.5
 [0.5]
 2389
Downgradi ent
   (n=10)

    5938
     32
     1.4
     409
     356
    0.18
     116
    0.48
     3.6
    [0.5]
     694
Development
  (n=10)

   11893
    492
     51
    1628
    9058
    0.50
     40
     58
     19
     32
   58600
    All values reported in units mg/kg dry weight. EPCs are the minimum of the UCL95  or maximum detected
    value as described in text.
    [ ]  Values in brackets represent ^> guantitation  (reporting) limit.
    See Appendix C for data and summary statistics.

Draft Final Ecological Risk Assessment                                 September  1997
Document Control Num 4500-090-AOKP                                       Page 3-2
THIS DOCUMENT WAS PREPARED BY ROY F. WESTON, INC. EXPRESSLY FOR EPA.  IT SHALL NOT BE RELEASED  OR
DISCLOSED IN WHOLE OR IN PART WITHOUT THE EXPRESS WRITTEN PERMISSION OF EPA.

-------
TABIiE 9 Exposure Point Concentrations for Riparian Soil
Chemical

Aluminum
Arsenic
Cadmium
Copper
Lead
Mercury
Nickel
Selenium
Silver
Thallium
Zinc
Upgradient
  (n=7)

  52611
    70
   1.8
   258
   771
  0.77
    67
   0.7
   6.6
  [0.5]
   685
Onsite
 (n=8)

 10780
  129
   6
  366
  3100
    1
   47
   11
  8.6
  1.1
  2332
Downgradient
   (n=5)

    9800
     55
      3
     193
     659
     0.67
     40
     0.9
     6.6
    [0.5]
     681
All values reported in units of mg/kg dry weight. EPCs are  the minimum of  the  UCL95  or  maximum detected value
as described in the text.
[ ]  Values in brackets represent ^> guantitation  (reporting)  limit.
See Appendix C for data and summary statistics.

-------
                 TABLE 10  Exposure Point Concentrations for
                          Benthic Macroinvertebrates
Chemical   Upgradient (n=6)   Onsite  (n=4)   Downgradient  (n=2)   Depression  (n=3)
Aluminum
Arsenic
Cadmium
Copper
Lead
Selenium
Silver
Thallium
Zinc
1100
12
3
60
58
17
0.56
[0.5]
373
                                 1169
                                  288
                                   5
                                  73
                                  175
                                  11
                                  1.3
                                 [0.5]
                                  595
865
 99
  3
 64
 50
 12
0.49
[0.5]
 425
 864
 133
  15
 122
 440
  58
 2.6
 9.3
3160
   All value reported in units of mg/kg dry weight. EPCs are the minimum of the UCL95 or maximum detected
   value as described in the text.
   See Appendix C for data and summary statistics.

Draft Final Ecological Risk Assessment                                 September 1997
Document Control Num 4500-090-AOKP                                       Page 3-2
THIS DOCUMENT WAS PREPARED BY ROY F. WESTON, INC. EXPRESSLY FOR EPA.  IT SHALL NOT BE RELEASED OR
DISCLOSED IN WHOLE OR IN PART WITHOUT THE EXPRESS WRITTEN PERMISSION OF EPA.

-------
                                              TABIiE 11
                         SUMMARY OF ARSENIC  CONCENTRATIONS IN SURFACE WATER
                              MEASURED  IN  QUARTERLY MONITORING EVENTS
                                        MURRAY  SMELTER SITE
SAMPLE
DATE
UPSTREAM
AVERAGE
UPSTREAM
AVERAGE
ONSITE   ON-SITE
AVERAGE  AVERAGE
DOWNSTREAM
  AVERAGE
DOWNSTREAM
  AVERAGE
                                                                             WETLANDS   WETLANDS
         DISSOLVED
                     TOTAL
                              DISSOLVED   TOTAL
                                                    DISSOLVED
7/22/96
12/6/96
1/14/97
4/11/97
7/15/97
10/8/97
0.007
<0.005
<0.005
<0.005
0.007
0.009
0.007
<0.005
<0.005
<0.005
0.008
0.008
0.167
0.173
0.288
0.176
0.051
0.123
0.146
0.201
0.299
0.161
0.046
0.11
0.129
0.164
0.2
0.181
0.042
0.053
                                                                   TOTAL

                                                                   0.107
                                                                   0.202
                                                                   0.255
                                                                   0.184
                                                                   0.043
                                                                   0.061
                                                                    DISSOLVED    TOTAL

                                                                      0.26       0.266
                                                                      0.201
                                                                      0.146
                                                                      0.232
                                                                      0.175
All results are reported in units  of milligrams  per liter.
Where no result is reported, no  sample was  collected on that date.

-------
                                                                          TABIiE 16
                                                                   CHEMICAL SPECIFIC ARARS
                 Requirement

Utah Primary Drinking Water Standards




National Primary Drinking Water Standards



National Primary Drinking Water Standards
   Citation

UAC R309-103-2
40 CFR 141 11
40 CFR 141
Definitions and General Requirements of Utah Water
Quality Act
UAC R317-1
Administrative Rules for Groundwater Quality Protection   UAC R317-6-6 4C and R317-6-
                                                          64D

                                                          UAC R317-6-2
            Description

Establishes maximum contaminant levels
of 0.015 mg/L for lead and 0.05mg/l for
arsenic as primary drinking water
standards

Establishes the maximum contaminant
level for arsenic of 0.05 mg/L
Establishes a lead action level of 0.015
mg/L. Regulations establish a treatment
technique triggered by exceedance of the
action level in more than 10 percent of tap
water samples collected during any
monitoring period.

Provides definitions and general
requirements for water quality in the State
of Utah

Establishes requirements for issuance of a
groundwater discharge permit at an
existing facility. Permit limits may be
either groundwater quality standards or
alternate concentration limits.
Groundwater quality standard for arsenic
is 0.05 mg/1, for lead is 0.015 mg/1.
Alternate concentration limits are
established on a site specific basis. The
Alternate Concentration Limit for the
Murray Smelter Site is 5.0 mg/L.
            Notes
relevant and appropriate for
groundwater at the Murray Smelter
Site
relevant and appropriate
groundwater at the Murray Smelter
Site

relevant and appropriate for
groundwater at the Murray Smelter
Site
Applicable to ground water and surface
water at the Murray Smelter Site
                                                                           Substantive requirements are relevant
                                                                           and appropriate for groundwater at
                                                                           Murray Smelter. Note that the
                                                                           groundwater quality standard need not
                                                                           be met if it is demonstrated that an
                                                                           alternate concentration limit (ACL)  is
                                                                           protective. At the Murray Smelter Site
                                                                           the ACL is the relevant and appropriate
                                                                           requirement for on site groundwater in
                                                                           the shallow aquifer.

-------
                                              TABIiE 16
                                       CHEMICAL SPECIFIC ARARS
Standards of Quality for Waters of the State
UAC R317-2-6, R317-2-7,
R312-2-13, and
R317-2-14
Establishes use designations of Class 2B,
Class 3A, and Class 4 for the segment of
Little Cottonwood Creek which borders
the Murray Smelter site. Establishes
water quality standards applicable to each
class. Water quality standards for trivalent
arsenic are 190 ug/1 (4 day average) and
360 ug/1  (1 hour average) for Class 3A
Water quality standard for dissolved
Arsenic is 100 ug/1 for Class 4 Water
quality standards for lead are 3 2 ug/1(4
day average)  and 82 ug/l(l hour average)
for Class 3A and 100 ug/1 for Class 4
Applicable to surface water of Little
Cottonwood Creek

-------
                                                               TABIiE 17
                                                         ACTION SPECIFIC ARARS
Emission Standards
Fugitive Dust Emission Standards
Ground Water Protection Standards for Owners and
Operators of Hazardous Waste Treatment, Storage, and
Disposal Facilities
General Facility Standards:
Construction Quality Assurance Program
                                                           UAC R307-1-4
    UAC R307-12
40 CFR Part 264.97
   UAC R315-8-6
40 CFR Part 264 99

  40 CFR 264 19
 Establishes air guality standards for visible
 emissions,  PM10,  and internal combustion
 engines

 Establishes air guality standards for
fugitive dust emissions

 Establishes general ground water
 monitoring reguirements for treatment
 storage and disposal facilities

 Establishes reguirements for compliance
 monitoring program

 Established reguirement for a construction
 guality assurance program to ensure that
 constructed units meet or exceed all
 design criteria and specifications
                                                                             Applicable to emissions generated
                                                                             during remedial activities
Applicable to fugitive dust emissions
generated during remedial activities

Relevant and appropriate to ground
water at Murray Smelter Site
underlying any on site waste
management units constructed as part
of the remedial action
Relevant and appropriate to
construction of surface impoundment.
waste pile, and land fill units
constructed as part of the remedial
action
General Facility Standards:
Location Standards for Hazardous Waste Facilities
  UAC R315-8-2.9

    40 CFR 264.18
Standards for Control of Installations, State Adoption of   UAC P307-1-3
National Ambient Air Quality Standards (NAAQS)
Off Site Management of CERCLA Wastes
    40 CFR 300.440
                                                              UAC R315-5
 Establishes site characteristics which are
 unsuitable for location of hazardous waste
 management units.

 Establishes NAAQS  as reguirements for
 air guality. NAAQS for PM10 is 50
 ug/m 3 annual arithmetic mean, and 150
 ug/m 3 24 hour maximum.
 NAAQS for lead is  1.5 ug/m 3 maximum
 guarterly average,

 Establishes reguirements for off site
 management of CERCLA wastes

 Establishes hazardous waste generator
Portions are relevant and appropriate to
alternatives which include
consolidation of wastes on site

Relevant and appropriate to air
emissions resulting from remedial
activities at Murray Smelter
Applicable to alternatives that involve
off site management of hazardous
waste
                                                      40 CFR 262.10 through 262-44   reguirements
Well Drilling Standards
      UAC R655-4
 Establishes standards for drilling and
 abandonment of wells
Applicable to installation or
abandonment of monitoring wells

-------
Utah Pollutant Discharge Elimination System
Requirements
ACTION SPECIFIC ARARS

     UAC R317-8
Closure and Post-Closure:
Post-closure Care and the Use of Property
Closure And Post Closure:
Post-closure notices
Closure and Post Closure:
Post-closure notices
                                                         40 CFR 264.117
                                                         40 CFR 264.118
                                                         40 CFR 204.119
Establishes general requirements,
definitions, and standards for point source
discharges of pollutants into surface water
bodies in Utah and establishes pre-
treatment requirements for discharge to a
publicly owned treatments works

Establishes minimum requirements for
monitoring, reporting, and maintenance of
closed hazardous waste management units

Establishes requirement for written plan
identifying activities that will be carried on
after closure of each disposal unit

Establishes requirement to record
certification of closure via a notation on
the property deed to the facility and
notification that the land has been used to
manage hazardous waste
Applicable to point source discharges to
Little Cottonwood Creek from the
Murray Street site
                                                                            Relevant and appropriate to
                                                                            consolidation units constructed as
                                                                            of the remedial action

                                                                            Portions are relevant and appropriate to
                                                                            consolidation units constructed as part
                                                                            of the remedial action

                                                                            Portions are relevant and appropriate to
                                                                            consolidation units constructed as part
                                                                            of the remedial action

-------
                                             TABIiE 18
                                     LOCATION SPECIFIC ARARS
Migratory Bird Treaty Act
16 USCS 703
Establishes that is unlawful to take or
possess any migratory nongame bird or
any part of such migratory nongame bird
Applicable to migratory birds at the
Murray Smelter site

-------
APPENDIX A

RESPONSIVENESS SUMMARY

PROPOSED PLAN FOR
MURRAY SMELTER PROPOSED NPL SITE

PART I:

COMMENTS RECEIVED FROM THE UTAH DEPARTMENT OF ENVIRONMENTAL QUALITY (UDEQ)

UDEQ stated concerns that selenium has been detected in the shallow ground water at the Site in
concentrations which exceed drinking water standards for that chemical.

EPA Response: The over-riding environmental concern associated with shallow ground water within the
on-facility boundaries is arsenic which has been detected at levels 100-1000 times drinking water MCL. In
comparison, selenium has been detected at various locations within the on-facility boundaries at levels twice
the drinking water MCL. Unlike arsenic, the selenium in shallow ground water has not affected the guality of
Little Cottonwood Creek. EPA's selected remedy includes continued monitoring of selenium in shallow ground
water and institutional controls which will prevent exposure to selenium by prohibiting the installation of
ground water wells except for the purpose of monitoring. The selected remedy is thus protective.

UDEQ also expressed concern about the arsenic loading of Little Cottonwood Creek as a result of a point
discharge from the 48 inch reinforced concrete pipe culvert which runs along State Street.

EPA Response: The selected remedy requires control of the Site related source(s) of this arsenic discharge
and further requires compliance with surface water quality standards for Little Cottonwood Creek. The details
of the source control activities will be developed as part of remedial design.

UDEQ provided an evaluation of responses to comments they submitted on the draft Feasibility Study. The
responses were prepared by Asarco. On the basis of Asarco's responses to UDEQ's and EPA's comments on the
document, EPA approved the Final Feasibility Study. EPA notes that UDEQ was provided Asarco's responses on
August 27, 1997.

UDEQ Comment 1: UDEQ requests numeric clean up levels and confirmatory sampling.

EPA Response: EPA established remediation levels in Section 8.3 of the ROD. The details of confirmatory
sampling will be developed as part of remedial design.

UDEQ Comment 2: UDEQ is concerned about the remedy's ability to comply with the ground water MCL for arsenic
given the long time (>150 years) for achieving the MCL predicted in the Feasibility Study.

EPA Response: In the ROD EPA provides the rationale for why the conditions at Murray Smelter meet those
established in CERCLA Section 122 for the establishment of an Alternate Concentration Limit in lieu of the
MCL for arsenic. The evaluation of how the selected alternative will meet this ACL within reasonable time
frame given the Site specific circumstances is contained in Section 9.1 of the ROD. EPA agrees with the
statements in the final Feasibility Study that the MCL will ultimately be met. The mechanisms of natural
attenuation will continue in perpetuity such that ground water quality will continue to improve resulting in
the achievement of restoration albeit in a very long time.

Also in Comment 2, UDEQ requests more specific information about how ACLs will be established at the Site.

EPA has included the development of the ACL for arsenic as Appendix C to the ROD.

UDEQ Comment 3: UDEQ objects to Asarco's statements which suggest that State ARARs were not identified in a
timely manner.

EPA Response: EPA notes that UDEQ has never responded to EPA's September, 1996 formal request for an ARARs
analysis from the State. While it is accurate to state that many discussions have occurred between the State
and EPA on the identification of State ARARs, UDEQ has only provided a table with no indication of whether
that table was to be considered official or final identification of State ARARs for this Site.

UDEQ also requested justification for why chemical specific RCRA ground water maximum concentration levels
were not identified as ARAR.

EPA Response: EPA did not identify these standards because they are not applicable (Murray Smelter is not a

-------
treatment storage or disposal facility) and are not relevant and appropriate given the Site circumstances are
appropriate for establishing an ACL.

EPA included the following ARARs in the ROD in response to UDEQ' comments:

A. UAC R315-8-6 is identified as relevant and appropriate.

B. Utah's ground water protection rule is identified as relevant and appropriate.

C. EPA's off-site rule is identified as applicable.

D. UAC R315-5 is identified as applicable.

E. UAC R3158-14 is identified as relevant and appropriate.

F. UAC R311-211-2 is identified as applicable.

G. UAC R317-1 is identified as applicable.

UDEQ Comment 4: UDEQ is concerned about the lack of detail regarding the cover design for the on-Site
repository system.

The reguirements for the cover are identified in the ROD. The further development of the details of the cover
is a remedial design activity.

UDEQ provided an evaluation of how well Asarco responded to UDEQ's comments on the draft Feasibility Study.
This evaluation is noted by EPA. EPA considers the responses provided by Asarco to be adeguate. It was on the
basis of Asarco's responses to these comments as well as EPA's comments that EPA approved the Feasibility
Study. We assume that this further evaluation by UDEQ is provided for the record and as such will be included
as part of the Administrative Record for the Site.

PART II

COMMENTS RECEIVED FROM ASARCO

Asarco commented that the monitoring reguirements included in the ROD to support efforts to reduce
uncertainties in the ecological risk assessment may not be reguired if the wetlands area of the Site are
filled during Site development. Asarco also suggests that there may be other options to monitoring which will
reduce the uncertainties in the ecological risk assessment.

EPA Response: EPA agrees with the comment and has included language in the ROD indicating that in the event
the wetlands are filled, the associated exposure pathways will be broken. The ROD also includes the
reguirement that if the wetlands remain, monitoring will be reguired. The majority of the ecological risk at
the Site is associated with the wetlands. As development plans become more clear, monitoring will be
incorporated into the remedial design or deleted as appropriate. Currently, there is not enough information
to assess how the planned Site development will affect the wetlands.

Asarco also commented that the Proposed Plan was not clear in describing whether the proposed cover for slag
is to be an interim or permanent cover. Asarco further guestioned the basis for reguiring a cover for slag.
Asarco also enclosed the attached memorandum supporting their view that a cover for slag is not reguired.

EPA Response: EPA agrees with Asarco's comments. Language has been added to the ROD to clarify that there is
no need to cover the slag as part of the remedy for the Site. The ROD also makes it clear that EPA expects
the slag will be covered in the near future as part of Site development.

-------

To:       Donald A. Robbins

From:     Rosalind A. Schoof, Ph.D

Date:     October 23, 1997

Subject:  Weathering of Slag at the Murray Smelter Site

Recently a guestion has arisen regarding potential future human health risks for workers who might contact
particles released from slag at the Murray Smelter Superfund Site in Murray, Utah due to weathering
processes, prior to completion of final remediation activities at the site within the next 5 to 10 years
(Lavelle 1997),  This technical memorandum addresses the possibility that such risks might differ from those
previously assessed by EPA in the baseline human health risk assessment for the site (Weston 1997).

The baseline risk assessment noted that there are extensive areas of the site where slag is exposed at the
surface, but concluded that on-facility workers were unlikely to spend much time in areas of exposed slag.
The only human receptors for whom slag exposure was determined to be of potential concern were teenagers who
might spend time near the slag piles up to 50 times per year for 7 years. These teenagers were assumed to
ingest 100 mg of slag at each visit. The fraction of lead and arsenic assumed to be absorbed from the slag
was based on studies conducted using fine particles collected from the slag piles. The risk assessment
concluded that these teenagers were unlikely to be at risk of adverse health effects from lead in the slag,
and that incremental cancer risks associated with arsenic in the slag were within EPA' s acceptable risk range
(i.e., between 1 x 10 -6 and 1 x 10 -4).

There are a number of reasons why continued weathering of the slag piles during an interim period prior to
implementation of remedial actions is not likely to pose unacceptable human health risks. The purpose of the
baseline risk assessment was to assess potential risks to workers and residents if no remedial actions were
ever taken at the site. Conseguently, risks from an interim period prior to implementing remedial action
cannot exceed those evaluated and judged to be low in the baseline risk assessment unless there is some
marked change in the nature of exposures to slag, or in the nature of releases of metals from slag that was
not foreseen in the baseline risk assessment. The nature of exposures to slag is not expected to change for
on-facility workers or for residents. Similarly, the nature of releases of metals from slag is also unlikely
to change for reasons described below,

One mechanism for release of metals is by weathering and breakdown of chunks of slag into fine particles. In
many area of the site, slag that has been at the surface for 50 to 100 years doesn't show any marked signs of
weathering. In areas where weathering may have occurred, the risk assessment already accounted for this
process by assuming that the ingested slag was in fine particles that might adhere to hands prior to
ingestion. Additionally, the bioavailability of fine particles collected from the slag piles was tested, and
the results were used in the risk assessment. Conseguently, the baseline risk assessment is already based on
weathered material, and continued weathering is not likely to lead to increased risks.  Indeed, as discussed
below, further weathering may serve to reduce the risks by changing the arsenic and lead to less bioavailable
forms.

To assess the risk posed by the exposed slag at the Murray site, the risk assessment used the bioavailability
estimate for the composite Murray slag sample tested in the EPA swine study (Weston 1997).  Greater than 70
percent of the lead mass in this sample (as determined by electron microprobe analysis) was associated with
the highly bioavailable lead form, lead oxide (Figure 1).  However, as the lead oxide weathers, it will form
secondary weathering products including lead phosphate, iron-lead oxides, and iron-lead sulfates  (Davis et
al. 1993). Because these weathering products will have lower solubility than the lead oxide mineral upon
which the risk assessment was based, the risk posed by the exposed slag will diminish as the lead oxide
weathers and forms these secondary minerals.

While no information was presented in the risk assessment describing the arsenic mineralogy of the Murray
slag sample, Dr. John Drexler of the University of Colorado has indicated that a large fraction of the
arsenic was associated with the arsenic oxide phase (Drexler 1997).  Assuming this is the case, then arsenic
bioavailability would also diminish with time because the arsenic bound in soluble arsenic oxide will
eventually repartition into iron oxide phases (PTI 1996),  which have a lower bioavailability than arsenic
oxide.

In conclusion, it appears that the evaluation of potential human health risks from exposure to slag in the
baseline risk assessment for the Murray Smelter Superfund site was sufficiently comprehensive to ensure that
no unforeseen risks will occur during an interim period prior to completion of remedial actions at the site.

-------
References

Davis, A, J.W. Drexler, M V Ruby, A. Nicholson 1993. Micromineralogy of mine wastes in relation to lead
bioavailability. Butte, Montana. Environ. Sci. and Tech.  (27) 1415-25.

Drexler, J.W. 1997. Personal communication between J.W. Drexler, University of Colorado and Christopher
Sellstone, PTI Environmental Services.

Lavelle, B. 1997. Personal communication between B. Lavelle, U.S. Environmental Protection Agency, Region
VIII, Denver, CO and R. Schoof, PTI Environmental Services, Bellevue, WA.

PTI. 1996. Appendix 0 - Electron Microprobe Analysis Results. Included in: Draft Site Characterization Report
for the Former Murray Smelter Site.

Weston. 1997. Baseline Human Health Risk Assessment for the Murray Smelter Superfund Site. Prepared by Roy F.
Weston. Inc. Lakewood, CO. Prepared for U.S. Environmental Protection Agency, Region VIII, Denver, CO.

-------
                               APPENDIX B

-------
TABIiE 2-1. SUMMARY OF PARAMETERS FOR ARSENIC EVALUATION

Parameter                                        Resident   NCI-Worker   Cl-Worker

Soil/dust intake rate as child  (ing/day)             200
Soil/dust intake rate as adult  (ing/day)             100           50         240
Fraction of total that is dust                      0.5          0.5           0
Relative bioavailability of arsenic  in  soil/dust  0.26         0.26        0.26
Body weight as child  (kg)                            15           -
Body weight as adult  (kg)                            70           70          70
Exposure frequency  (days/yr)                        350          250         250
Exposure duration as child  (yrs)                      6           -
Exposure duration as adult  (yrs)                     24           25          25
Averaging time for cancer  (yrs)                      70           70          70
Oral slope factor                                   1.5          1.5         1.5

-------
   PRG for Arsenic in Soil a (ppm)
Population     1E-04    IE-OS    1E-06

Residential     290       29      2.9
NCI-Worker     1,200     120       12
Cl-Worker       180       18      1.8

   a All values expressed to two significant figures

2.4  Uncertainty in the PRG Values

It is very important to recognize that guantitative risk calculations and PRG derivations are both inherently
uncertain due to lack of knowledge regarding a number of key parameters. These uncertainties (discussed in
Section 6 of the Baseline Human Health Risk Assessment)  include lack of knowledge regarding actual human
exposure rates to soil, dust and slag, uncertainty in the extent of absorption (bioavailability) of arsenic
from soil and slag, and uncertainty in the exposure levels of arsenic that are actually likely to cause
significant adverse effects.

In most cases, conservative approaches are used to fill these knowledge gaps. Therefore, the PRG values
calculated above are more likely to be low than high. Because of this, the PRG values should not be viewed as
concentrations which form a clear boundary between acceptable and unacceptable soil levels. Rather, values
below the PRG should be viewed as very likely to be protective, with a gradually decreasing probability of
protection as soil values exceed the PRG.

3.0  EVALUATION OF LEAD

3.1  PRGs for Residents

Basic Approach

The USEPA has developed an Integrated Exposure, Uptake and Biokinetic (IEUBK) model for evaluating the risks
of lead to children (age 0-7) exposed under residential circumstances. This model was used to calculate the
concentration of lead in soil which would correspond to a 5% probability that a child living at a location
with that concentration in soil would have a blood lead value greater than 10 ug/dL. All input assumptions to
the model were those recommended by EPA as defaults  (EPA 1994a),  except for 1) the Geometric Standard
Deviation (GSD),  2) the ratio of lead in dust compared to soil, 3)  the relative bioavailability of lead in
soil and dust, and 4)  the amount of lead ingested in the diet. The basis for each of these site-specific
values is detailed in the Baseline Human Health Risk Assessment (WESTON, 1997) and is summarized briefly
below.



GSD

A study of blood lead levels in Sandy, Utah, indicate that variability between different children can be
described by an individual geometric standard deviation of 1.4 (EPA 1995b). Because the population of Sandy
is believed to be generally similar to the population of Murray,  this value  (a GSD of 1.4) is considered to
be more relevant and a better approximation of the true site-specific value than the default value (1.6), so
the site-specific value is used in place of the default value.

Soil/Dust Relationship

The normal assumption used in the IEUBK model is that the concentration of lead in indoor dust is 70% of that
in outdoor soil (EPA 1994a). However, this assumption has been found to overestimate lead concentrations in
dust at some mining-related sites. As described in the Baseline Human Health Risk Assessment (WESTON 1997),
paired soil-dust samples were collected from 22 off-facility locations,  and these data were used to analyze
the average relationship between levels of lead in soil and in dust. The slope of the best-fit straight line
through the data calculated by linear regression is 0.32 ppm per ppm. However, as noted above,  analysis of
soil/dust relationships by linear regression is complicated by the problem of measurement error, which tends
to lead to an underestimate of slope and an overestimate of intercept. On this basis, the best-fit slope was
rounded upwards to 0.35 ppm per ppm, and this value was used in place of the default of 0.70 in the IEUBK
model.

RBA

The IEUBK model employs a default relative bioavailability factor of 60% for lead absorption from soil and
dust (compared to that for water or food) (EPA 1994a).  However, there are several studies which provide

-------
evidence that lead in soil from mining/smelting sites may be absorbed less-extensively than this default. The
EPA has conducted a study of the bioavailability of lead in a composite soil sample from the Murray Smelter
site  (EPA 1996a).  Preliminary results are summarized below:

                       RBA in Site Soil        Value

                   Plausible Range           0.67-0.84
                   Preferred Range           0.67-0.75
                   Suggested Point Estimate    0.71

As seen, although there is uncertainty in the estimate, the relative bioavailability for soil is probably
about 70%, slightly higher than the default value used in the IEUBK model. Based on this value,  and assuming
that lead in food and water is about 50% absorbed by children (EPA 1990),  this RBA value corresponds to an
absolute bioavailability of 35% (0.35).

-------
Results

Based on the parameters summarized in Table 3-1, the levels of lead in soil that will be protective for adult
on-site workers are:

                Population          PRG for Lead (ppm)

                NCI-Workers               5600
                Cl-Workers                 930

3.3  Uncertainty in the PRG values

As discussed above, it is important to stress that there is substantial uncertainty in the soil lead PRG
values calculated for both residential children and for on-site workers. These uncertainties are related to
lack of knowledge regarding true soil and dust intake rates, lack of certainty in the true absorption
fraction for lead, and uncertainty in the true level of health risk posed by low level lead exposures to
children and fetuses. In addition, there is uncertainty associated with the accuracy of the mathematical
models used to make the calculations  (the IEUBK model and the Bowers model).  These "model uncertainties"
arise because human exposure, absorption, distribution and clearance of lead are very complicated and dynamic
processes, and any mathematical model which seeks to guantify the processes must always be an
over-simplification. In addition, many of the pharmacokinetic parameters relating to lead metabolism in
humans are difficult to study and measure, so there is uncertainty whether the values used in the models are
accurate. Because of these uncertainties, the PRG values calculated for lead should not be thought of as a
clear boundary between acceptable and unacceptable soil levels. Rather, values below the PRG should be viewed
as very likely to be protective, with a gradually decreasing probability of protection as values exceed the
PRG.

-------
4.0 REFERENCES

Bowers TS, Beck BD, Karam HS. 1994. Assessing the Relationship Between Environmental Lead Concentrations and
Adult Blood Lead Levels. Risk Analysis 14:183-189.

CDC. 1991. Preventing Lead Poisoning in Young Children. US Department of Health and Human Services, Centers
for Disease Control.

CEPA. 1992. California Environmental Protection Agency, Department of Toxic Substances Control. Supplemental
Guidance for Human Health Multimedia Risk Assessment of Hazardous Waste Sites and Permitted Facilities,
Sacramento, California.

EPA. 1990. U.S. Environmental Protection Agency, Technical Support Document on Lead. ECAO-CIN-757.
Cincinnati, OH: EPA Office of Environmental Criteria and Assessment Office. September.

EPA. 1991a. U.S. Environmental Protection Agency, Office of Solid Waste and Emergency Response. Human Health
Evaluation Manual, Supplemental Guidance: "Standard Default Exposure Factors". Washington, D.C. OSWER
Directive 9285.6-03.

EPA. 1991b. U.S. Environmental Protection Agency, Office of Solid Waste and Emergency Response. Role of the
Baseline Risk Assessment in Superfund Remedy Selection Decisions. Washington, D.C. OSWER Directive 9355.0-30.

EPA. 1991d. U.S. Environmental Protection Agency, Office of Emergency and Remedial Response. Risk Assessment
Guidance for Superfund. Volume I. Human Health Evaluation Manual (Part B, Development of Risk-Based
Preliminary Remediation Goals). EPA Document EPA/540/R-92/003.

EPA. 1994a. U.S. Environmental Protection Agency, Office of Emergency and Remedial Response. Guidance Manual
for the Integrated Exposure Uptake Biokinetic Model for Lead in Children. EPA Publication No. 9285.7-15-1.

EPA. 1994b. U.S. Environmental Protection Agency, Technical Review Workgroup for Lead. Comments and
Recommendations on a Methodology for Estimating Risk Associated with Acute Lead Exposures at Superfund Sites.

EPA. 1994c. Revised Interim Soil Lead Guidance for CERCLA Sites and RCRA Corrective Action Facilities. U.S.
Environmental Protection Agency, Office of Solid Waste and Emergency Response. Memorandum from Elliot P.
Laws. Assistant Administrator. July 14, 1994.

EPA. 1995a. U.S. Environmental Protection Agency, Region 8 Superfund Technical Section. Standard Operating
Procedure. Evaluating Exposure from Indoor Dust.

-------
APPENDIX A

REVISION OF DIETARY IiEAD INTAKES
IN IEUBK MODEL

MEMORANDUM

TO: Bonnie Lavelle
Remedial Project Manager
Murray Smelter Site

FROM: Susan Griffin, PhD. DABT
Regional Toxicologist
Program Support Group

SUBJECT: Revision of Dietary Lead Intakes in IEUBK Model

This memorandum is in response to ASARCO's reguest to update the dietary lead intake default values in the
IEUBK Model for the Murray Smelter Site. As you are aware, the IEUBK dietary lead intake values are based on
FDA Total Diet Study data from 1986 to 1988. A number of scientific papers have been published recently by
Dr. Ellis Gunderson (Gunderson, 1995) and Dr. Michael Bolger (Bolger et al. 1996) of the U.S. FDA which
contain more recent information from the FDA's Total Diet Studies. These papers list the mean daily intake of
lead from the diet for the years from 1986-1991.

I spoke with Dr. Rob Elias of the USEPA who was responsible for the dietary lead intake component of the
IEUBK model. He indicated it would be appropriate to use the more recent FDA data to update the dietary input
values in the IEUBK model. As you may note from the FDA papers, dietary intakes are provided for children
6-11 months of age and 2 years of age. The next age group studied are teenagers 14-16 years of age. The IEUBK
model contains age-adjusted dietary lead intakes for each year up to 7 years of age. This is because the age
groups other than 6 months and 2 years were extrapolated. Originally, Dr. Elias did this by using the
information from the FDA Total Diet Studies of 1986-1988 and the data from the Pennington studies of 1975 on
food consumption rates for each age group. ASARCO is proposing to perform this extrapolation by a simpler
ratio method between the older IEUBK model values and the more recent FDA data. Dr. Elias indicated that this
was a satisfactory method and would probably not yield significantly different results from the more
complicated method of combining the FDA data with food consumption rate data. Dr. Elias did indicate that he
will be updating the dietary intake component of the IEUBK model in the near future. Those values may be
slightly different from those proposed here, because he will be combining the most recent FDA data with a new
1996 study on food consumption rates in the U.S. which is just coming out. Using the more recent FDA data to
update the IEUBK model values results in the following intakes:

Age Dietary Lead Intake (ug/day)

6-11 mos 1.82
1 year'" 1.90
2 years 1.87
3 years* 1.80
4 years* 1.73
5 years* 1.83
6 years* 2.02

'"Derived from IEUBK 99d value for 1 year divided by the ratio of the IEUBK 99d value for 6 months/ 1990-91
FDA data for 6 months
*Derived from IEUBK 99d value for that age divided by the ratio of the IEUBK 99d value for 2 years/1990-91
FDA data for 2 years

When these more recent values are input to the IEUBK model the current PRG range of 550 -1100 ppm will be
changed to 630-1260 ppm.
References

Bolger, PM, Yess, NJ, Gunderson, EL, Troxell, TC and Carrington, CD. 1996. Identification and Reduction of
Sources of Dietary Lead in the United States. Food Additives and Contaminants, Vol. 13, No 1, 53-60.

Gunderson EL. 1995. FDA Total Diet Study, July 1986-April 1991, Dietary Intakes of Pesticides, Selected
Elements, and Other Chemicals. Journal of AOAC International Vol. 78, No. 6, 1353-1363.

-------
APPENDIX B

PRELIMINARY REMEDIATION GOALS
FOR THE MURRAY SMELTER SITE

MEMORANDUM

TO: Bonnie Lavelle
RPM, Murray Smelter Site

FROM: Susan Griffin, PhD. DABT
Regional Toxicologist

SUBJECT: Preliminary Remediation Goals for the Murray Smelter Site

Development of risk-based preliminary remediation goals (PRGs) are part of the risk assessment process.
The first step involves a baseline risk assessment which uses contaminant concentrations and exposure
variables in conjunction with toxicity criteria, to estimate exposure and risk for a defined population at a
Site. At lead sites, a risk assessment is conducted by inputting contaminant concentrations into a simulation
model, the Integrated Exposure Uptake Biokinetic (IEUBK) Model, which predicts blood lead levels in children
6 months to 7 years of age. If greater than 5% of those blood lead levels exceed 10 ug/dl, the risk is
considered to be unacceptable. Risk-based PRG calculations are basically the reverse of the risk assessment
calculations. These calculations use a selected acceptable risk (e.g., no more than 5% > 10 ug/dl) and
exposure variables to estimate a desired contaminant concentration.

A single PRG could be estimated for the site using the IEUBK model with single values for both default and
site-specific parameters. Using the data from the 1996 Baseline Risk Assessment for the Murray Smelter
Superfund Site, (e.g., IEUBK model default values except for a site-specific Geometric Standard Deviation of
1.4, a soil/dust correlation coefficient of 0.35, and a soil/dust bioavailability of 35%) this single PRG
would be 550 ppm. However, we know there is variability and uncertainty in both analytical measurements
(e.g., bioavailability estimates, soil concentration, etc.), as well as population behavior and exposure. For
example, all children do not ingest the exact same amount of soil, or spend 100% of their time in one
location. Concentrations of lead in house dust are not identical for each home. These are examples of
variability. Use of randomly collected soil samples to predict the true value of lead concentrations in the
soil is an example of uncertainty. Therefore, development of PRG's which attempt to capture this uncertainty
and variability convey more information about risk at a site, than a single PRG estimate.

EPA-Region 8 is currently in the process of guantitating this uncertainty in the risk estimate and PRG
estimate for the Murray Smelter Site via a Monte Carlo analysis. This is a complex process, however, and will
not be completed until late Spring 1997. In the interim, a more simplified approach may be useful. This
approach looks at the variability around the estimate of the mean values which are used as inputs to the
IEUBK model. As you are aware, the default inputs to the IEUBK model represent average or typical values for
intake and uptake. Rather than evaluate all of the IEUBK model inputs, it is more efficacious to evaluate
those which most significantly affect the outcome. At the Murray Smelter Site the lead in soil and house dust
are the most significant sources of exposure. From this exposure pathway, the variables which impact soil and
dust exposure the most are (1) bioavailability, (2) the correlation between lead in soil and house dust, and
(3) soil ingestion rate. Based on site-specific data from the swine bioavailability study and the paired soil
and dust concentrations, the variability around the mean estimates for (1) and (2) are fairly small. This
variability would result in PRG's which ranged from 500 - 640 ppm. However, based on information from
technical documents for the National Ambient Air Quality Standard (NAAQS) for lead, the Guidance Manual for
the IEUBK Model and information from the Anaconda Childhood soil ingestion study, the variability surrounding
the mean estimate for soil ingestion is fairly significant. At the Murray Smelter site it results in a range
of PRGs from 550 - 1100 ppm for lead in soil. As you are aware, the IEUBK model was utilized originally by
the Office of Air Quality Planning and Standard (OAQPS) for the development of the lead NAAQS. Rather than
utilizing a single value for soil ingestion, the model employed a range of average estimates. As part of the
technical documentation of the NAAQS, these were reviewed and approved by the EPA's Science Advisory Board.
These ranges are documented in the 1989 OAQPS report, "Review of the National Ambient Air Quality Standards
for Lead: Exposure Analysis Methodology and Validation" and the 1994 Guidance Manual for the IEUBK Model for
Lead in Children. It wasn't until the modification of the IEUBK model by the Superfund program, that the
maximum value in that range was selected as the single soil ingestion input for the IEUBK model. In addition
a recent soil ingestion study conducted by Dr. Edward Calabrese from the University of Massachusetts for
children at the Anaconda Smelter site, yielded similar estimates of variability around a mean soil ingestion
rate. The four best tracers resulted in average estimates ranging from 89 - 126 mg/day with upper and lower
95% confidence limits around the averages ranging from 15 to 218 mg/day.

-------
   In summary,  the quantitation of variability surrounding the mean soil ingestion rate is based on
technically sound scientific data. The precedence for it's use is the development of the NAAQS for lead.  In
addition,  various points along the range have also been used on a site-specific basis at both the Leadville
and Butte NPL sites. By using a range of PRGs which take into account the variability in mean soil ingestion
rates,  more realistic information is conveyed about the variability surrounding lead exposure and risk from
soil and dust.  The range does not imply that there is greater risk at the high end of the range,  and less
risk at the low end of the range. Instead,  it suggests that any point on the range can represent EPA's risk
goal of no greater than 5% exceedance of 10 ug/dl.

   At Murray, the PRPs have suggested that the 1988 dietary default values of the model be updated and that
an in vitro bioavailability study be conducted.  In terms of how these new data may affect the PRG range of
550 - 1100 ppm,  the updated dietary information will provide only a small impact. The new range will be 600 -
1200 ppm.  Depending on the results of the in vitro study,  the change could range from minimal to significant.
Changes in bioavailability are linear with changes in PRG estimates, provided soil lead is the only or major
source of exposure. For example a reduction in bioavailability from 30% to 15% will result in a doubling of
the PRG estimate.

-------
                              APPENDIX C


     As per your request, I have taken a preliminary look at the determination of ACLs for arsenic in ground
water at the Murray Smelter Site. I have focused my effort on arsenic as this analyte is the driver for any
risk determinations from the ground-water pathway at the Site. ACLs for other analytes can easily be
determined based on this work for arsenic.

    I will provide a discussion of the concepts utilized in this ACL determination and then provide the
preliminary ACL results based on various scenarios.

ALTERNATE CONCENTRATION LIMITS AS APPLIED TO THE SITE

    Ground water at various locations on the Murray Smelter Site is contaminated with arsenic at ppm-levels.
There are potentially three distinct plumes which have migrated a relatively short distance downgradient of
the source areas. These plumes show zones of high arsenic concentrations with a significant drop-off in most
cases to background levels over a relatively short distance. The plumes are present in the water-table
aguifer of the terrace and fill deposits near Little Cottonwood Creek as well as within the floodplain
deposits of the Creek. The ultimate fate of the arsenic-contaminated ground water is discharge to Little
Cottonwood Creek.

    Historically, impacts to the Creek from Site-specific contamination have been present. Monitoring over
time has shown levels of arsenic approaching and exceeding the ambient water guality criteria (AWQC) of 190
ppb. However, recent studies have shown that this surface-water contamination can be attributed to discharges
from a drainage conduit that is present at the State Street bridge. This conduit has been found to run
southward along State Street and to have an arm that runs through the Site in the area of the former Baghouse
where one of the arsenic plumes is present. Therefore, the mechanism for measurable contaminant migration to
Little Cottonwood Creek looks to be ground-water seepage from the Baghouse plume into the drainage conduit
with rapid transport to its discharge point at the State Street bridge.

    SARA allows for the setting of ACLs for contaminants where "1)  there are known and projected points of
entry of such ground water into surface water, 2) on the basis of measurements or projections, there is or
will be no statistically significant increase of such constituents from such ground water in such surface
water at the point of entry or at any point where there is reason to believe accumulation of constituents may
occur downstream, and 3) the remedial action includes enforceable measures that will preclude human exposure
to the contaminated ground water at any point between the facility boundary and all known and projected
points of entry of such ground water into surface water". Since the impacts to Little Cottonwood Creek are
presently believed to be attributable to the drainage conduit pathway, ACLs are applicable at the Site.
However, it is recommended that a contingency plan be developed in the event that remedial actions to stop
contaminant migration in the drainage conduit do not result in significant reductions in contaminant
concentrations in Little Cottonwood Creek.

    ACLs at the Site will then be developed for the protection of surface-water guality in Little Cottonwood
Creek. The AWQC of 190 ppb will be applied to this determination. The logic behind this determination is to
assure that arsenic-contaminated ground water upon discharge to Little Cottonwood Creek will be diluted by
streamflow such that the AWQC is never exceeded in the Creek. The determination is simply a mass balance
calculation based on theoretical ground-water and surface-water flow conditions.

    If this ACL approach is accepted as the remedial action for contaminated ground water at the Site, then
the point of compliance for maintenance of the ACL is within the water-table aguifer adjacent to Little
Cottonwood Creek. That is a line of monitoring wells completed within the water-table aguifer will have to be
installed on the floodplain along the Creek and be routinely monitored for the contaminants of concern.

PRELIMINARY ALTERNATE CONCENTRATION LIMIT DETERMINATION FOR THE SITE

    I have looked at a number of hydrologic scenarios -- all based on Site-specific data in this preliminary
ACL determination. In all scenarios I considered the zone of contaminated ground-water discharge potentially
impacting the Creek to be the stretch from SW-2 downstream to SW-3 or a distance of approximately 3500 feet.
This assumption is based on a combined analysis of the ground-water flow directions and contaminant plume
distributions at the Site; if both the Baghouse and Arsenic Storage Bin plumes were to migrate to the Creek,
based on the existing ground-water flow information, their discharge and impacts to the Creek would occur
between SW-2 and SW-3. Also, in all scenarios I have only considered ground-water discharge to the Creek from
the Site or south side as the Site-specific database focuses on this ground-water flow system; this is a
conservative assumption as based on the conceptual model for the area a component of ground-water flow from
the north discharging to the Creek will exist. Lastly, a background arsenic level in surface water of 0.007
ppm was used; this was the maximum value detected in samples from SW-2 where most of the sampling results

-------
were below the detection limit of 0.005 ppm. I will summarize each scenario and the ACL for arsenic result.

Scenario 1

Under Scenario 1 the determinations of ground-water flow and surface-water discharge utilized in
ground-water flow and solute transport modeling for the Site Characterization and Feasibility Study reports
were input into a mass balance eguation. The values for ground-water flow from the Site to the Creek range
from 0.02 to 1.92 cfs based on this analysis. A low-flow discharge rate for Little Cottonwood Creek was
estimated based on Site-specific data to be 3.0 cfs.

Using the above values for flow conditions, the background arsenic level, and the AWQC criteria, the ACL
for arsenic under this scenario would range from 0.476 to 27.6 mg/1. (See attachment for calculations.)

Scenario 2

Under Scenario 2 determinations of ground-water flow and surface-water discharge to be utilized were
based on my assessment of the Site-specific database. Data used included that from the Site Characterization
and Feasibility Study reports as well as the guarterly monitoring program results. The evaluation focused on
ground-water flow within the floodplain alluvium of the Creek. The hydraulic conductivity for MW-112 was used
and the hydraulic gradient was determined based on ground-water flow between MW-112 and Well 2. The value for
ground-water flow from the Site to the Creek was determined to be 0.0075 cfs based on this analysis. A
low-flow discharge rate for Little Cottonwood Creek was estimated based on Site-specific data for SW-2 to be
2.5 cfs.

Using the above values for flow conditions, the background arsenic level, and the AWQC criteria, the ACL
for arsenic under this scenario would be 61.2 mg/1. (See attachment for calculations.)

DISCUSSION OF ALTERNATE CONCENTRATION LIMIT RESULTS

The results of this exercise are ACLs for arsenic at the Site ranging from 0.476 to 61.2 mg/1. In theory
these ACLs if attained at the POC should assure that the AWQC of 190 ppb is not exceeded in Little Cottonwood
Creek due to contaminated ground-water discharge from the Site. These values are conservative in that no
ground-water discharge from north of the Site was considered in this determination.

Based on the existing database for the Site, only the lowest determined ACL (0.476 mg/1) is exceeded in
monitoring points (monitoring wells or hydropunch sample sites). The other values exceed any detected
concentrations on-Site.

These ACLs show a range of over two orders of magnitude (0.476 to 61.2 mg/1). This range provides an
indication of the levels of uncertainty in this type of determination. As a result, it is imperative that if
this ACL approach is accepted as the remedial action for contaminated ground water at the Site, then a
significant monitoring network needs to be established within the water-table aguifer on the floodplain along
the Creek. This network will need to be routinely monitored for the contaminants of concern.

If you should have any guestions, please feel free to contact me at x6595.

-------
                            ACL CALCULATIONS
Scenario 1

AWQC:

Background Surface-Water:

Ground-Water Flow:
Surface-Water Flow:

ACL

    For Q GW = 0.02 cfs



    For Q GA = 1.92 cfs
C AWQC =0.19 ppm

C BKG = 0.007 ppm

Q GW = KiA

K = 5 ft/d
i = 0.008 ft/ft
A = 43,200 ft 2

Q GW = 0.02 cfs

Q SW = 3.0 cfs
    (SW-2  maximum)

    (FS  modeling work)

     K = 154  ft/d
     i = 0.028  ft/ft
     A = 43,200 ft  2

     Q GW = 1.92 cfs

     (Estimated)
Q SWC BKG + Q GWC ACL =  (Q SW + Q  GW)C AWQC

(3.0 cfs) X  (0.007 ppm) +  (0.02 cfs)C ACL  =  (3.0  cfs  +
0.02 cfs) X  (0.19 ppm)
C ACL= 27.6 ppm

(3.0 cfs) X  (0.007 ppm) +  (1.92 cfs)C ACL  =  (3.0  cfs  +
1.92 cfs) X  (0.19 ppm)
C ACL = 0.476 ppm
Scenario 2

AWQC:

Background Surface-Water:

Ground-Water Flow:
Surface-Water Flow:

ACL
C AWQC =0.19 ppm

C BKG = 0.007 ppm

Q GW = KiA

K = 14 ft/d
i = 0.0012 ft/ft

A = 38,500 ft 2

Q GW = 0.0075 cfs

Q SW = 2.5 cfs
     (SW-2  maximum)
                                                   (MW-112  slug tests)
                                                   (1/97  ground-water flow between
                                                   MW-112 and Well-2)
                                                   (A = bXI = 11 ft X 3500 ft)
(Estimated)
                             Q  SWC  BKG  +  Q  GWC  ACL = Q SW + Q GW)C AWQC

                              (2.5 cfs)  X  (0.007 ppm)  + (0.0075cfs)C ACL = (2.5 cfs +
                             0.0075  cfs)  X  (0.19 ppm)
                             C  ACL=  61.2  ppm

-------

As per your request, I have taken a final look at the determination of an ACL for arsenic in ground water
at the Murray Smelter Site. ACLs for other analytes can easily be determined based on this work, if
necessary.

I will not provide a thorough discussion of the concepts of the ACL determination as this information is
detailed in my memorandum on this subject to you dated November 5, 1997. I will provide the ACL results based
on the various scenarios detailed in that memo.

AN ALTERNATE CONCENTRATION LIMIT FOR ARSENIC AS APPLIED TO THE SITE

An ACL for arsenic at the Site will be developed for the protection of surface-water quality in Little
Cottonwood Creek. The Utah Agricultural Water Standard of 0.1 ppm for arsenic will be applied to this
determination. The logic behind this determination is to assure that ground water contaminated with arsenic
upon discharge to Little Cottonwood Creek will be diluted by streamflow such that the 0.1 ppm concentration
is not exceeded in the Creek. The determination is simply a mass balance calculation based on theoretical
ground-water and surface-water flow conditions.

ALTERNATE CONCENTRATION LIMIT DETERMINATION FOR THE SITE

I have looked at a number of hydrologic scenarios - all based on Site-specific data in the ACL
determination. In all scenarios I considered the zone of contaminated ground-water discharge potentially
impacting the Creek to be the stretch from SW-2 downstream to SW-3 or a distance of approximately 3500 feet.
This assumption is based on a combined analysis of the ground-water flow directions and contaminant plume
distributions at the Site; if both the Baghouse and Arsenic Storage Bin plumes were to migrate to the Creek,
based on the existing ground-water flow information, their discharge and impacts to the Creek would occur
between SW-2 and SW-3. Also, in all scenarios I have only considered ground-water discharge to the Creek from
the Site or south side as the Site-specific database focuses on this ground-water flow system; this is a
conservative assumption as based on the conceptual model for the area a component of ground-water flow from
the north discharging to the Creek will exist. Lastly, a background arsenic level in surface water of 0.007
ppm was used; this was the maximum value detected in samples from SW-2 where most of the sampling results
were below the detection limit of 0.005 ppm. I will summarize each scenario and the ACL for arsenic below.

Scenario 1

Under Scenario 1 the determinations of ground-water flow and surface-water discharge utilized in
ground-water flow and solute transport modeling for the Site Characterization and Feasibility Study reports
were input into a mass balance equation. The values for ground-water flow from the Site to the Creek range
from 0.02 to 1.92 cfs based on this analysis. A low-flow discharge rate for Little Cottonwood Creek was
estimated based on Site-specific data to be 3.0 cfs.

Using the above values for flow conditions, the background arsenic level, and the Agricultural Standard,
the ACL for arsenic under this scenario would range from 0.245 to 14.05 mg/1. (See attachment for
calculations.)

Scenario 2

Under Scenario 2 determinations of ground-water flow and surface-water discharge to be utilized were
based on my assessment of the Site-specific database. Data used included that from the Site Characterization
and Feasibility Study reports as well as the quarterly monitoring program results. The evaluation focused on
ground-water flow within the floodplain alluvium of the Creek. The hydraulic conductivity for MW-112 was used
and the hydraulic gradient was determined based on ground-water flow between MW-112 and Well 2. The value for
ground-water flow from the Site to the Creek was determined to be 0.0075 cfs based on this analysis. A
low-flow discharge rate for Little Cottonwood Creek was estimated based on Site-specific data for SW-2 to be
2.5 cfs.

Using the above values for flow conditions, the background arsenic level, and the Agricultural Standard,
the ACL for arsenic under this scenario would be 31.1 mg/1. (See attachment for calculations.)

DISCUSSION OF ALTERNATE CONCENTRATION LIMIT RESULTS

The results of this exercise are ACLs for arsenic at the Site ranging from 0.245 to 31.1 mg/1. In theory
these ACLs if attained at the POC should assure that the Utah Agricultural Standard for arsenic of 0.1 ppb
are not exceeded in Little Cottonwood Creek due to contaminated ground-water discharge from the Site. These
values are conservative in that no ground-water discharge from north of the Site was considered in this
determination.

-------
    The arsenic ACLs show a range of over two orders of magnitude (0.245 to 31.1 mg/1),  This range provides
an indication of the levels of uncertainty in this type of determination.  As a result,  it is imperative that
if this ACL approach is accepted as the remedial action for contaminated ground water at the Site, then a
significant monitoring network needs to be established within the water-table aguifer on the floodplain along
the Creek.  This network will need to be routinely monitored for the contaminants of concern.

    If you should have any guestions, please feel free to contact me at x6595.

-------
                          ARSENIC ACL  CALCULATIONS

Scenario 1

Utah Agricultural Standard:  C AG =  0-1  ppm

Background Surface-Water:    C BKG = 0.007  ppm
Ground-Water Flow:
Surface-Water Flow.

ACL:

    For Q GW = 0.02 cfs



    For Q GW = 1.92 cfs
Q GW = KiA

K = 5 ft/d
i = 0.008 ft/ft
A = 43,200 ft 2

Q GW = 0.02 cfs

Q SW = 3.0 cfs
 (SW-2 maximum)

 (FS modeling work)

K = 154 ft/d
i = 0.028 ft/ft
A = 43,200 ft 2

Q GW = 1.92 cfs

 (Estimated)
Q SWC BKG + Q GWC ACL =  (Q SW + Q GW)C AG

(3.0 cfs) X  (0.007 ppm) +  (0.02 cfs)C ACL =  (3.0  cfs  +
0.02 cfs) X  (0.1 ppm)
C ACL = 14.05 ppm

(3.0 cfs) X  (0.007 ppm) +  (1.92 cfs)C ACL =  (3.0  cfs  +
1.92 cfs) X  (0.1 ppm)
C ACL= 0.245 ppm
Scenario 2

Utah Agricultural Standard:  C AG =  0.1  ppm

Background Surface-Water:    C BKG = 0.007  ppm
Ground-Water Flow:
Surface-Water Flow:

ACL
Q GW = KiA

K = 14 ft/d
i = 0.0012 ft/ft

A = 38,500 ft 2

Q GW = 0.0075 cfs

Q SW = 2.5 cfs
                           (SW-2 maximum)
                                                        (MW-112  slug tests)
                                                        (1/97 ground-water flow between
                                                        MW-112 and Well-2)
                                                        (A = bXI = 11 ft X 3500 ft)
 (Estimated)
                             Q SWC  BKG  +  Q  GWC ACL = (Q SW + Q GW)C AG

                              (2.5 cfs)  X  (0.007  ppm)  +  (0.0075 cfs)C ACL = (2-5 cfs +
                             0.0075  cfs)  X  (0.1  ppm)
                             C ACL  =31.1 ppm

-------