Site-wide Baseline Ecological Risk Assessment Libby Asbestos Superfund Site Libby, Montana


Site-wide
Baseline Ecological
Risk Assessment

Libby Asbestos Superfund Site

Libby, Montana

December 2014

-------
This page left blank intentionally

-------
Site-wide Baseline Ecological Risk Assessment
Libby Asbestos Superfund Site
Libby, Montana

December 2014

Prepared for:

U.S. Environmental Protection Agency, Region 8

Prepared by

5>"

SRC, Inc.
And

TrM.h

CDM Federal Programs Corporation

-------
This page left blank intentionally

-------
Site-Wide Baseline Ecological Risk Assessment for
Asbestos

Libby is a community in northwestern Montana that is located near a former vermiculite mine.
Vermiculite from the mine contains varying concentrations of amphibole asbestos, referred to as
"Libby amphibole asbestos" or LA. In October 2002, the Libby Asbestos Superfund Site (Site] was
listed on the U.S. Environmental Protection Agency (EPA] National Priority List. The Site includes
properties that may have become contaminated with LA as a result of the vermiculite mining and
processing conducted in and around Libby, as well as other areas at and around the mine that may
have been affected by mining-related releases of LA.

This document presents the Site-wide baseline ecological risk assessment (BERA] for asbestos. The
purpose of this document is to describe the likelihood, nature, and extent of adverse effects in
ecological receptors exposed to asbestos at the Site as a result of releases of LA to the environment
from past mining, milling and processing activities at the Site. This information, along with other
relevant information, will be used by risk managers to decide whether remedial actions are needed to
protect ecological receptors from the effects of exposure to mining-related environmental
contamination. If actions are warranted, the results of the BERA will be used, along with other
relevant information, to assess the appropriate remedial actions needed to protect ecological
receptors.

Due to the complexity of the Site and to facilitate a multi-phase approach to remediation, the Site has
been divided into eight operable units (OUs], This document presents the BERA in two parts; Part 1 is
the BERA for Operable Unit 3 (OU3], which includes the mine and the surrounding areas, and Part 2 is
the BERA for the other seven OUs (i.e., all non-OU3 OUs],

-------
This page left blank intentionally

-------
Part 1 Baseline Ecological Risk Assessment for
Asbestos - Operable Unit 3

|

-------
This page left blank intentionally

-------
FINAL

BASELINE ECOLOGICAL RISK ASSESSMENT
FOR EXPOSURE TO ASBESTOS
LIBBY ASBESTOS SUPERFUND SITE

PART 1
OPERABLE UNIT 3

Prepared by
U.S. Environmental Protection Agency
Region 8
Denver, CO

^eD_sX



With Technical Assistance from:

SRC, Inc.
Denver, CO

and

CDM Smith
Denver, CO

c&

December 2014

-------
FINAL

APPROVALS

, n

1/ ///	, I¦ i

Approved by:	'				 Dale:	[ LjPJ ''*"2.

U.S. EPA Region 8, Superfunu Ideological Risk Assessor

1--"	^

Dan Wall



' ^ /' /"->

/ // ' / ;

Approved by; 	;	"* 'T				 Date;

z. .

''•"Y

C'ltrisfiua Proi^vs

i\S. EPA Rt%ipK X. Reraed'al Project Manager

'J/'" f"	Date: i.V. .T

Approved by:		 Date:

\\ Jlliain Bruit in

SRC' Work -vx-agisnitrii Marker

i

-------
FINAL

TABLE OF CONTENTS

1.0 INTRODUCTION	1

1.1	Purpose of this Document	1

1.2	Document Organization	1

2.0 SITE CHARACTERIZATION	2

2.1	Overview	2

2.2	Physical Setting	3

2.2.1	Topography	3

2.2.2	Land Ownership or Stewardship	3

2.2.3	Climate	3

2.2.4	Surface Water Features	4

2.3	Current Condition of the Mine Site	5

2.4	Nature and Extent of LA Contamination at the Site	5

2.4.1	Mineral Characteristics of LA	5

2.4.2	Concentrations of LA in Environmental Media	6

2.5	Ecological Setting	10

2.5.1	Terrestrial Setting	10

2.5.2	Aquatic Setting	11

2.5.3	Federal and State Species of Special Concern	11

3.0 PROBLEM FORMULATION	 13

3.1	Conceptual Site Model	13

3.1.1	Potential Sources of Contamination	13

3.1.2	Migration Pathways in the Environment	13

3.1.3	Potentially Exposed Ecological Receptors	14

3.1.4	Exposure Pathways of Chief Concern	14

3.2	Management Goal and Assessment Techniques	17

3.2.1	Management Goal	17

3.2.2	Assessment Endpoints	18

3.2.3	Measures of Effect	18

3.2.4	Statistical Methods	20

3.2.5	Data Evaluation	22

3.3	Role of the BT AG	23

4.0 RISKS TO FISH	24

4.1	Reported Effects	24

4.2	Site-Specific Toxicity Tests	25

4.2.1	In Situ Eyed Egg and Alevin Exposure Studies	25

4.2.2	In Situ Juvenile Fish Study	33

4.3	Population Studies	37

4.3.1	Demographic Studies	38

4.3.2	Habitat Studies	40

4.4	In Situ Lesion Studies	42

4.5	Weight of Evidence Evaluation for Fish	47

5.0 RISKS TO BENTHIC MACROINVERTEBRATES	49

5.1	Reported Effects	49

5.2	Laboratory Toxicity Tests	49

5.2.1	Study Design	49

5.2.2	Results for Hyalella	51

ii

-------
FINAL

5.2.3	Results for Chironomus	53

5.2.4	Discussion	55

5.3	Population Studies	55

5.3.1	Demographic Measurements	55

5.3.2	Habitat Studies	59

5.4	In Situ Lesion Studies	60

5.5	Weight of Evidence Evaluation	60

6.0 RISKS TO AMPHIBIANS	62

6.1	Reported Effects	62

6.2	Laboratory Toxicity Tests	62

6.2.1	Study Design	62

6.2.2	Results	63

6.3	Population Studies	65

6.4	In Situ Lesion Studies	65

6.4.1	Study Design	65

6.4.2	Results	67

6.5	Weight of Evidence Evaluation	71

7.0 RISKS TO MAMMALS	72

7.1	Reported Effects	72

7.2	Laboratory Toxicity Tests	72

7.3	Population Studies	73

7.4	In Situ Lesion Studies	73

7.4.1	Study Design	73

7.4.2	Results	75

7.5	Weight of Evidence Evaluation	79

8.0 RISKS TO BIRDS	81

8.1	Reported Effects	81

8.2	Laboratory Toxicity Tests	81

8.3	Population Studies	81

8.4	In Situ Lesion Studies	82

8.5	Weight of Evidence Evaluation	83

9.0 SUMMARY AND CONCLUSIONS	85

10.0 REFERENCES	87

TABLES AND FIGURES	96

ATTACHMENTS

Attachment A: Wildlife Species That May Occur in OU3

Attachment B Summary of Laboratory-Based Surface Water Toxicity Tests

Attachment C: Avian Respiratory System: Overview of Anatomy and Function as Related to
Particulate Inhalation

Attachment D Study Reports (electronic)

iii

-------
FINAL

LIST OF FIGURES

Figure 2-1	Site Location

Figure 2-2	OU3 Study Area

Figure 2-3	Land Ownership

Figure 2-4	Average Temperature and Precipitation in Libby

Figure 2-5	Wind Rose at Mine Site

Figure 2-6	Surface Water Features

Figure 2-7	Mined Area Features

Figure 2-8	Ambient Air Monitoring Locations

Figure 2-9	Surface Water and Sediment Sampling Stations

Figure 2-10	LA Concentration vs. Flow in Lower Rainy Creek

Figure 2-11	LA Concentrations in Soil, Duff and Tree Bark

Figure 2-12	LA Concentrations in Bark and Duff as a Function of Distance from the Mine

Figure 3-1	Conceptual Site Model for Ecological Exposure to Asbestos

Figure 4-1	Design and Function of a Whitlock-Vibert Box

Figure 4-2	Example of Whitlock-Vibert Boxes Buried in Sediment in Lower Rainy Creek

Figure 4-3	2013 Eyed Egg Exposure Study Temperature and Flow Data

Figure 4-4	2013 Eyed Egg Study Exposure Concentrations

Figure 4-5	2013 Eyed Egg Study Results

Figure 4-6	2013 Alevin Size and Weight Data

Figure 4-7	Example Juvenile Trout Cages

Figure 4-8	Juvenile Trout In Situ Exposure Conditions

Figure 4-9	Juvenile Trout Size and Growth Data

Figure 4-10	Fish Density, Weight, and Biomass

Figure 4-11	Habitat Quality Metrics

Figure 5-1	Laboratory Toxicity Results for Hyalella azteca

Figure 5-2	Laboratory Toxicity Results for Chironomus tentans

Figure 5-3	RBP Biological Condition Scores

Figure 5-4	Mountain MMI Scores

Figure 5-5	Habitat Quality Scores

Figure 5-6	Correlation between Community Status and Habitat Quality

Figure 6-1	Survival and Metamorphosis in Exposed Organisms

Figure 6-2	Size and Weight of Pre-Metamorphic Amphibians Field Stages 1-2

Figure 6-3	Size and Weight of Proto-Metamorphic Amphibians Field Stages 3-6

Figure 6-4	Size and Weight of Metamorphosed Amphibians Field Stage 8

Figure 7-1	Small Mammal Trap Line Locations in OU3

Figure 7-2	Small Mammal Trap Line Locations for the Reference Area

Figure 7-3	Histology Scores for Deer Mice

IV

-------
FINAL

LIST OF TABLES

Table

2-1

Table

2-2

Table

2-3

Table

2-4

Table

2-5

Table

2-6

Table

4-1

Table

4-2

Table

4-3

Table

4-4

Table

4-5

Table

4-6

Table

4-7

Table

4-8

Table

4-9

Table

4-10

Table

4-11

Table

4-12

Table

4-13

Table

4-14

Table

4-15

Table

5-1

Table

5-2

Table

5-3

Table

5-4

Table

5-5

Table

5-6

Table

5-7

Table

5-8

Table

5-9

Table

6-1

Table

6-2

Table

6-3

Table

6-4

Table

6-5

Table

6-6

Table

6-7

Table

6-8

Table

6-9

Table

6-10

Table

7-1

Table

7-2

Table

7-3

Table

7-4

Table

7-5

Table

7-6

Summary Statistics for LA in Mine Waste

Summary Statistics for LA in Ambient Air

Summary Statistics for LA in Surface Water

Summary Statistics for LA in Sediment

Federal Species of Concern in the Kootenai National Forest

State Species of Concern Occurring In or Near OU3

2013 Eyed Egg Survival Data
2013 Eyed Egg Study Statistical Comparisons
Abnormal Swimming Behavior in 2013 Study
2013 Alevin External Lesion Frequency Data
Description of Lesions Observed in Alevins
Juvenile Trout Survival Data

External Lesion Scoring System for Caged Juvenile Trout

Juvenile Trout External Lesion Data

Number of Fish Captured by Electroshocking

Fish Species Captured by Electroshocking

Barriers to Fish Movement in Rainy Creek

Resident Trout Captured and Evaluated

Resident Trout External Lesion Data

Resident Trout Histological Lesion Data

Weight of Evidence Summary for Fish

Physical Characteristics of Site and Reference Sediments
Concentration Data for Site-Specific Sediments
Concentration of LA in Sediment Porewater
Kick Net Benthic Macroinvertebrate Community Data
RBP BCS Calculations Based on Kick Net Data
Surber Benthic Macroinvertebrate Community Data
Mountain MMI Scores Based on Surber Data
Benthic Habitat Quality Data and Scores
Weight of Evidence Summary for Benthic Invertebrates

Growth and Survival Endpoints in Amphibian Laboratory Study

Measurement Endpoints for Amphibian Field Study

Estimated Concentrations of LA in Sediment

Exposure Conditions in Water

Amphibians Collected During Field Study

Metamorphs Sent for Histological Examination

List of Tissues Examined Histologically

Frequency of Histologic Lesions in Field-Collected Metamorphs
Severity of Histologic Lesions in Field-Collected Metamorphs
Weight of Evidence Summary for Amphibians

Small Mammal Species Captured
Size Data for Deer Mice
Gender Distribution of Mice
Estimated Age of Mice

Small Mammal Lesion Frequency and Severity
Weight of Evidence Summary for Mammals

v

-------
FINAL

LIST OF ACRONYMS

BERA

Baseline Ecological Risk Assessment

BCS

Biological Condition Score

BMI

Benthic Macroinvertebrate

BTAG

Biological Technical Assistance Group

BTT

Bobtail Creek Tributary

CC

Carney Creek

CSM

Conceptual Site Model

DO

Dissolved Oxygen

DQO

Data Quality Objective

EMAP

Environmental Monitoring and Assessment Program

EPA

U.S. Environmental Protection Agency

EPT

Ephemeroptera, Plecoptera, Trichoptera

ERT

Environmental Response Team

FEL

Fort Environmental Laboratory, Inc.

GI

Gastrointestinal

HBI

Hilsenhoff Biotic Index

HQ

Hazard Quotient

ISO

International Organization for Standardization

KDC

Kootenai Development Corporation

LA

Libby Amphibole

LRC

Lower Rainy Creek

MDEQ

Montana Department of Environmental Quality

MFL

Million Fibers per Liter

MMI

Multimetric Index

MNHP

Montana National Heritage Program

MFWP

Montana Fish, Wildlife, and Parks

MLE

Maximum Likelihood Estimate

NBF

Neutral Buffered Formalin

NSY

Noisy Creek

ou

Operable Unit

PAH

Polycyclic Aromatic Hydrocarbon

PCB

Polychlorinated Biphenyl

PLM

Polarized Light Microscopy

QAPP

Quality Assurance Project Plan

RBP

Rapid Bioassessment Protocol

RI

Remedial Investigation

SAP

Sampling and Analysis Plan

SVL

Snout-Vent Length

SOP

Standard Operating Procedure

TEM

Transmission Electron Microscopy

TP-TOE

Tailings Pond Toe

TTM

Time to Metamorphosis

URC

Upper Rainy Creek

USGS

U.S. Geological Survey

USFS

U.S. Forest Service

USFWS

U.S. Fish and Wildlife Service

vi

-------
FINAL

EXECUTIVE SUMMARY

1.0 INTRODUCTION

This document is a Baseline Ecological Risk Assessment (BERA) for Operable Unit 3 (OU3) of
the Libby Asbestos Superfund Site, located near Libby, Montana. The purpose of this BERA is
to describe the likelihood, nature, and extent of adverse effects in ecological receptors exposed to
asbestos in OU3 as a result of releases of asbestos to the environment from past mining, milling
and processing activities at the Site. This information, along with other relevant information, is
used by risk managers to decide whether remedial actions are needed to protect ecological
receptors in OU3 from the effects of exposure to mining-related environmental asbestos
contamination. If actions are warranted, the results of the BERA will be used with other relevant
information to assess the appropriate remedial actions needed to protect ecological receptors.

An evaluation of potential ecological risks due to other (non-asbestos) contaminants in OU3 is
presented in a separate report (EPA 2013a).

2.0 SITE CHARACTERIZATION

Overview

Libby is a community in northwestern Montana that is located near a large open-pit vermiculite
mine, referred to as the Zonolite Mine. The mine began limited operations in the 1920s and was
operated on a larger scale from approximately 1963 to 1990. The mine is now closed and all
buildings have been removed.

Vermiculite is a naturally-occurring silicate mineral that has found a range of commercial
applications such as packing material, attic and wall insulation, various garden and agricultural
products, and various cement and building products.

The vermiculite ore deposit at the mine in Libby contains a form of asbestos referred to as Libby
Amphibole (LA). Historic mining, milling, and processing of vermiculite at the site are known
to have caused releases of vermiculite and LA to the environment. Inhalation of LA is known to
have caused a range of adverse health effects in exposed humans, including workers at the mine
and processing facilities as well as residents of Libby. Exposure to asbestos released to the
environment may also be having adverse effects on aquatic and/or terrestrial wildlife near the
mine. Based on these concerns, the U.S. Environmental Protection Agency (EPA) listed the
Libby Asbestos Superfund Site on the National Priorities List in October 2002.

ES-1

-------
FINAL

Given the size and complexity of the site, EPA divided the site into a series of Operable Units
(OUs). This Section presents an evaluation of risks to ecological risks from exposure to LA
within OU3, which includes the property in and around the Zonolite Mine and any area impacted
by the release and subsequent migration of LA or other contaminants from the mine.

An evaluation of ecological risks for other OUs within the Libby Superfund Site is presented in
Part 2 of the Site-wide BERA.

Physical Setting

The terrain in OU3 is mainly mountainous with dense forests and steep slopes. Figure ES-1
shows the main surface features in the vicinity of the mine. There are a number of areas where
mine wastes have been disposed, including waste rock dumps (mainly on the south side of the
mine), coarse tailings (mainly to the north of the mine), and fine tailings (placed in a tailings
impoundment on the west side of the site). The main surface water bodies within OU3 include
Rainy Creek, Fleetwood Creek, Carney Creek, the large tailings impoundment on the west side
of the mine, and a smaller pond on Rainy Creek below the tailings impoundment.

Nature and Extent of LA Contamination at the Site

A large number of environmental samples from OU3 have been collected and analyzed for LA.
These samples have revealed the following general conclusions:

LA in Ore and Mine Wastes: The concentration of LA in veins of amphibole within the
vermiculite deposit can be as high as 50-75%. Concentrations of LA in mine waste samples
generally are in the range of about 0.2% to 1%, although some samples may be higher.

LA in Ambient Air Near the Mine: LA concentrations in air near the mine are generally low,
often below the detection limit. The average concentration is about 0.0002 fibers per cubic
centimeter of air (f/cc). LA fibers that occur in air near the mine presumably arise due to wind or
other disturbances that release fibers from existing sources (contaminated soil, tailings, waste
rock, duff, etc.) into air.

LA in Surface Water: Concentrations of LA in surface waters of OU3 are variable, but are often
in the range of 5 to 50 million fibers per liter (MFL), although some samples are higher.
Concentrations tend to be highest during the high flows typically associated with the spring
runoff.

ES-2

-------
FINAL

LA in Tailings and Sediment: LA can be detected in nearly all samples of tailings and sediment
from streams and ponds in OU3, with estimated concentrations ranging from less than 0.2% up
to as high as 10%.

LA in Forest Soil Duff and Tree Bark: Concentrations of LA in soil, duff (forest litter), and tree
bark in forest areas around the mine are variable, but show a clear tendency to decrease as a
function of distance from the mine.

Ecological Setting

Terrestrial Setting

The mined area was heavily disturbed by past mining activity and some areas remain largely
devoid of vegetation. Outside the mined area, the forested area of OU3 is suitable habitat for a
wide range of terrestrial species, including a variety of mammals, birds, and reptiles.

Aquatic Setting

The streams and ponds within OU3 provide habitat for a range of aquatic species including fish,
benthic macroinvertebrates, and amphibians. Fish surveys performed in OU3 streams indicate
that the most common species of fish are western cutthroat trout, rainbow trout, and "cutbow"
trout (a rainbow/cutthroat hybrid). Aquatic invertebrate community surveys in OU3 indicate that
the most common types of aquatic invertebrates observed include mayflies, stoneflies,
caddisflies, true flies, and beetle larvae. The most common amphibian species observed are the
tree frog, spotted frog, and western toad.

3.0 PROBLEM FORMULATION

Conceptual Site Model

Based on the information that is available on the nature and extent of LA contamination in the
environment in OU3 and the types of species that are known or expected to be present, it is
considered likely that many species of ecological receptors, both aquatic and terrestrial, may be
exposed to LA. The main focus of this risk assessment includes the following groups:

Fish

Benthic macroinvertebrates

•	Amphibians

•	Mammals
Birds

ES-3

-------
FINAL

Management Goal

The overall management goal identified for ecological receptors at the Libby OU3 site for
asbestos contamination is:

Ensure adequate protection of ecological receptors within OU3 from the adverse effects
of exposures to mining-related releases of asbestos to the environment.

For most species, "adequate protection" is defined as the reduction of risks to levels that will
result in the recovery and maintenance of healthy local populations and communities of biota.
For threatened or endangered species, "adequate protection" is generally interpreted to mean
minimizing risks to individual members of the population.

Assessment Endpoints

Assessment endpoints are the characteristics of the ecological systems that are to be protected.
Because the risk management goal is formulated in terms of the protection of individuals and
populations of ecological receptors, the assessment endpoints selected for use in this problem
formulation focus on parameters that are directly related to the management goal. This includes:

• Mortality

• Growth

• Reproduction

If effects on these three assessment endpoints are absent or minimal, it is likely that ecologically
significant effects will not occur.

Measures of Effect

There are a number of alternative measures of effect that may be investigated as part of an
ecological risk assessment. The primary alternative strategies for characterizing measures of
effects are described below.

Hazard Quotients

For most environmental contaminants, the first line of investigation is usually the Hazard
Quotient (HQ) Approach. A Hazard Quotient (HQ) is the ratio of the estimated exposure of a
receptor to a "benchmark" exposure that is believed to be without significant risk of unacceptable
adverse effect. However, there are no established benchmarks for the evaluation of ecological

ES-4

-------
FINAL

receptors to any form of asbestos, and most of the studies that are available that might potentially
serve as a basis for development of a benchmark are based on studies of chrysotile asbestos
rather than amphibole asbestos. Consequently, HQ values were not calculated for any exposure
scenario, and ecological investigations of the potential effects of LA on ecological receptors in
OU3 focused on other measures of effect, as discussed below.

Site-Specific Toxicity Tests

Site-specific toxicity tests measure the response of receptors that are exposed to site media. This
may be done either in the field {in situ) or in the laboratory using media collected from the site.
The chief advantage of either type of study is that site-specific conditions that can influence
toxicity are usually accounted for, and that the cumulative effects of all exposure pathways to the
medium and all contaminants in the medium are evaluated simultaneously. One potential
limitation of this approach is that, if toxic effects are observed when test organisms are exposed
to site media, it may not be possible to specify which contaminant or combination of
contaminants is responsible for the effect without further testing or evaluation. A second
limitation is that it may be difficult to perform tests on site samples that reflect the full range of
environmental conditions which may occur in the field across time and space, so it may not be
possible to fully identify all conditions that are and are not of concern.

Population and Community Demographic Observations

Another approach for evaluating possible adverse effects of environmental contamination on
ecological receptors is to make direct observations on the receptors in the field, seeking to
determine whether any receptor population has unusual numbers of individuals (either lower or
higher than expected), or whether the diversity (number of different species) of a particular
category of receptors is different than expected. The chief advantage of this approach is that
observation of community status relate directly to the management goal (protection of
populations). However, there are also a number of limitations to this approach. The most
important of these is that both the abundance and diversity of a receptor depend on many site-
specific factors (habitat suitability, availability of food, predator pressure, natural population
cycles, meteorological conditions, etc.), and it is often difficult to know what the expected (non-
impacted) abundance and diversity should be in a particular area. This problem is generally
approached by seeking an appropriate "reference area" (either the site itself before the impact
occurred, or some similar site that has not been impacted), and comparing the observed
abundance and diversity in the reference area to that for the site. However, it is sometimes
difficult to locate reference areas that are a good match for all important habitat and ecological
characteristics.

ES-5

-------
FINAL

In-Situ Measures of Exposure and Effects

An additional approach for evaluating the possible adverse effects of environmental
contamination on ecological receptors is to make direct observations on receptors in the field,
seeking to determine if individuals residing in areas of contamination have an increased
frequency and/or severity of lesions and/or deformities compared to organisms residing in
uncontaminated reference areas. This method has the advantage of integrating most factors that
influence the true level of exposure and toxicity of contaminants in the field. However, if an
increased incidence or severity of lesions is observed, it may not be possible to identify with
certainty which environmental contaminant(s) is (are) responsible, and it may also be difficult to
determine with confidence whether the observed lesions are likely to cause an ecologically
significant population-level impact.

Weight of Evidence Evaluation

As noted, each of these alternative strategies for characterizing ecological risks has some
advantages and some limitations. Because of this, the risk assessment for OU3 sought to collect
information from two or more lines of evidence whenever feasible. If two or more lines of
evidence are available, and if the lines of evidence are in general agreement, then confidence in
risk conclusions is increased. If two or more lines of evidence do not agree, then careful
attention must be given to likely reasons for the disparity, and to decide which line(s) of
evidence provide the highest confidence.

Detailed descriptions of the studies performed to investigate potential risks to each group of
exposed receptors are presented in Section 4 (fish), Section 5 (benthic macroinvertebrates),
Section 6 (amphibians), Section 7 (mammals), and Section 8 (birds) of the main risk assessment.
The weight of evidence conclusions are summarized below.

4.0 RISKS TO FISH

Four lines of evidence are available to help evaluate the effects of exposure of fish to LA in site
waters, including:

•	In situ toxicity studies of eyed eggs and alevins

•	In situ toxicity tests of juvenile trout

•	Fish population studies

•	Resident fish lesion studies

ES-6

-------
FINAL

The population studies indicates that trout population structure in LRC is different from
reference streams, with decreased fish density, increased fish size, and decreased biomass. This
observation could be consistent with a hypothesis that LA in site waters is toxic to trout and
results in a decreased number of fish, but several observations suggest that LA is not the likely
cause of the difference:

• There are several habitat quality factors that are lower in LRC than reference streams
(especially spawning gravel, woody debris, water temperature, and pool availability).
These habitat factors show a relatively strong correlation with trout density, suggesting
that habitat likely accounts for much of the apparent difference.

• In situ toxicity studies of early life stage trout indicate there might be a small decrease in
hatching success of eyed eggs in lower Rainy Creek than in reference streams, but this
cannot be attributed to LA. Moreover, the difference is sufficiently small (<10%) that a
substantial effect on population density would not be expected (Toll et al. 2013).

• No effects that might contribute to decrease survival of larger fish have been detected,
either in caged juvenile fish studies or studies of resident fish. This is consistent with
numerous other studies which indicate that early life stages of fish are usually more
sensitive to toxicants that larger fish.

Taken together, the weight of evidence suggests that LA in waters of LRC is not causing adverse
effects on resident trout. By extension, effects of LA on fish in the Kootenai River (including
sensitive species such as the white sturgeon and bull trout) are therefore not of concern, since
concentrations of LA in the Kootenai River are substantially lower than in LRC.

Confidence in this conclusion is medium to high. However, observations from the in situ
exposure studies are limited to the conditions and concentration values that occurred during the
studies, and if substantially higher concentrations were to occur in other years, the consequences,
if any, cannot be predicted. While observations from fish population surveys are often variable
between years, results at this site were relatively consistent across two years, so confidence in
these studies is good.

5.0 RISKS TO BENTHIC MACROINVERTEBRATES

Two lines of evidence are available to evaluate effects of site contaminants on benthic
macroinvertebrates, including:

• Laboratory-based site-specific sediment toxicity tests in two species of organism (H.
azteca, and C. tentans)

• Site-specific benthic community population studies, augmented with habitat quality
studies

ES-7

-------
FINAL

The site-specific sediment toxicity tests indicate that effects on growth and reproduction were
not apparent in H. azteca, and were minor in C. tentans. However, an effect of site sediment on
survival was noted in both species, with C. tentans being more impacted (9-25% decrease) than
H. azteca (4-6% decrease). It is difficult to judge if LA is the likely cause, because quantitative
estimates of LA concentration in the two site sediments are sufficiently uncertain that the
presence of a dose-response relationship cannot be ascertained. Even if LA is the cause, the
applicability of these results to other species, and hence the potential magnitude of effects on the
benthic invertebrate community as a whole, are difficult to judge from this line of evidence
alone, and are best determined by evaluating the site specific population studies presented below.

The site-specific population studies suggest that benthic macroinvertebrate communities along
lower Rainy Creek may occasionally rank as slightly impaired compared to off-site reference
locations, but are not impaired compared to upper Rainy Creek. The differences are not
extensive and might be due, at least in part, to differences in habitat quality.

Taken together, these findings support the conclusion that LA contamination in lower Rainy
Creek may be causing small to moderate effects on survival of some species, but the overall
benthic macroinvertebrate community is not substantially impacted.

Confidence in this conclusion is medium to high. One potential limitation to the site-specific
studies is that the test species are not expected to occur in mountain streams, and native species
(mainly mayflies, stoneflies, caddisflies, true flies, and beetle larvae) might have differing
sensitivities. While benthic community and habitat surveys often display considerable variability
between years, in this case the results are relatively consistent between two years, providing
good confidence in the survey results.

6.0 RISKS TO AMPHIBIANS

Two lines of evidence are available to evaluate potential effects of LA on amphibians in OU3:

• A site-specific laboratory-based sediment toxicity test

• A field survey of gross and histologic lesion frequency and severity in amphibians
collected from OU3 and from reference areas

ES-8

-------
FINAL

development is not certain, but assuming the effect is only a lag (as opposed to an actual
cessation of development), it is suspected the effects would likely not be ecologically
meaningful. However, it is plausible that the delay might become important if ponds in high
exposure areas were to dry up during this critical stage of development.

The survey of external and histological lesions in field-collected organisms indicates that lesions
in organisms from OU3 are not more frequent or more severe that in organisms from reference
sites, and that all lesions observed are likely the result of parasitism rather than asbestos
exposure. This supports the conclusion that LA is not causing any external or internal
malformations of concern.

Taken together, these findings support the conclusion that sediments and waters in OU3 are not
likely to be causing any ecologically significant adverse effects on amphibian populations.

Confidence in this conclusion is medium to high. The most significant uncertainty is whether the
apparent delay in the final stages of metamorphosis might be of concern. Further studies would
be need to determine if the apparent lag in final stage development is reproducible, and whether
complete metamorphosis is ultimately achieved in exposed organisms.

7.0 RISKS TO MAMMALS

One line of evidence is available to evaluate risks to mammals from LA contamination in
forested areas near the mine:

• An evaluation of lesion prevalence and severity in mice captured from OU3 compared to
mice from a reference area

This is considered to be a relatively strong line of evidence because a) mice are likely to have
high exposure to LA in duff and soil, b) the area selected for study was at the high end of LA
contamination observed in duff, and c) the mice collected would have been exposed by all
relevant exposure routes (inhalation, ingestion of soil, ingestion of food items).

Although the prevalence or mean severity of some types of lesions was higher in mice from OU3
than the reference area, none of the lesions were judged to be attributable to LA exposure, none
were judged to be associated with significant decrements to overall animal health, and no
evidence of meaningful differences in body size or age of the mice was detected. Based on this,
it is considered likely that LA exposures in OU3 are not causing any ecologically significant
effects on populations of small mammals residing in the forest areas of OU3.

ES-9

-------
FINAL

Confidence in this conclusion is high. However, there are several uncertainties in extrapolation
of the results from this study to other mammals that may be exposed in OU3, including the
following:

•	Larger mammals generally have longer life spans than mice, and consequently might
have higher cumulative exposures than mice. Because effects of inhalation exposure to
asbestos are usually found to be related to cumulative exposure in humans and laboratory
animals (ATSDR 2001), this raises the possibility that risk of effect might be higher in
larger mammals with longer lifespans than mice. However, numerous studies have
shown that while effects of asbestos exposure in humans usually take many years to
develop, the same effects occur in rats and mice within 1-2 years (ATSDR 2001).
Moreover, home range is often much larger for large mammals than small mammals, so
longer-lived species such as deer, elk, bear, lynx, etc., would generally be expected to
spend only a fraction of their lifespan in the impacted areas near the mine, thereby
reducing their tendency for exposure. Although uncertain, there is no compelling
evidence to presume that mammals with longer life spans than mice would likely be more
at risk than mice.

•	The mice that were evaluated were trapped in an area near the mine where concentration
levels of LA in duff are at the high end of the range that has been observed in the forest
area. However, LA levels on the mine site itself are likely higher due to the presence of
LA veins in the ore body as well as in waste rock and tailing deposits onsite.
Consequently, mammals residing in the mined area (as opposed to the forest area around
the mine) may have higher exposures.

8.0 RISKS TO BIRDS

One line of evidence is available to evaluate the effect of LA exposure on birds exposed in OU3:

•	A literature-based evaluation of the relative sensitivity to the effects of inhaled
particulates in birds compared to mammals.

Based on the available information, it is concluded that birds are not more sensitive, and are
probably less sensitive, to the effects of inhaled particulates than mammals. Because a site-
specific study of the effects of LA on small mammals did not detect any evidence for increased
incidence or severity of asbestos-related lesions in the respiratory tract (see above), it is
concluded that ecologically significant adverse effects are not likely to be of ecological concern
in populations of birds exposed to LA in OU3. Although a comparable comparative study was
not attempted with regard to relative sensitivity by the oral exposure route, because no effects
were noted in the gastrointestinal system of mice exposed in OU3, there is no reason to expect
that effects in the gastrointestinal system of birds would be of concern.

ES-10

-------
FINAL

Confidence in this conclusion is medium. However, in the absence of direct studies of birds
from OU3, several possible uncertainties remain including the following:

• The relative LA exposure levels of birds compared to mice in OU3 is not certain. It is
assumed that of the wide variety of bird species that occur in OU3, ground foraging birds
with small home ranges would tend to be most exposed, both by inhalation of fibers
released to air and by ingestion of prey or food items capture in duff or soil. However,
considering that mice are likely exposed nearly continuously in the duff or soil, while
birds are likely to be exposed only while foraging, and would likely have low exposure
while in trees or bushes, it is considered likely that birds are not more exposed, and might
be less exposed, than mice.

• Much of the available information on the relative effects of inhaled particulates in birds is
derived from studies of domestic poultry (chickens, ducks). Respiratory demands in wild
birds may tend to be higher than in domestic fowl, which might tend to increase
exposure. However, wild birds tend to be more robust than domestic fowl, which would
tend to decrease sensitivity. Moreover, the basic physiology of the respiratory system is
the same in both domestic and wild birds, so the conclusion that birds are not likely to be
more sensitive than mammals is considered to be reliable.

9.0 SUMMARY AND CONCLUSION

EPA planned and performed a number of studies to investigate whether ecological receptors in
OU3 of the Libby Asbestos Superfund Site were adversely impacted by the presence of LA in
the environment.

Studies of fish, benthic invertebrates, and amphibians exposed to LA in surface water and/or
sediment revealed no evidence of ecologically significant effects that were attributable to LA.
Likewise, in the terrestrial environment, a study of mice exposed to LA in soil and duff in an
area of high LA contamination revealed no evidence of effects attributable to LA. These studies
indicate that ecological receptors are unlikely to be adversely impacted by LA released to the
aquatic or terrestrial environments by previous vermiculite mining and milling activities.

Although there are some uncertainties and limitations associated with this conclusion, these
uncertainties do not result in significant uncertainty in the overall finding that ecological
receptors in OU3 are unlikely to be adversely impacted by LA released to the environment by
previous vermiculite mining and milling activities.

ES-11

-------
FINAL

1.0 INTRODUCTION

1.1 Purpose of this Document

This document is a Baseline Ecological Risk Assessment (BERA) for Operable Unit 3 (OU3) of
the Libby Asbestos Superfund Site, located near Libby, Montana. The purpose of this BERA is
to describe the likelihood, nature, and extent of adverse effects in ecological receptors exposed to
asbestos in OU3 as a result of releases of asbestos to the environment from past mining, milling
and processing activities at the site. This information, along with other relevant information, is
used by risk managers to decide whether remedial actions are needed to protect ecological
receptors in OU3 from the effects of exposure to mining-related environmental asbestos
contamination. If actions are warranted, the results of the BERA will be used with other relevant
information to assess the appropriate remedial actions needed to protect ecological receptors.

An evaluation of potential ecological risks due to other (non-asbestos) contaminants in OU3 is
presented in a separate report (EPA 2013a).

1.2 Document Organization

In addition to this introduction, this report is organized into the following main sections.

• Section 2 - This section describes the location, history, and environmental setting of
OU3, including information on the nature and extent of asbestos contamination in the
environment.

• Section 3 - This section presents the ecological problem formulation, including the site
conceptual model for exposure to asbestos, the selection of assessment endpoints, and a
description of the measures of effect used to characterize the effects of asbestos exposure.

• Section 4 - This section presents the risk characterization for fish.

• Section 5 - This section presents the risk characterization for benthic macroinvertebrates.

• Section 6 - This section presents the risk characterization for amphibians.

• Section 7 - This section presents the risk characterization for mammals.

• Section 8 - This section presents the risk characterization for birds.

• Section 9 - This section provides citations for all data, methods, studies, and reports
utilized in the BERA.

All tables and figures are presented at the end of the document (following the references).

All site-specific study reports that provide data used in the risk assessment are provided
electronically in Attachment D.

-------
FINAL

2.0 SITE CHARACTERIZATION

2.1 Overview

Libby is a community in northwestern Montana (see Figure 2-1 Panel A) that is located near a
large open-pit vermiculite mine (Figure 2-1 Panel B). The mine began limited operations in the
1920s and was operated on a larger scale by the W.R. Grace Company (Grace) from
approximately 1963 to 1990. Before the mine closed in 1990, Libby produced approximately
70-80% of the world's supply of vermiculite.

Vermiculite is a naturally-occurring silicate mineral that exhibits a sheet-like structure similar to
mica. When heated to approximately 870°C, water molecules between the sheets change to
vapor and cause the vermiculite to expand like popcorn into a light porous material. This
process of expanding vermiculite is termed "exfoliation" or "popping." Both unexpanded and
expanded vermiculite have found a range of commercial applications, the most common of
which include packing material, attic and wall insulation, various garden and agricultural
products, and various cement and building products.

The vermiculite ore deposit at the mine in Libby contains a form of asbestos referred to as Libby
Amphibole (LA). Historic mining, milling, and processing of vermiculite at the Site are known
to have caused releases of vermiculite and LA to the environment. Inhalation of LA is known to
have caused a range of adverse health effects in exposed humans, including workers at the mine
and processing facilities (McDonald et al. 1986a, McDonald et al. 1986b, Amandus and Wheeler
1987, McDonald et al. 2004, Whitehouse 2004, Sullivan 2007, Rohs et al. 2007, Larson et al.
2010a, 2010b, 2012a), as well as residents of Libby (Peipins et al. 2003, Whitehouse et al. 2008,
Larson et al. 2012b, Antao et al. 2012). Exposure to asbestos released to the environment may
also be having adverse effects on aquatic and/or terrestrial wildlife near the mine.

Based mainly on concerns for public health, the U.S. Environmental Protection Agency (EPA)
listed the Libby Asbestos Superfund Site on the National Priorities List in October 2002. Given
the size and complexity of the site, EPA divided the site into a series of Operable Units (OUs).
This document focuses on Operable Unit 3 (OU3), which is defined as follows:

OU3 includes the property in and around the Zonolite Mine owned by W.R. Grace or
Grace-owned subsidiaries (excluding OU2) and any area (including any structure, soil,
air, water, sediment or receptor) impacted by the release and subsequent migration of
hazardous substances and/or pollutants or contaminants from such property, including,
but not limited to, the mine property, the Kootenai River and sediments therein, Rainy
Creek, Rainy Creek Road and areas in which tree bark is contaminated with such
hazardous substances and/or pollutants and contaminants.

-------
FINAL

Because the extent of mine-related contamination in tree bark could not be determined until data
were collected, EPA established an initial study area for OU3, as shown by the red line in Figure
2-2.

2.2 Physical Setting

2.2.1 Topography

The terrain in OU3 is mainly mountainous with dense forests and steep slopes. Based on the
USGS topographic map of the area1, the mined area is at an elevation of about 3,400 to 4,200
feet, and the Kootenai River is at an elevation of about 2,100 feet.

2.2.2 Land Ownership or Stewardship

OU3 is located within the Kootenai National Forest. Current land ownership in the area is
shown in Figure 2-3. Kootenai Development Corporation (KDC), a subsidiary of Grace, owns
about 3,500 acres of land that includes the mine and the surrounding area to a distance of about 1
mile. Land surrounding the KDC property is mainly within the Kootenai National Forest and is
managed by the U.S. Forest Service. Some land parcels are owned by the State of Montana and
some are owned by Plum Creek Timberlands LP for commercial logging. Small areas of private
properties near the southern border of the OU3 study area are included in OU4 rather than OU3.

2.2.3 Climate

Northern Montana has a climate characterized by relatively hot summers, cold winters, and low
precipitation. Figure 2-4 presents temperature and precipitation data collected at the Libby NE
Ranger Station, which is located just west of the town of Libby near the Kootenai River. As
indicated, long-term (100-year) average summer high temperatures (degrees Fahrenheit) are in
the upper 80s, and long-term average low temperatures are in the 40s. Long-term average winter
high temperatures are in the 30s, with average lows less than 20. The western mountain ranges
cause Pacific storms to drop much of their moisture before they reach the area, resulting in
relatively low precipitation, averaging about 18 inches total per year. The most abundant rainfall
occurs in late spring and early summer. In the winter months, snowfall averages 54 inches per
year and snow cover typically remains on the ground from November through March.

A meteorological station was installed at the mine site in January 2007, and data are available for
seven years of monitoring (through December 2013). Figure 2-5 is a wind rose that summarizes
the average speed and direction of winds at the mine over this time interval. As indicated, the

1 http://www.mytopo.com/products/quad.cfm?code=o48115d5

2 http://www.wrcc.dri.edu/cgi-bin/cliMAIN.pl7mtlibb

-------
FINAL

winds blow predominantly (about 45% of the time) from the southwest toward the northeast,
usually at speeds of less than 17 knots (19.5 mph). Winds in the opposite direction (from
northeast to southwest) occur about 15% of the time, usually at speeds less than 10 knots (11.5
mph).

2.2.4 Surface Water Features

The mine is located within the Rainy Creek watershed, an area of approximately 17.8 square
miles. Figure 2-6 shows the main surface water features of OU3. Primary surface water bodies
include:

• Rainy Creek originates between Blue Mountain and the north fork of Jackson Creek at an
elevation of about 5,000 feet, and falls to an elevation of 2,080 feet at the confluence with the
Kootenai River (Zinner 1982). The average gradient for Rainy Creek is about 12% (Parker
and Hudson 1992), and the banks are well vegetated (MWH 2007). The reach of Rainy
Creek that occurs up-gradient of the mine site is referred to as Upper Rainy Creek (URC),
while the reach adjacent to and down-gradient of the mine site is referred to as Lower Rainy
Creek (LRC).

• Fleetwood Creek flows westward along the northern edge of the mined area. The average
stream gradient for Fleetwood Creek is about 11% (Parker and Hudson 1992). Under current
site conditions, Fleetwood Creek flows through a portion of mine waste before discharging
into a large tailings impoundment which was constructed within the former Rainy Creek
channel (see below). A small ponded area was identified along Fleetwood Creek during
reconnaissance surveys by EPA in 2007.

• Carney Creek flows westward along and through mine waste on the southern side of the
mined area before joining Rainy Creek. A small pond is present that was formed when waste
piles were deposited in the drainage and blocked the flow of the creek. The pond is
vegetated on one side. Several small springs are reported along Carney Creek (Zinner 1982).

• Tailings Impoundment. In 1972, Grace constructed a tailings impoundment (also referred to
as the tailings pond) along Rainy Creek to receive tailings produced by a new wet milling
process and to recover water for reuse. The height of the dam which forms the impoundment
is about 135 feet. The impoundment occupies 70 acres. The impoundment receives water
from both upper Rainy Creek and Fleetwood Creek. The impoundment drains through 12 toe
drains directly into lower Rainy Creek, and may also discharge to lower Rainy Creek via an
overflow channel during high flow events (Parker and Hudson 1992).

-------
FINAL

• Mill Pond. A pond in the Rainy Creek channel downstream of the tailings impoundment was
constructed to provide a water supply for mining operations. This pond, sometimes referred
to as the Lower Pond, discharges to Rainy Creek where it mixes with flow from Carney
Creek and flows downstream to the Kootenai River.

• Kootenai River. The Kootenai River flows from east to west along the south side of OU3.
Flows in the Kootenai River are controlled by the Libby Dam, which was constructed in the
late-1960s and early-1970s as part of the Columbia River development for flood control,

power generation, and recreation. Daily water flow from the dam generally ranges from
4,000 to 12,000 cubic feet per second (cfs), with maximum discharge flows in late May/early
June up to 30,000 cfs.

2.3 Current Condition of the Mine Site

Figure 2-7 shows an aerial view of the current condition of the mine site and the main surface
features. As indicated, the mined area was heavily disturbed by the open-pit mining activities,
and some areas remain largely devoid of vegetation. There are a number of areas where mine
wastes have been disposed, including waste rock dumps (mainly on the south side of the mine),
coarse tailings (mainly to the north of the mine), and fine tailings (placed in the tailings
impoundment on the west side of the site). All former buildings and mine works at the site have
been demolished and removed.

2.4 Nature and Extent of LA Contamination at the Site

2.4.1 Mineral Characteristics of LA

Asbestos is the generic name for a group of naturally-occurring silicate minerals that crystallize
in long thin fibers. The basic chemical unit of asbestos is [Si04]"4. This basic unit consists of
four oxygen atoms at the apices of a regular tetrahedron surrounding and coordinated with one
silicon ion (Si+4) at the center. The silicate tetrahedra can bond to one another through the
oxygen atoms, leading to a variety of crystal structures. Different forms of asbestos differ from
each other in their crystal structures, and also in the types of cations that bind to the un-bonded
oxygen atoms along the silicate chains (EPA 2014).

The U.S. Geological Survey (USGS) performed electron probe micro-analysis and X-ray
diffraction analysis of 30 samples obtained from exposed asbestos veins at the mine to identify
the type of asbestos present (Meeker et al. 2003). The results indicated that there were several
mineral varieties of amphibole asbestos present, including winchite, richterite, tremolite, and
magnesioriebeckite. Meeker et al. (2003) noted that, depending on the valence state of iron and

3 http://waterdata.usgs.gov/mt/nwis/inventory?search_site_no=12301933

-------
FINAL

data reduction methods utilized, some minerals may also be classified as actinolite. The EPA
refers to this mixture of amphibole asbestos minerals as Libby Amphibole asbestos (LA).

2.4.2 Concentrations of LA in Environmental Media

As part of the Remedial Investigation (RI) in OU3, Grace and their contractors, working in
cooperation with EPA and with EPA oversight, have collected a large number of environmental
samples and analyzed them for LA. All of the sampling and analytical methods have been
planned in Sampling and Analysis Plans (SAPs) with associated Quality Assurance Project Plans
(QAPPs) and detailed Standard Operating Procedures (SOPs) for sampling and analysis methods.
Consequently, all data collected under these governing SAP/QAPPs/SOPs are considered to be
appropriate for use in the risk assessment, unless otherwise noted.

Overview of Sampling and Analysis Methods

Air and Water

Samples of air and water are typically collected by drawing a known volume of air or water
through a filter, and then examining the filter under a microscope to determine the number of
asbestos fibers in the sample. For studies in OU3, analysis was performed using transmission
electron microscopy (TEM) in basic accord with the counting and recording rules specified in
International Organization for Standardization (ISO) 10312 (ISO 1995). A particle is identified
as an LA fiber if it satisfies the following three criteria:

• Morphology: The particle is elongated with roughly parallel sides, a length >0.5 [j,m, and
an aspect ratio (length/width) >3:1

• Crystallography: The particle has an X-ray diffraction pattern consistent with amphibole
asbestos

• Chemistry: the particle has an energy dispersive X-ray spectrum consistent with known
samples of LA from the mine (SRC 2008)

Results are generally expressed as fibers per cubic centimeter (f/cc) in air, or million fibers per
liter (MFL) in water4. Accuracy and precision of concentration estimates tend to increase as the
number of fibers counted increase.

4 In some samples, fibers occur in complex structures classified as bundles, clusters, or matrix particles. ISO 10312
provides rules for quantifying the contribution of these complex structures to concentration estimates. For
simplicity, the term "fiber" is used here to include not only fibers but the more complex structures as well.

-------
FINAL

Mine Waste. Soil and Sediment

For studies at OU3, samples of mine waste, soil, and sediment were analyzed by polarized light
microscope (PLM) in accordance with Libby-specific SOPs (SRC 2012). Prior to analysis,
samples are sieved and ground to reduce maximum particle size to < 250 [j,m. LA fibers and
particles are identified based on their optical characteristics (color, pleochroism, refractive index,
birefringence, and extinction angle). The microscopist estimates the area fraction of particles in
a field of view that are LA based on a visual comparison of the sample to site-specific standards
with known levels of LA, and this is used as an estimate of the mass fraction. Because the visual
area estimates are largely subjective, this is a semi-quantitative method, and the amount of LA
present is characterized by assigning a semi-quantitative "bin" designation:

Bin

Approximate Range

Non-detect

< 0.2%

0.2% to <1%

> 1%

Samples in Bin C are assigned a quantitative estimate (expressed as mass percent), but these
estimates may not be highly accurate or precise.

Duff and Tree Bark

Samples of duff (forest floor litter) and tree bark were analyzed in accord with SOPs developed
for use at the Libby site (EPA 2012c, 2012d). In brief, samples are prepared for LA analysis by
ashing at high temperature to fully oxidize all organic material. The ashed residue is then
suspended in acid (this helps dissolve residual salts in the residue), diluted as needed, filtered,
and analyzed by ISO 10312, similar to the method used for water. Results are usually expressed
as million fibers per gram dry weight for duff and million fibers per square centimeter of surface
area for tree bark.

Summary of LA Concentration Data

A complete database of LA measurements in environmental media in OU3 is provided in CDM
Smith (2013a). Results of the sampling and analysis efforts are summarized below, stratified by
medium.

-------
FINAL

LA in Ore and Mine Wastes

LA occurs in cross-cutting veins and dikes that occur throughout the deposit. These veins and
dikes generally range from a few millimeters to several meters in thickness, and the LA
concentration in these zones is estimated to range between 50-75% (Meeker et al. 2003).

Concentrations of LA measured in several categories of mine waste s collected from in and about
the mined area (intentionally excluding samples that were judged to be from amphiboles veins)
are summarized in Table 2-1. As indicated, almost all samples contain detectable levels of LA,
ranging from PLM Bin B1 (<0.2%) up to Bin C (>1%). Concentration estimates for the Bin C
samples ranged from 2% to 8%.

LA in Ambient Air Near the Mine

Data on the concentration of LA in ambient air near the mined area were collected at 12
sampling stations (see Figure 2-8). One round of sampling (four sequential 5-day samples) was
collected during the month of October 2007 (EPA 2007), and a second round (8 sequential 5-day
samples) was collected in the interval from July to October 2008 (EPA 2008b). The relatively
long sampling duration (five days) was used to ensure the samples were representative of long
term average concentrations.

Summary statistics are presented in Table 2-2. As shown, LA concentrations were often below
the detection limit (typically about 0.0005 f/cc), with an overall average of about 0.0002 f/cc.
These concentrations are much lower than were present during the time the mine was active,
when concentrations in air often ranged from 1 to more than 100 f/cc (Amandus et al. 1987).
The current low levels in air are presumably due to wind or other disturbances that release fibers
from existing sources (contaminated soil, tailings, waste rock, duff, etc.) into air.

LA in Surface Water

EPA has collected samples of surface water at a number of on-Site locations (Figure 2-9 Panel
A) as well as at two reference streams located several miles west or northwest (cross-wind) of
the mine (Figure 2-9 Panel B). Summary statistics are presented in Table 2-3. As shown, in
lower Rainy Creek (LRC-1 to LRC-6), mean concentrations commonly range from 3 to 44 MFL,
although individual samples may be higher. Generally similar values occur in Fleetwood Creek
and Carney Creek at stations adjacent to the mined area (CC-2, FC-2, CC-Pond), although levels
in FC-Pond may be somewhat higher (81 MFL). In upper Rainy Creek, concentrations are
generally low in the upstream portions (URC-1 and URC-1 A), although elevated concentrations
have occasionally been observed at URC-2 (this is below a mine roadway constructed in part of
mine wastes). LA is generally non-detect or very low in reference creeks and ponds.

8

-------
FINAL

Concentrations in the Kootenai River are generally low, with little apparent difference between
samples collected upstream and downstream of the confluence with Rainy Creek.

The concentrations of mining-related contaminants in surface waters near the mine site are not
constant over time, but tend to vary as a function of flow rates, especially the high flows
typically associated with the spring runoff. Figure 2-10 shows the concentrations of LA
measured at four stations along lower Rainy Creek in 2008 as a function of time of year. As
shown, an increase in concentration was observed during the spring runoff at three of the four
stations. Similar increases (of a smaller magnitude) were also noted in Fleetwood Creek and
Carney Creek. The reason that no increase was detected at LRC-2 is not known, but might be
due to the effect of the Mill Pond which is located a short distance upstream.

LA in Sediment

EPA has collected samples of sediment at a number of locations, typically the same as those
where surface water samples were collected (see Figure 2-9). Summary statistics on bin
assignments are presented in Table 2-4. As shown, essentially all sediment samples from Lower
Rainy Creek, Fleetwood Creek, and Carney Creek, as well as from the tailings impoundment and
the Mill Pond, contain detectable levels of LA. The highest frequency of high concentration
samples (Bin C) were observed in Carney Creek (which flows adjacent to and downhill of the
mined area) and in Rainy Creek just below the tailings impoundment dam (TP-TOE1 and TP-
TOE2). Levels in lower Rainy Creek were mainly Bin B1 (<0.2%) or Bin B2 (0.2 to 1%),
although several Bin C samples (>1%) were observed. Quantitative estimates for the 62 Bin C
samples range from 1% to 10%, with an average of 3%. As noted above, estimates of LA in
sediment are semi-quantitative.

Sediment samples from the upper reaches of Upper Rainy Creek (URC-1 and URC-1 A) appear
to contain little LA, with 5 of 6 being Bin A (non-detect), although one sample was ranked as
Bin B1 (<0.2%). Samples from URC-2 do appear to have low levels (mainly < 0.2%). The
source of this LA is uncertain, but might either be mining-related or natural levels eroding from
the ore body.

Samples of sediment from off-site reference areas and ponds did not contain any detectable
levels of LA.

LA in Forest Soil, Duff, and Tree Bark

EPA has collected samples of forest soil, duff, and/or tree bark at a variety of distances and
directions from the mine (CDM Smith 2013a, 2013b, 2013c, 2014; EPA 2012b). Several
samples have also been collected in the area of Souse Creek by the U.S. Public Health Service

-------
FINAL

(USPHS 2013). The sampling locations and the resulting LA concentration values are shown in
Figure 2-11. In this figure, the results at each station are indicated in a triangular set of symbols:

Top symbol = soil

Bottom left symbol = tree bark

Bottom right symbol = duff

An "x" indicates that no data for that media type are available for that location, while a grey
circle indicates the sample was non-detect. Detects are indicated as colored circles, with low
values in green, medium values in yellow, and high values in red.

As illustrated, the highest concentrations tended to occur close to the mine, mainly in the primary
downwind (northeast) direction, although some high values were also detected in the secondary
downwind direction (to the southwest). Although there is moderate variability in the
measurements, concentrations in all three media tend to decrease as a function of distance from
the mine. This tendency is shown more clearly in Figure 2-12, which plots concentrations in duff
and tree bark as a function of distance from the mine. As illustrated, concentrations tend to
decrease exponentially as a function of distance from the mine.

2.5 Ecological Setting

2.5.1 Terrestrial Setting

The mined area was heavily disturbed by past mining activity and some areas remain largely
devoid of vegetation. Outside the mined area, most of OU3 is forested, with only 4% of the land
being classified as non-vegetated (USDAFSR1 2008). Data for the Kootenai National Forest
indicate Douglas-fir forest type is the most common, covering nearly 35% of the National Forest
land area within OU3. Next in abundance are the lodgepole pine forest and spruce-fir forest
types at 17% each, and the western larch forest type at 11%. Other tree species reported in the
area are the Black Cottonwood (Populus trichocarpa), Quaking Aspen (Populus tremuloides),
Western Paper Birch (Betulapapyrifera) and Pacific Yew (Taxus brevifolia) (USDAFSR1
2008).

The forested area of OU3 is suitable habitat for a wide range of terrestrial species, including
mammals, birds, and reptiles. In order to identify wildlife species likely to occur in OU3, data
available from the Montana National Heritage Program (MNHP) was consulted. First, using the
MNHP Animal Tracker web page (http://nhp.nris.mt.gov/Tracker/), all species known to occur
within Lincoln County, Montana, were identified. Next, the MNHP and Montana Fish, Wildlife
and Parks Animal Field Guide (http://fieldguide.mt.gov/) were consulted to determine if a
particular species has been observed in the vicinity of OU3. Species not identified within the

10

-------
FINAL

vicinity of 0U3, and those not expected to occur at OU3 based on a consideration of available
habitat, were removed. The species that remained are listed in Attachment A, along with
information on general habitat requirements, habitat type for foraging and nesting, feeding guild,
typical food, migration and hibernation, longevity, home range, and size. The species identified
as residing all or part of the year within OU3 include 29 invertebrates (26 terrestrial and three
aquatic), seven amphibians, seven reptiles, 175 birds, and 48 mammals.

2.5.2 Aquatic Setting
Rainy Creek Watershed

Within the Rainy Creek watershed there are streams and ponds that provide habitat for a range of
aquatic species including fish, invertebrates, and amphibians. Species identified during site-
specific ecological population surveys performed as part of the RI at OU3 are summarized in
Section 4.3 (fish), Section 5.3 (benthic macroinvertebrates), and Section 6.3 (amphibians). In
brief, fish surveys performed in OU3 streams indicate that the most common species of fish are
westslope cutthroat trout (Oncorhynchus clarkii lewisi), rainbow trout (Oncorhynchus mykiss),
and "cutbow" trout (a rainbow/cutthroat hybrid). Brook trout {Salvelinus fontinalis) were not
observed in OU3, but were observed in nearby reference streams. Aquatic invertebrate
community surveys in OU3 indicate that the most common types of aquatic invertebrates
observed include mayflies, stoneflies, caddisflies, true flies, and beetle larvae. The most
common amphibian species observed are the northern tree frog (Pseudacris regilla), Columbia
spotted frog (Rana luteiventris), and western toad (Bufo boreas).

Kootenai River

No site-specific studies of aquatic receptors in the Kootenai River have been performed as part
of the OU3 RI. However, EPA's Environmental Monitoring and Assessment Program (EMAP)
has collected aquatic community data at a station on the Kootenai River about one mile
downstream of the confluence with Rainy Creek. This location was sampled in August 2002.
Forty-four species of aquatic invertebrates have been observed, including oligochaetes, insects
(diptera, ephemeroptera, trichoptera and hemiptera), coelenterates (hydra), mollusks, and
nematodes. Eleven species of fish were observed, including mountain whitefish (Prosopium
williamsoni), rainbow trout, sockeye salmon (Oncorhynchus nerka), cutthroat trout, bull trout
{Salvelinus confluentus), and several species of forage fish (dace, shiner, sculpin).

2.5.3 Federal and State Species of Special Concern

Table 2-5 lists the animal and plant species currently identified by the U.S. Fish and Wildlife
Service (USFWS) as being of Federal concern in the Kootenai Nation Forest (USFWS 2014).

-------
FINAL

Table 2-6 lists species currently listed by the Montana Natural Heritage Program (MNHP) as
being of concern to the state that occur in the general area of OU3 (Montana Township 3 IN,
Range 30W) (MNHP 2014). Based on an evaluation of habitat requirements, the following
listed species are considered to be the most likely to occur in OU3:

Federal

•	Bull Trout (Sa/ve/inus confluentus)

•	White Sturgeon (Acipenser transmontanus) (Kootenai River only)

•	Grizzly Bear (Ursus arctos horribilis)

•	Canada Lynx (Lynx canadensis)

State

•	Coeur d'Alene Salamander (Plethodon idahoensis)

•	Boreal Toad, Green (also known as Western Toad) (Bufo boreas)

•	Flammulated Owl (Otus flammeolus)

•	Northern Goshawk (Accipiter gentilis)

•	Bull Trout (Salvelinus confluentus)

•	Torrent Sculpin (Cottus rhotheus)

•	Westslope Cutthroat Trout (Oncorhynchus clarkii lewisi)

•	Canada Lynx (Lynx canadensis)

The Kootenai River is ranked as critical habitat for the bull trout, and the north-central portion of
the OU3 study area includes critical habitat for the Canada lynx.

12

-------
FINAL

3.0	PROBLEM FORMULATION

Problem formulation is a systematic planning step that identifies the major concerns and issues to
be considered in an ecological risk assessment, and describes the basic approaches that will be
used to characterize ecological risks that may exist (EPA 1997). As discussed in EPA (1997),
problem formulation is generally an iterative process, undergoing refinement as new information
and findings become available.

3.1	Conceptual Site Model

A Conceptual Site Model (CSM) is a schematic summary of what is known about the nature of
source materials at a site, the pathways by which contaminants may migrate through the
environment, and the scenarios by which ecological receptors may be exposed to site-related
contaminants. When information is sufficient, the CSM may also indicate which of the exposure
scenarios for each receptor are likely to be of greatest potential concern, and which (if any) are
likely to be sufficiently minor that detailed evaluation is not needed. This diagram is generally
prepared at the start of the risk assessment process, and is used to help identify the types of
studies and data collection efforts that are likely to be useful in evaluating ecological risks at the
site.

Figure 3-1 presents the CSM that was developed for exposure of ecological receptors to LA in
OU3. The following sections provide a more detailed discussion of the main elements of this
CSM.

3.1.1	Potential Sources of Contamination

In the past (when the mine was operating), vermiculite mining and milling activities resulted in
releases of LA fibers to air as well as the generation of various types of LA-containing solid
waste. Fibers released to air would have been carried downwind in air (mainly to the northeast),
followed by deposition of the fibers to soil or duff, or entrapment in tree bark. Various solid
wastes that were generated during mining and milling operations (waste rock, waste ore, and
tailings) were deposited on-site.

3.1.2	Migration Pathways in the Environment

On-site solid wastes that remain at the site may be a source of on-going release of asbestos to the
environment by two main pathways:

13

-------
FINAL

Airborne Transport. Asbestos fibers that are present in solid wastes may become
suspended in air as the result of various types of disturbances, including wind action and
mechanical disturbances caused by human activities (vehicle traffic, operation of heavy
machinery, etc.). Once airborne, suspended fibers move with the wind and then settle
and become deposited onto surface soils, tree bark, and duff.

Erosion. Asbestos that is present in on-site solid wastes may be carried in surface water
runoff (e.g., from rain or snowmelt) into local streams (especially Fleetwood Creek,
Carney Creek and Rainy Creek below the tailings impoundment), resulting in
contamination of waters and sediments in the streams.

3.1.3 Potentially Exposed Ecological Receptors

As discussed in Section 2.3, there are a large number of ecological species that are likely to
occur in OU3 and that could be exposed to mine-related contaminants. However, it is generally
not feasible or necessary to evaluate risks to each species individually. Rather, it is usually
appropriate to group receptors with similar behaviors and exposure patterns, and to evaluate the
risks to each group.

For aquatic and semi-aquatic receptors, organisms are often evaluated in four groups:

Fish

Benthic macroinvertebrates
Amphibians (aquatic life stages)

Aquatic plants

For terrestrial receptors, organisms are often grouped into the following broad categories:

Birds
• Mammals

Terrestrial plants
Soil invertebrates
Reptiles

3.1.4 Exposure Pathways of Chief Concern

Most ecological receptors are likely to be exposed to LA in the environment by several pathways
(ingestion, inhalation, and/or direct contact), but not all scenarios are equally likely to be of
concern and not all require equal levels of investigation. In Figure 3-1, solid circles identify the
pathways that were judged to be of greatest potential concern in term of exposure potential.

-------
FINAL

Open circles identify exposure pathways that are likely to be complete, but are considered likely
to be have low exposure or risk potential. Open boxes identify exposure pathways that are
judged to be incomplete, negligible, or not applicable. The rationale for these judgments is
summarized below.

Fish

The primary exposure pathway of concern for fish is direct contact with asbestos fibers
suspended in surface water. Fish may also be exposed to asbestos by incidental ingestion of
sediment while feeding, ingestion of contaminated prey items, and direct contact with sediment.
Incidental ingestion of sediment is likely to be a minor source of exposure, especially for fish
(e.g., trout) that feed mainly in the water column. Likewise, ingestion of prey items is likely to
be minor because asbestos is not expected to bioaccumulate in food web items. Direct dermal
contact with sediment is also likely to be minor, at least for fish that reside mainly in the water
column.

Studies that were performed to evaluate these potential exposure pathways of fish are described
in Section 4.0.

Benthic Invertebrates

The exposure pathways of primary concern for benthic invertebrates that reside in stream
sediment are direct contact with sediment and with sediment porewater. For organisms that
reside in the uppermost layers of the sediment, exposure may also include surface water flowing
over and through the sediment. In addition, benthic organisms may be exposed by ingestion of
fibers while feeding in the sediment. For this type of organism, distinguishing between direct
contact with sediment and ingestion exposure is often not possible, so these pathways are often
evaluated together.

Studies that were performed to evaluate these potential exposure pathways of benthic
macroinvertebrates are described in Section 5.0.

Amphibians

Amphibians (e.g., frogs, toads) inhabit both aquatic and terrestrial (mainly riparian)
environments, with early life stages being primarily aquatic and later life stages being semi-
aquatic or terrestrial. In their aquatic life stages, the exposure pathways most likely to be
significant are direct contact with surface water and sediment. As for fish, exposure by ingestion
of sediments and/or prey items may also occur, and studies of this exposure scenario usually
include both pathways. Numerous studies suggest that aquatic early life stages are usually more

-------
FINAL

susceptible to environmental contaminants than older life stages, so exposures of adult
amphibians in the terrestrial environment is likely to be of lesser concern than the exposures that
occur during development in the aquatic environment.

Studies that were performed to evaluate these potential exposure pathways of amphibians are
described in Section 6.0.

Aquatic and Terrestrial Plants

Aquatic plants might be exposed to LA both by direct contact of foliage with fibers in surface
water contact of roots with fibers in sediment. Similarly, terrestrial plants may be exposed to
asbestos mainly by direct contact of roots with fibers that have been deposited into soil, or by
deposition of airborne fibers onto bark or foliar surfaces. Because asbestos exists as solid fibers
that are not likely to be taken up into plant tissues, either by foliar contact or through roots, it is
not expected that asbestos contamination is of concern for either aquatic or terrestrial plants.

Consequently no studies were planned to evaluate impacts of LA on plants.

Mammals and Birds

Mammals and birds may be exposed to asbestos by ingestion of contaminated soils, surface
water, sediment, and food, and by inhalation when feeding or foraging activities result in the
disturbance of asbestos-contaminated soils, sediments, or duff. Studies in humans and laboratory
animals indicate that inhalation exposures are likely to be the main exposure route of
toxicological concern, with oral exposure often tending to cause few significant effects (ATSDR
2001). However, there are some reports of potential effects of asbestos following oral exposure
in mammals (see Section 4.1), so oral exposure is indicated by a solid circle in Figure 3-1, both
for mammals and for birds.

Direct contact (dermal exposure) of birds and mammals to fibers in soil or other contaminated
media may occur, but this exposure route is suspected to be or minor concern, since asbestos is
expected to remain mainly on the surface of fur or feathers and is not expected to cross the skin
barrier.

Studies that were performed to evaluate exposures of mammals are described in Section 7.0, and
an evaluation of potential hazards to birds is discussed in Section 8.

16

-------
FINAL

Soil Invertebrates

Soil invertebrates (e.g., worms) may be exposed by direct contact with fibers in soil, and also by
ingestion of soil detritus that contains fibers. While the likelihood of LA effects on worms is not
known, it was considered likely that benthic macroinvertebrates would have higher exposure
than terrestrial invertebrates because concentrations of LA are generally higher in OU3
sediments than in soils. Based on this, no studies of earthworms or other terrestrial invertebrates
were planned or performed.

Reptiles

Turtles have been observed in OU3 ponds, and other types of reptiles (snakes) are also present in
OU3. These organisms may be exposed to site-related contaminants by direct contact and
ingestion of water or sediment, and by ingestion of prey items. While the likelihood of LA
effects on reptiles is not known, it was considered likely that amphibians would be more at risk
than reptiles, especially considering that reptilian skin is covered in scales that would be
expected to decrease exposure from direct contact pathways. Based on this, no studies of reptiles
were planned or performed.

3.2 Management Goal and Assessment Techniques

3.2.1 Management Goal

A management goal is a statement of the basic objectives that the risk manager wishes to achieve
at a site. The overall management goal identified for ecological health at the Libby OU3 site for
asbestos contamination is:

Ensure adequate protection of ecological receptors within the Libby OU3 Site from the
adverse effects of exposures to mining-related releases of asbestos to the environment.

For most species, "adequate protection" is generally defined as the reduction of risks to levels
that will result in the recovery and maintenance of healthy local populations and communities of
biota (EPA 1999). For the Libby OU3 Site, the assessment populations are defined as the
groups of organisms that reside in locations that have been impacted by mining-related releases.
For exposure to asbestos, this is believed to include the mined area and the drainages associated
with the mined area, as well as surrounding forest lands that were impacted by airborne releases
of asbestos.

For threatened or endangered species, "adequate protection" is generally interpreted to mean
minimizing risks to individual members of the population.

-------
FINAL

3.2.2	Assessment Endpoints

Assessment endpoints are explicit statements of the characteristics of the ecological systems that
are to be protected. Because the risk management goals are formulated in terms of the protection
of populations and communities of ecological receptors, the assessment endpoints selected for
use in this problem formulation focus on endpoints that are directly related to the management
goals. This includes:

•	Mortality

•	Growth

•	Reproduction

If effects on these three assessment endpoints are absent or minimal, it is likely that ecologically
significant effects will not occur.

3.2.3	Measures of Effect

Measures of effect are quantifiable ecological characteristics that can be measured, interpreted,
and related to the valued ecological components chosen as the assessment endpoints (EPA 1997,
1998). There are a number of alternative measures of effect that may be investigated as part of
an ecological risk assessment. The primary alternative strategies for characterizing measures of
effects are described below.

Hazard Quotients

For most environmental contaminants, the first line of investigation is usually the Hazard
Quotient (HQ) Approach. A Hazard Quotient (HQ) is the ratio of the estimated exposure of a
receptor to a "benchmark" exposure that is believed to be without significant risk of unacceptable
adverse effect:

HQ = Exposure / Benchmark

If the site exposure does not exceed the benchmark (HQ < 1), it is usually concluded that site-
related exposures are of low concern.

However, there are no established benchmarks for the evaluation of ecological receptors to any
form of asbestos, and most of the studies that are available that might potentially serve as a basis
for development of a benchmark are based on studies of chrysotile asbestos rather than
amphibole asbestos. In particular, there are no studies on the toxicity of LA on any class of

18

-------
FINAL

ecological receptors. Consequently, HQ values were not calculated for any exposure scenario,
and ecological investigations of the potential effects of LA on ecological receptors in OU3
focused on other measures of effect, as discussed below.

Site-Specific Toxicity Tests

Site-specific toxicity tests measure the response of receptors that are exposed to site media. This
may be done either in the field {in situ) or in the laboratory using media collected from the site.
The chief advantage of either type of study is that site-specific conditions that can influence
toxicity are usually accounted for, and that the cumulative effects of all exposure pathways to the
medium and all contaminants in the medium are evaluated simultaneously. One potential
limitation of this approach is that, if toxic effects are observed to occur when test organisms are
exposed to site media, it may not be possible to specify which contaminant or combination of
contaminants is responsible for the effect without further testing or evaluation. A second
limitation is that it may be difficult to perform tests on site samples that reflect the full range of
environmental conditions which may occur in the field across time and space, so it may not be
possible to fully identify all conditions that are and are not of concern.

Population and Community Demographic Observations

Another approach for evaluating possible adverse effects of environmental contamination on
ecological receptors is to make direct observations on the receptors in the field, seeking to
determine whether any receptor population has unusual numbers of individuals (either lower or
higher than expected), or whether the diversity (number of different species) of a particular
category of receptors is different than expected. The chief advantage of this approach is that
observation of community status relate directly to the management goal (protection of
populations). However, there are also a number of important limitations to this approach. The
most important of these is that both the abundance and diversity of a receptor depend on many
site-specific factors (habitat suitability, availability of food, predator pressure, natural population
cycles, meteorological conditions, etc.), and it is often difficult to know what the expected (non-
impacted) abundance and diversity should be in a particular area. This problem is generally
approached by seeking an appropriate "reference area" (either the site itself before the impact
occurred, or some similar site that has not been impacted), and comparing the observed
abundance and diversity in the reference area to that for the site. However, it is sometimes
difficult to locate reference areas that are a good match for all important habitat characteristics.

In-Situ Measures of Exposure and Effects

An additional approach for evaluating the possible adverse effects of environmental
contamination on ecological receptors is to make direct observations on receptors in the field,

19

-------
FINAL

seeking to determine if individuals residing in areas of contamination have an increased
frequency and/or severity of physiological lesions and/or deformities compared to organisms
residing in uncontaminated reference areas. This method has the advantage of integrating most
(if not all) factors that influence the true level of exposure and toxicity of contaminants in the
field. However, if an increased incidence or severity of lesions is observed, it may not be
possible to identify with certainty which environmental contaminant(s) is (are) responsible, and
it may also be difficult to determine with confidence whether the observed lesions are likely to
cause an ecologically significant population-level impact.

Weight of Evidence Evaluation

As noted, each of these alternative strategies for characterizing ecological risks has some
advantages and some limitations. Because of this, it is generally desirable to obtain information
using two or more alternative strategies, and to seek to reach a weight of evidence conclusion
that considers the strengths and limitations of each available line of evidence, including the
magnitude and statistical significance of any observed effects. If two or more lines of evidence
are available, and if the lines of evidence are in general agreement, then confidence in risk
conclusions is increased. If two or more lines of evidence do not agree, then careful attention
must be given to likely reasons for the disparity, and to decide which line(s) of evidence provide
the highest confidence.

3.2.4 Statistical Methods

When appropriate, statistical tests were used to help evaluate the data obtained from the site-
specific studies performed in OU3. In studies where there are replicates for the various
treatments [e.g., two or more measurements at a station, two or more stations within a category
("Site" and "Reference")], there are often several options for performing statistical tests. In
general, if the differences between replicates within a station and/or between stations within a
category are small, it is often useful to combine the data to increase statistical power. However,
even in cases where there may be differences between replicates or stations within a category, it
may still be useful to group the data by category, assuming that risk management decisions are
more likely to be made on a category basis (Site vs Reference) than on a station-by-station basis.

The choice of the most appropriate statistical test(s) depends on the nature of the measurement
endpoints. For studies that measure a discrete endpoint (e.g., mortality), there are two basic
options that may be illustrated using the following hypothetical data:

20

-------
FINAL

Replicate

Site

Reference

N

Dead

Rate

N

Dead

Rate

A

10

3

30.0%

15

1

6.7%

B

10

1

10.0%

15

4

26.7%

C

10

4

40.0%

15

2

13.3%

The first approach treats an individual organism as the unit of observation, combining the data
across replicates, and comparing the rates between categories using a one-tailed Fisher exact test
(FET):

Option 1: By Organism

Category

N

Dead

Rate

FET
p value

Site

30

8

26.7%

0.188





Reference

45

7

15.6%



The second option treats each replicate as the unit of observation and compares the mean of the
rates between categories using a one-tailed t-test:

Option 2: By Replicate

Category N Mean Stdev

t-test
p value

Site 3 26.7% 15.3%
Reference 3 15.6% 10.2%

0.181

Each approach has some statistical advantages and potential limitations, and each may provide
useful information. Option 1 has the advantage of large sample sizes (which helps increase
statistical power) and does not make any assumptions about distributional form. Option 2 avoids
pseudoreplication that may occur if unusual conditions occur in one exposure chamber compared
to the others within the category, but sample size is small and a normal distribution of the means
is assumed.

For continuous measurement endpoints (e.g., fish density, benthic organism growth, lesion
severity, etc.), the preferred test is usually a one-tailed Wilcoxon rank sum (WRS) test (also
known as the Mann Whitney test). This test evaluates whether measurements from one
population consistently tend to be larger (or smaller) than those from the other population. The
test is non-parametric, so it is not necessary to make any assumptions about the distributional
form of the individual measurements.

21

-------
FINAL

3.2.5 Data Evaluation

In general, a three step process was used to evaluate the results of the studies performed, as

follows:

1)	Was a difference observed? This question was assessed mainly by considering the results of
the statistical test(s) used to compare the magnitude and/or severity of the effect in organisms
exposed to OU3 media to that for organisms exposed to reference media. The likelihood that
an observed difference is due to treatment (i.e., exposure to an OU3 medium) rather than
random variation may be judged by the statistical p value. A p value reflects the probability
of obtaining the observed difference between site and reference if the null hypothesis (site <
reference) is true. The smaller the value of p, the less likely it is that the site and reference
are the same and the observed effect occurred by chance. In many cases, a difference is not
considered to be "significant" unless the p value is 0.05 or smaller (i.e., there is no more than
a 5% chance the observed difference occurred at random). However, as discussed in EPA
(2002), while use of a p value of 0.05 as the criterion for "significant" effect ensures that any
effect that is identified as significant has a high probability (>95%) of being treatment-
related, this criterion also runs the risk that some real effects may be overlooked, which
increases the chances of a Type I decision error (deciding the site is not impacted, when it
really is). For example, in the case of a p value of 0.07, this would not be considered
"significant" even though there is a 93% chance the effect is due to treatment. For this
reason, EPA (2002) recommends that when the null hypothesis is "Ho: site < reference", a p
value of < 0.20 should be used to define "significant", and this approach has been used in this
risk assessment.

However, use of a p value of < 0.20 to help minimize risk of a Type I decision error does not
necessarily mean that an effect with a p value of 0.01 and an effect with a p value 0.19 are
equally likely to be treatment-related. Rather, confidence tends to decrease as a continuous
(rather than discrete) function of increasing p. For this reason, in this risk assessment, while
all effects with p value < 0.20 are considered "significant", a distinction in confidence is
indicated by the use of the phrase "statistically significant" to describe differences with p
values < 0.05, and the phrase "marginally significant" to characterize effects with p values of
0.06 to 0.20.

2)	If a difference was observed, is exposure to LA the cause of the effect? While a low p value
indicates that an observed difference is likely to be due to differences between site and
reference exposure conditions, it does not necessarily prove that LA in site media is the cause
of the effect. This is because there may be several differences (other than the

22

-------
FINAL

presence/absence of LA) between site and reference. The question of causality is generally
evaluated by considering the following:

a. Is the nature of the observed difference characteristic of the known effects of asbestos
on the exposed organisms?

b. Does the magnitude or severity of the effect appear to depend on the level of LA?

c. Are there other recognizable differences (e.g., habitat factors) that might explain
some or all of the observed difference?

3) If a difference was observed, and it is or might reasonably be attributed to LA, is the effect
ecologically significant? In a well-designed and well-performed study, small differences
may sometimes be detected and declared to be "significant". However, statistical
significance does not necessarily imply that a difference is of ecological significance. For
example, small decreases in the hatch rate of trout eggs might not lead to a meaningful
difference in the number of trout surviving to adulthood, since only a small fraction of fish
survive to maturity even under normal conditions (Toll et al. 2013). Likewise, increased
prevalence of mild external or internal lesions that would not impair the ability of an
organism to survive, grow, and reproduce would be unlikely to be of concern. An evaluation
of ecological significance is generally based largely on professional judgment, considering
the observed magnitude, nature, and severity of the effect to estimate the expected
consequences of the effect.

3.3 Role of the BTAG

All studies to investigate the potential effects of LA on ecological receptors in OU3 were
planned by EPA working in close cooperation with the Libby OU3 Biological Technical
Assistance Group (BTAG). The BTAG included technical and managerial representatives from
all of the stakeholders at the site including:

• The U. S. EPA Region 8

• The U.S. EPA headquarters Environmental Response Team (ERT)

• The U.S. Fish and Wildlife Service (USFWS)

• The Montana Department of Environmental Quality (MDEQ)

• W.R. Grace and Co., represented by Remedium Group

Input from the various members of the BTAG was used to strengthen the design of all studies
that were performed, thereby maximizing the probability of deriving scientifically reliable data,
taking costs and feasibility into account.

-------
FINAL

4.0 RISKS TO FISH

4.1 Reported Effects

Adverse effects in fish resulting from exposure to asbestos have not been extensively studied, but
several relevant reports were located. Brief summaries are presented below.

• Woodhead et al. (1983) exposed Amazon mollies (Poecilia formosa) to 0.01-10 mg/L of
chrysotile asbestos for six months. Epidermal hypertrophy and necrosis were observed in
kidney and gill in many of the fish at exposure concentrations of 0.1 mg/L or higher, and
in heart at a concentration of 1 mg/L in three of 20 fish. No adverse effects were
observed in liver, muscle or skin.

• Belanger (1985) studied the effects of chrysotile asbestos on adult and juvenile fathead
minnows (Pimephales promelas). Neither adult nor juvenile minnows suffered acute
toxicity at concentrations up to 1E+06 MFL or differential mortality relative to controls
up to 100 MFL for 30 days. Length, weight, and swimming performance of adult
minnows exposed to asbestos were not significantly affected relative to controls.

Juvenile minnows exposed to 1-100 MFL had significantly lower weight.

• Belanger et al. (1990) studied the effects of chrysotile asbestos at concentrations of 0,
0.0001, 0.01, 1, 100 or 10,000 MFL on egg and larval Japanese Medaka (Oryzias
latipes). Eggs were exposed to chrysotile until hatching (13-21 days) and larval were
exposed for thirteen weeks. Exposure of eggs to concentrations of 1 MFL or higher
tended to delay hatching, but egg survival (hatching success) was not grossly or
significantly impaired. Larval Medaka experienced growth reduction at concentrations of
1 MFL or higher. Fish exposed to 10,000 MFL suffered 100% mortality by 56 days.

Fish exposed to 1 MFL or higher developed thickened epidermal tissue. Concentrations
of chrysotile as low as 0.01 MFL tended to reduce successful spawns per female and eggs
per females, although the differences in eggs per female were not statistically significant.

• Belanger et al. (1986a) exposed coho salmon (Oncorhynchus kisutch) and green sunfish

(Lepomis cyanellus) to chrysotile asbestos at concentrations of 1.5 or 3 MFL. Coho were
exposed for 40 or 86 days, while sunfish were exposed for 52 or 67 days. No treatment-
related increases in mortality were detected. Coho larvae exposed to 1.5 MFL were
significantly more susceptible to an anesthetic stress test, becoming ataxic and losing
equilibrium faster than control fish. Two of 106 coho larvae exposed at 3 MFL developed
tumorous swellings in the gill region and 3 additional fish developed coelomic distentions
leading to death. Larval coho and juvenile green sunfish exposed to 3.0 MFL had
epidermal hypertrophy superimposed on hyperplasia, necrotic epidermis, lateral line
degradation, and lesions near the branchial region. Lateral line abnormalities were
associated with a loss of the ability to maintain normal orientation in the water column.

-------
FINAL

No studies were located on the toxicity of LA to fish.

4.2 Site-Specific Toxicity Tests

The EPA, working in concert with the Libby OU3 BTAG, determined that site-specific studies of
the toxicity of LA-contaminated water would provide one valuable line of evidence to evaluate
risks to fish in OU3. Several alternative study designs were pursued. However, all attempts to
expose fish to LA under laboratory conditions were judged to be unsuccessful because of a
tendency for LA to form clumps and bind to bottles, tubing, and aquaria walls, as described in
Attachment B. Based on these difficulties in exposing fish to controlled levels of LA under
laboratory conditions, EPA and the BTAG decided the best alternative strategy was to evaluate
the toxicity of site waters to fish using an in situ exposure design. Because toxicity of water-
borne chemicals to fish may depend on the age of the fish exposed (with early life stages often
tending to be more sensitive that older life stages), two separate in situ studies were planned,
with the first focusing on trout that were exposed from the eyed egg stage through hatching and
alevin swim-up, and the second focusing on juvenile trout. These studies are described below.

4.2.1 In Situ Eyed Egg and Alevin Exposure Studies

An initial study to investigate the effect of in situ exposure of eyed eggs and hatched alevins was
planned and performed in 2012. Detailed descriptions of the study design and the results are
presented in Golder (2013). However, as discussed in Golder (2013), this study was complicated
by the fact that a number of organisms went missing during the study, and the conclusions of the
study depended strongly on what was assumed about the survival status of these missing
organisms. If missing organisms were excluded from the evaluation, or if it were assumed that
most missing organisms did not survive, then the data suggested that effects of exposure might
be important. In contrast, if it were assumed that most missing organisms did survive but
escaped, then the data suggested that any effects of in situ exposure would likely not be
important.

Because of the uncertainty in the 2012 eyed egg study resulting from the missing organisms,
EPA and the BTAG decided that a repeat of the study was necessary, taking care to make
changes to minimize the problems encountered in the first study. The design and results of the
repeat study are reported in Golder (2014b), and the main findings are summarized below.

Study Design

The data quality objectives (DQOs) for the study are presented in Section 5 of the Phase V, Part
B SAP/QAPP (EPA 2012a), and the detailed study protocol is presented in Appendix A.3 of the

-------
FINAL

SAP/QAPP. Changes that were implemented in 2013 to minimize problems encountered in 2012
are summarized in an addendum to the Phase V Part B SAP (EPA 2013b).

In brief, eyed eggs from native westslope cutthroat trout were obtained from the Montana Fish,
Wildlife, and Parks (MFWP) fish hatchery in Anaconda, Montana. The hatchery carefully
inspected all eggs and eliminated any that were observed to be cloudy, have no eyes, or have
"double eyes".

Eggs were placed in Whitlock-Vibert boxes (30 eggs per box). As illustrated in Figure 4-1,
Whitlock-Vibert boxes contain small chambers in the upper portion of the box to house the eggs.
After the eggs hatch and after some of the yolk sac has been absorbed, the larval fish fall from
the upper egg chamber into a lower protected "nursery" chamber where they rest on the bottom
until they develop to the swim-up stage (yolk fully resorbed). Although unaltered boxes allow
alevins to escape after swim-up, for this study, each box was modified by attaching rigid plastic
mesh (100 openings per in ) to the inside of each box, using zip-ties to ensure a secure fit. This
prevented the escape of the swim-ups and also provided protection from predators.

Exposure Locations

A total of six Whitlock-Vibert boxes were placed in lower Rainy Creek (LRC), with two boxes
each at stations LRC-2, LRC-4, and LRC-5. Likewise, a total of six boxes were placed into
reference streams, with three boxes each at upper Rainy Creek (URC) station URC-2 and in
Noisy Creek (NSY). These locations are shown in Figure 2-9.

In addition, one "dummy" box (i.e., one that did not contain any organisms) was placed in the
stream bed at each sampling station between the two boxes with organisms. This "dummy" box
was fitted with a sampling port (a PVC tube extending from with the box to above the water
surface) to allow sampling of water within the box while located in the stream bed.

At each station, the exact locations for Whitlock-Vibert box deployment were selected to
approximate a natural redd that fish could use for spawning. Typically, such areas had gravel or
cobble substrates and were outside locations with high stream velocity. Sites were prepared by
raking out a depression in the streambed at the selected deployment location. In some cases,
structures such as boulders, rocks, or logs were placed upstream to create a breakwater area for
placement that ensured flow velocities were not excessive. Each box was placed into a steel
cage filled with coarse gravel for burial in the stream bed. The steel cages containing the boxes
were placed in the streambed depression, oriented parallel to creek flow, and then covered with
gravel (Figure 4-2).

-------
FINAL

Timing and Duration of Exposure

As discussed in Section 2.4.2, available data indicate that concentrations of LA in OU3 streams
tend to increase during the spring runoff. Therefore, the study was implemented as close as was
feasible to the peak of the spring hydrograph in order to achieve exposures at the high end of the
concentration range. The boxes were left in place until all of the viable eggs had hatched and all
living fry had fully resorbed yolks and had reached the swim-up stage.

Field Observations

Each box in LRC, URC, and NSY was observed twice per week until study termination. During
each examination, the number of dead eggs and alevins was recorded, along with water
temperature and oxygen saturation level. Dead organisms (eggs and alevins) were removed after
each observation and submitted for external examination. Remaining organisms were placed
into clean Whitlock-Vibert boxes and re-buried (in the gravel-filled cages) in the streambed.

Negative Controls

Three groups of eggs were placed in Whitlock-Vibert boxes and were maintained in aquaria in a
temperature-controlled refrigerator in a local laboratory. The temperature was adjusted twice per
week to match the temperature observed in LRC.

As was the case in the field, the Whitlock-Vibert boxes in the negative control group were
observed and changed twice a week, moving organisms from the old box to a new box in the
same manner as for field organisms. At this time, a 70% change in aquarium water was
performed. Oxygen levels were also measured twice per week.

All organisms in these negative control groups were monitored for the same biological endpoints
evaluated in the field (mortality, hatch rate, etc.).

Laboratory Observations

At the end of exposure, the cages were removed from the streambed and transported in site water
to an on-site laboratory where all remaining living alevins were transferred into aquaria. After a
brief acclimation period, the swimming behavior of the alevins was observed for 30 minutes.
Then, the fish were sacrificed and the weight and length of each fish was measured. Each fish
was then placed in preservative for transport to a pathology laboratory for external examination.

-------
FINAL

Exposure Characterization

Exposure of eggs and alevins was characterized by collecting samples of water from inside the
"dummy" Whitlock-Vibert box at each station. To avoid potential bias due to suspended
sediment in the water, samples of water from the boxes were withdrawn and discarded until no
visible sediment was apparent. In addition, samples of water from the overlying stream were
also collected. For the boxes in LRC, water samples were collected twice per week. For the
boxes in the reference locations (URC and NSY), water samples were collected once per week.
All water samples from site and reference locations were analyzed for LA, treating the water
with ozone and ultraviolet light prior to analysis to remove any biological material that might
cause fiber clumping and interfere with the analysis.

Data Evaluation

Hatching Success

Egg hatching success was calculated as:

Neggs = starting number of eggs at the exposure location
Deadeggs = total number of eggs that died before hatching

Missingns = number of missing organisms whose life stage is not specified (that is, the

Hatching success (%) = 100 ¦

^eggs Deadeggs Missi7igns
(Neggs-Missingns)

where:

missing organisms may have been eggs).

Alevin Survival

Alevin survival to the end of the study was calculated as:

Alevin survival (%) = 100 ¦

Alive

Hatched-Missing aievin

where:

Alive = number of alevins alive in the chamber on the last day of the study
Hatched = the number of eggs which are known to have hatched (see above)
Missingaievin = the number of alevins that are missing

28

-------
FINAL

Overall Survival

Overall survival (accounting for the combined mortality in both the egg and alevin life stages)
was calculated as:

Overall survival (%) = 100

N-Missingau

where:

Alive = number of alevins alive in the chamber on the last day of the study
N = the number of eggs at the start of the study

Missingan = the total number of missing organisms (not specified plus alevins)

Results

Detailed results of the 2013 study are presented in Golder (2014b). The main findings are
summarized below.

Exposure Conditions

Figure 4-3 Panel A shows flow data for LRC in 2013. As shown, the spring runoff began in
early April and continued through late May. The eyed eggs were placed into the stream on May
6, approximately at the peak of the runoff.

Figure 4-3 Panel B shows temperatures monitored in LRC and the reference streams during the
eyed egg study. As shown, there is a clear diurnal cycle in water temperature in all streams, with
a slow warming trend as the spring progresses. Temperatures in LRC were very similar at all
stations, and were several degrees warmer than in the reference reaches. Consequently, fish
developed more rapidly in LRC, and exposure in LRC was terminated on May 30 but continued
until June 17 at Noisy Creek and June 19 at URC-2.

Figure 4-4 shows measured LA concentrations in water samples collected from inside the
Whitlock-Vibert boxes. As indicated, there was variability over time (Panel A). On average
across the study duration, exposure levels in LRC ranged from about 40 to 45 MFL, with no
apparent spatial pattern (Panel B). Concentrations at the URC-2 and NSY stations were
consistently much lower (< 0.1 MFL).

Average concentrations of LA (MFL) inside the Whitlock-Vibert boxes tended to be somewhat
higher than in the overlying water:

-------
FINAL

Sampling

Inside

Overlying

Station

Box

Water

LRC-2

LRC-4

LRC-5

Hatching and Survival

Table 4-1 summarizes the hatching and survival data from the 2013 repeat eyed egg study. The
data shown in Table 4-1 were used to calculate hatching and survival statistics as described
above. The results are shown in Figure 4-5.

Data from replicate Whitlock-Vibert boxes at a station were combined, and results between
stations within a category (LRC, Reference, Negative Controls) were compared using a two-
tailed Fisher exact test (Golder 2014b). Although some marginally significant differences were
noted (e.g., hatching rate was lower in LRC-5 than LRC-4 or LRC-2) (Golder 2014b), none of
the differences were statistically significant at the p < 0.05 level, so the data were combined into
three data sets (LRC, Reference, and Negative Controls), and the data were compared using a
one-tailed Fisher Exact Test and by one-tailed t-test, as described previously.

The results are shown in Table 4-2. As indicated, there is some variability in the results between
statistical test methods, but the pattern of results suggests a small but marginally significant
decrease in overall survival in LRC compared to both the Reference Group and the Negative
Control Group. This decrease is due mainly to a marginally significant decrease in hatching
success in organisms exposed in LRC-5 (see Figure 4-5).

As discussed previously (see Section 3.2.4), when an effect is observed in an in situ study, it is
sometimes difficult to identify the causal factor(s), which might include both site-related
contaminants as well as localized variation of environmental stressors or conditions. In this case,
because the average exposure concentrations of LA in water were similar between LRC stations
(see Figure 4-4), the lower hatching success in LRC-5 cannot be attributed to LA exposure.
Furthermore, the decrease in overall survival is relatively small in magnitude (less than 10%),
and effects of this magnitude are unlikely to lead to an ecologically significant decrease in trout
population density (Toll et al. 2013).

Size and Growth

Data on the length and weigh of alevins surviving to the end of the study are shown below and
are plotted graphically in Figure 4-6.

-------
FINAL

Station

Size

Wt(g)

Length (mm)

LRC-2

0.10

23.9

LRC-4

0.11

23.4

LRC-5

0.10

24.4

URC-2

0.11

24.2

NSY

0.11

24.8

NC

0.11

24.7

As shown, values were very similar between stations, although the mean values for LRC are
slightly lower than for reference stations. In some cases the differences are statistically different
(Golder 2014b), but these differences are not considered to be large enough to be of significant
ecological concern and are most likely explained by the differences in water temperatures and
study durations for the LRC and reference stations.

Swimming Behavior

All surviving alevins from each Whitlock-Vibert box were transported to a laboratory where
each fish was placed into an individual 1-gallon aquarium filled with water from the stream of
origin. After 5 minutes of acclimation, swimming behavior was observed for 30 minutes.
Abnormal swimming behaviors included:

•	Erratic swimming (e.g., swimming into walls)

•	Inability to swim in a straight line

•	Floating on side, not moving

•	Loss of equilibrium, difficulty maintaining orientation

•	Other abnormal swimming patterns

Each abnormal behavior was classified as occasional ("O"), frequent ("F"), or continuous ("C")
during the 30 minute period. The data are shown in Table 4-3. Statistical comparisons did not
reveal meaningful differences in the frequency of abnormal behaviors between boxes or stations
within LRC or within the reference reaches (Golder 2014b), so the data were grouped into LRC,
Reference, and Negative Controls. The highest abnormal rate (27%) was observed in the
negative control group, with lower rates in LRC (8%) and Reference (4%). Based on the one-
tailed Fisher Exact Test, the frequency of abnormal swimming in LRC was marginally
significantly higher compared to Reference (p = 0.139), but is not higher than the Negative
Control group. In a number of fish (12 out of 31 total), the cause of the abnormal swimming was
attributed to physical deformities (e.g., body or tail crimps) that prohibited normal swimming
(Golder 2014b).

31

-------
FINAL

External Lesion Frequency

All alevins were examined by a pathologist for the occurrence of external lesions or
abnormalities. A wide variety of lesions were observed, both in fish from LRC stations and from
reference stations and negative controls.

Statistical comparisons performed by Golder (2014b) indicated that there were no statistical
differences between stations within LRC or within reference locations, except for the skin,
caudal fin, yolk sac, and body form, with LRC-5 tending to be different than LRC-2 or LRC-4,
and NSY being different than URC. However, based on a p value of 0.20, it would be expected
that about 20% of the values would be different on a purely random basis, and the observed
frequency (7 out of 44 tests = 16%) is within this range. To further evaluate these differences,
EPA chose to perform a statistical evaluation in which the data were stratified by reach rather
than by station. Statistical comparisons by station are presented in Golder (2014b, Table 3-15
and Appendix C).

Table 4-4 (Panel A) summarizes the data. As shown, 34 of 122 fish (28%) from LRC stations
had one or more lesion, compared to 25 of 132 (19%) for Reference stations and 16 of 67 (24%)
for Negative Controls. Based on the one-tailed Fisher exact test, the difference between LRC
and Reference was marginally significant (p = 0.062), while the difference compared to Negative
Controls was not statistically significant.

Table 4-4 (Panel B) summarizes the data stratified by reach and by lesion type. As shown, the
frequency of lesions was low for most tissues, with abnormalities being noted most often in
yolk sack, caudal fin, or body form. Based on the one-tailed Fisher Exact Test, the difference
between LRC and Reference was statistically significant for lesions of the caudal fin and
marginally significant for lesions of the yolk sack and the skin. Compared to the Negative
Control group, none of the differences were statistically significant.

Nature and Etiology of Lesions

Table 4-5 provides the descriptions of the lesions in yolk sack, tail fin, body form, and skin that
were assigned severity scores by the pathologist. As shown, the nature of the lesions ranged
from minor (e.g., notched tail fin, skin discoloration) to severe (missing tail, severe body
deformity). However, there is no clear pattern of differences in the nature of the abnormalities
observed in fish exposed in LRC compared to fish from the Reference or Negative Control
groups.

-------
FINAL

Most of the minor lesions of fins and skin were judged to be attributable to trauma and/or
conspecific aggression. Abnormal body forms were attributed to genotypic mutations, but the
cause for the mutagenic event could not be determined from a gross pathology perspective. The
proliferative epidermal and gill lesions that have been observed during experimental asbestos
exposure in fish (Belanger et al. 1986a) were not observed in any study fish.

4.2.2 In Situ Juvenile Fish Study

As noted above, effects of exposure of fish to toxicants often depends on life stage, so an in situ
study of exposures of juvenile trout was planned and performed in 2012.

Study Design

The DQOs for the juvenile trout study are presented in Section 5 of the Phase V, Part B
SAP/QAPP (EPA 2012a), and the detailed study protocol is presented in Appendix A.3 of the
SAP/QAPP. The major aspects of the study design are summarized below.

Exposed Organisms

The exposed organisms were juvenile cutthroat trout obtained from the MFWP Murray Springs
Hatchery, near Eureka, Montana. The trout ranged in length from about 7.5 to 12.5 cm (mean =
10.5 cm) at test initiation.

Cages

Juvenile trout were exposed to surface water in floating cages. The cages were wooden boxes
with metal mesh on the bottom and sides, and a solid top that sealed the box. The dimensions of
the cage were roughly 13-inches tall, 10-inches wide, and 12-inches long. Floats were attached
along the top of the sides to keep the box suspended in the water column (see Figure 4-7). There
were 15 fish per cage.

Exposure Stations

Juvenile trout cages were deployed at exposure locations close to the locations used in the eyed
egg study. This included two cages each at LRC-2, LRC-4 and LRC-5, and three cages each at
URC-2 and NSY. Deployment locations were selected to occur in natural pools (some with
modifications by study personnel to decrease flow through the cage if flow velocity was too
high).

-------
FINAL

Field Observations

The cages were checked every day during the study period and cleaned. Cleaning involved
gently removing anything trapped against the outside netting and brushing the mesh sides if
needed using a bristle brush. Daily field activities included measuring stream flow, dissolved
oxygen, and temperature, feeding the juveniles with food provided by MFWP, and recording fish
observations. Any dead fish were noted in the field notes, removed from the cage, and
transported to a processing facility that was established in Libby. Each dead juvenile from the
field was weighed and measured and then preserved for subsequent pathological examination.

Surface water samples were collected twice a week in each LRC location and once weekly for
each reference site location. Water samples were collected from a randomly selected cage at
each location. All water samples from site and reference locations were analyzed for LA by
TEM in basic accordance with ISO 10312 counting and recording rules, treating the water with
ozone and ultraviolet light prior to analysis (per Libby laboratory modification LB-000020) to
remove any biological material that might cause fiber clumping and interfere with the analysis.

Observations of Swimming Behavior

At the end of the field study period, all surviving juvenile fish were transported to the laboratory
in Libby to allow for an observation of swimming behavior. Swimming observations were
conducted by placing the surviving trout into a 20-gallon aquarium filled with water from the
fish's corresponding creek. The fish were allowed to acclimate for a 15-minute period prior to
the start of the swimming observations. Swimming observations were then performed at 2, 10,
20, and 30 minutes from the end of the acclimation period. Swimming behaviors were classified
as follows:

Normal

Abnormal

•	holding on the bottom of the tank with
a vertical orientation

•	holding static or moving very slowly in
the water column

•	swimming very fast around the tank

•	lying on the tank bottom

•	floating on their side (with no movement)

•	having difficulty maintaining
vertical/horizontal orientation

•	other unusual activity

Pathology Laboratory Examination

Following completion of observations, the fish were humanely euthanized and preserved for
subsequent pathological examination. Preserved fish were sent to an off-site pathology

34

-------
FINAL

laboratory for evaluation of the frequency and severity of external abnormalities, focusing on the
skin, mouth, lateral line, and fins.

Results

Results of the in situ juvenile trout study are presented in Golder (2013). The main findings are
summarized below.

Exposure Conditions

Juvenile trout were deployed into the floating cages on May 11, 2012, and were exposed in situ
for 33-34 days.

Figure 4-8 (Panel A) shows the mean temperature measured during the study. As indicated,
temperatures were about 1.5 to 2 °C warmer in the LRC stations than in the reference stations.

Figure 4-8 (Panel B) plots concentration of LA from surface water (sampled from within the
floating cages) at each station as a function of time. As shown, there was substantial between-
day variability, with an apparent trend for decreasing concentrations over time. Panel C shows
the mean concentration of LA at each station. As indicated, average LA concentrations in LRC
ranged from about 10 MFL to 30 MFL, with an apparent tendency to increase in the downstream
direction. LA was occasionally detected in URC-2 (mean = 2.9 MFL) but was only rarely
detected at NSY (mean <0.02 MFL).

Survival

Table 4-6 summarizes the juvenile trout survival data. As shown, no deaths occurred in any of
the LRC stations, while 6 out of 90 juvenile trout died in the reference locations.

Length. Weight and Growth

Figure 4-9 summarizes data on the size (length and weight) and growth of fish surviving to the
end of the study. As indicated, fish exposed in LRC (especially LRC-2 and LRC-4) grew faster
and were larger at study termination than fish in the reference streams. When combined across
stations, both body weight and length were statistically higher (p < 0.01) in fish from LRC
compared to reference (Golder 2013). This difference is attributed to the warmer water
temperature in LRC than in the reference streams.

35

-------
FINAL

Swimming Behavior

Detailed descriptions of the swimming behavior at each station and at each time point of
observation are provided in Golder (2013). The results are summarized below.

Total Fish
Observed

Abnormal Swimming

Station

Number

% Abnormal

At any time

After 30 min.

At any time

After 30 min.

LRC-2

LRC-4

33%

LRC-5

URC-2

NSY

Site

12%

Ref

As shown, observations were collected for 89 fish from LRC and 75 fish from reference
locations. Of the 89 fish alive from the LRC floating cages, 78 (88%) showed consistently
normal behaviors and 11 (12%) showed occasional abnormal behavior. Of the 75 fish alive from
the reference areas, 73 {91%) showed consistently normal behaviors and 2 (3%) showed
occasional abnormal behaviors at one or more times during the observation period. Based on the
Fisher Exact test, the frequency of fish displaying abnormal swimming behaviors at any time
during the observation period is statistically higher in LRC than reference streams (p = 0.02).
However, if the data are grouped by station and analyzed by t-test or Wilcoxon rank Sum test,
the difference is not statistically significant (p = 0.23 or 0.38, respectively). This is because the
difference between LRC and reference was due mainly to abnormal behavior in fish from one
station (LRC-4).

Importantly, the abnormal behaviors were mainly transitory, with all but one having disappeared
by the end of the 30-minute observation period. Based on observations at the 30-minute time
period, differences were not statistically different (p > 0.20) by any test.

These results are somewhat difficult to interpret with confidence because of the dependence of
outcome on the statistical test employed and the apparent transient nature of the presumptive
effect. However, because LA concentrations at LRC-4 were lower than at LRC-5, while
prevalence of abnormal swimming was lower at LRC-5 than LRC-4, these differences cannot be
attributed to LA. In addition, because effects were relatively infrequent (12% vs 3%) and were
nearly entirely transitory in nature, it is considered unlikely that the effects on swimming will
result in an ecologically significant impact on survival in the wild.

-------
FINAL

External Lesions
Frequency and Severity

Juvenile trout from all locations had a spectrum of traumatic and idiopathic gross lesions. Each
fish was assigned a severity score for mouth, lateral line, fins, skin, and gills, using the scoring
system summarized in Table 4-7.

The data are summarized in Table 4-8. As indicated in Panel A, based on a one-tailed Fisher
Exact test, the frequency of lesions was not significantly higher in fish exposed in LRC
compared to the reference streams for lesions of the mouth, gills, lateral line, or pelvic, anal, or
caudal fins, but was significantly higher (p < 0.05) for lesions (notching/fraying) of the dorsal
and pectoral fins.

The mean severity scores for fish with lesions are shown in Panel B. In most cases, the average
severity of the lesions was similar in fish from site and reference streams, and based on a one-
tailed Wilcoxon rank Sum test, none of the differences are statistically significant except for
dorsal fin.

Etiology

The fin lesions were judged by the pathologist to be associated with the confined cage conditions
and/or conspecific aggression. The cause of an increased tendency for aggression in fish in LRC
is not known, but might be related their increased size compared to fish in reference station
cages. However, other factors (e.g., differences in flow rate through the cage) might also be
contributing.

Regardless of the cause(s), the fin lesions are not sufficiently severe to cause a serious
impairment of swimming ability in juvenile fish and hence are unlikely to be of significant
ecological concern.

4.3 Population Studies

As discussed in Section 3.2.4, population studies are one way to determine if an environmental
contaminant appears to be adversely impacting on-site populations of exposed ecological
receptors. The EPA, working in concert with the BTAG, determined that site-specific studies of
fish populations and habitat in OU3 streams compared to reference streams would provide a
valuable line of evidence to evaluate risks to fish in OU3. Consequently, fish population studies
were performed in two consecutive years, as described below. The basic requirements of the

-------
FINAL

site-specific fish population studies were specified in the Phase II Part C SAP for the OU3 RI
(EPA 2008c). Key elements of these studies are summarized below.

4.3.1 Demographic Studies

Detailed information on the fish community studies is provided in Parametrix (2009d, 2010).
Key finding are summarized below.

Study Dates and Locations

Surveys of fish density and diversity were performed in October of 2008 and September 2009 at
the following reaches (see Figure 2-9):

• TP-TOE-2

• LRC-1

• LRC-2

• LRC-3

• LRC-5

• URC-1A

• URC-2

• BTT-R1

• NSY-R1

Capture Methods

Fish were collected using electroshocking equipment. Multiple passes of electroshocking were
performed at each sampling location. In 2009, minnow traps were also used in addition to the
electroshocking passes in an effort to increase the effectiveness of capturing smaller fish.

Length, weight, and species type were recorded for each fish collected. Table 4-9 summarizes
the number of fish captured during these sampling efforts.

Of potential significance is the observation that fish < 65 mm in length were not detected in
lower Rainy Creek stations (LRC-1 to LRC-5) during either of these studies. Because young-of-
the-year fish usually fall into this size category, this observation suggests that young-of-the-year
are not present, which in turn implies the population in this reach is not reproducing. However,
lower Rainy Creek is isolated from upward migration of fish from the Kootenai River by a
hanging culvert and is usually (except in times of high water overflow) isolated from downward
migration of fish from Upper Rainy Creek by the tailings impoundment (Parametrix 2010).
Consequently, it is most likely that the population in Lower Rainy Creek is largely self-
sustaining and that young-of-the-year are present. EPA and the BTAG discussed several
alternative hypotheses that might explain the apparent absence of small fish, and decided the

-------
FINAL

most likely explanation was that, because the water in lower Rainy Creek is several degrees
warmer than in reference creeks, fish in lower Rainy Creek grow faster than in reference
locations and exceed the 65 mm length criterion by the time of year the sampling occurred
(September, October). This hypothesis is supported by the finding of numerous trout < 65 mm
in lower Rainy Creek when sampling occurred in August, as well as a clear difference in growth
rates between site and reference streams (see Section 4.4, below). Consequently, no special
importance is attributed to this observation.

Predominant Species

Raw data on the species of trout that could be reliably identified by species are shown in Table 4-
10. As indicated, lower Rainy Creek stations are populated mainly by rainbow trout, with
cutthroat and cutbow trout (a hybrid of rainbow and cutthroat trout) in lower numbers. Cutthroat
trout and cutbow trout tend to be predominant in upper Rainy Creek and Noisy Creek, while
Bobtail Creek is populated mainly by a mixture of brook trout and rainbow trout.

Population Estimates

Fish caught by electroshocking represent only a subset of the total population present in a
sampling reach, even after 2 or 3 passes. For this reason, the total fish population was estimated
using a mathematical model available in an application referred to as "Microfish" (v3.0) using a
maximum likelihood estimate (MLE) method (Van Deventer and Platts 1989). The calculations
were based on all fish captured by electroshocking, but did not include data from the minnow
traps5. This is because minnow traps were not used in both years, and because the openings on
these minnow traps may have been too large (-25 mm in diameter) to effectively capture smaller
fish (Parametrix 2010). These MLE population estimates were used to derive an estimated fish
population density (total fish per acre) for each sampling station by dividing by the area of the
reach evaluated.

Population Attributes

Figure 4-10 provides a graphical summary of the fish density (fish per acre), size (grams) and
biomass (kg/acre), stratified by reach. Although there was variability between years, density
values for LRC stations were consistently lower than for reference stations (Panel A). However,
fish in LRC stations tended to be larger than fish from reference stations (Panel B), so biomass
was only slightly decreased, especially compared to BTT and URC-2 (Panel C).

5 Other methods for estimating fish population density were also evaluated, including the MLE method
with the minnow trap data included (as presented in Parametrix 2010) and the CapPost (vl.0) estimation
method developed by Peterson and Zhu (2004). All methods yielded generally similar results.

-------
FINAL

Data for TP-TOE-2 and LRC-1 to LRC-4 were combined into one group (LRC) and data for
URC-1 A, URC-2, BTT-R1 and NSY-R1 were combined into a second group (Ref). In order to
determine if there was a statistically significant difference, the data sets were compared using a
two-tailed t-test and a two-tailed WRS test. The results are shown below.

Parameter

Mean Value

Statistical Significance

Ref

LRC

t-test

WRS

Density (fish/acre)

3955

654

<0.01

Weight (grams)

6.3

21.2

<0.01

Biomass (kg/acre)

21.7

13.4

0.047

0.034

As indicated, differences are statistically significant (p < 0.05) by both tests for all of the
endpoints. These data support the conclusion that the fish population structure in LRC is
different from that in reference streams, with decreased density, increased size, and decreased
biomass.

4.3.2 Habitat Studies

As noted in Section 3.2.4, one of the potential limitations to a site-specific population study is
that habitat conditions may not be truly equal in the site and reference reaches, and observed
differences in fish density might be related, at least in part, to habitat factors rather than exposure
to LA. Two types of habitat factors are of potential importance:

• Barriers to fish movement

• Habitat quality in the reaches being evaluated

Barriers to Movement

A fish barrier assessment along upper and lower Rainy Creek was conducted in the summer and
fall of 2009 (Parametrix 2010). The barrier assessment consisted of walking the stream to look
for waterfalls, culverts and other structures that may affect fish passage. The most important
determinants of a barrier are the height of the barrier and the depth of the plunge pool. When the
ratio of the two is less than 0.5, it is unlikely that fish can migrate from downstream to upstream
past the barrier, especially when the plunge pool itself is shallow.

As shown in Table 4-11, a total of 17 absolute or potential barriers were identified along LRC.
Of these, five were judged to pose little impediment to fish movement, but the others were
judged to be potentially significant, with the most important being:

-------
FINAL

• A hanging culvert just downstream of LRC-5. This creates an absolute barrier to upward
migration of fish from the Kootenai River.

• A weir at LRC-6. This also likely prevents upward migration from the Kootenai River
because there is no plunge pool at all.

• The dam that forms the tailings impoundment. This 135-foot tall structure represents a
complete barrier to upstream movement, and is also a barrier to downstream movement
except during times of overflow from the impoundment into lower Rainy Creek.

These potential and absolute barriers limit the migration of fish between different reaches of
Rainy Creek, and may be a factor that influences population density within certain reaches.

Habitat Quality

In order to evaluate the potential effect of habitat quality on fish population parameters, EPA
collected data on a number of key habitat variables that are considered to be important
determinants of fish population density (Raleigh et al. 1984). Potential influences of habitat
parameters on fish populations were evaluated based on a comparison of measured habitat
parameters to ranges that are considered to be optimum for sustaining healthy trout populations
(Harig and Fausch 2002, Adams et al. 2008, Hickman and Raleigh 1982, Raleigh 1984, Varley
and Gresswell 1988). Figure 4-11 summarizes the findings. In these figures, the optimum
ranges are shown by solid red and green lines. As indicated, there are several habitat parameters
where conditions in LRC are different from and more frequently outside the optimal range than
for the reference streams. This includes:

• Summer temperatures in LRC are warmer than is optimum for cutthroat trout, are near
the upper end of the range for rainbow trout, and are higher than in reference streams.

• The amount of large woody debris is lower in LRC than is optimal, and is lower than in
reference streams.

• Both the number of pools and the percent of pools in LRC are usually lower than is
optimal, and both tend to be lower than in reference streams.

The statistical correlations between population density and biomass and each of the habitat
metrics are summarized below:

-------
FINAL

Habitat Metric

Correlation Coefficient (R)

Density

Biomass

Max July/August Temp

-0.66

-0.45

Spawning Gravel

0.89

0.60

% Fines

-0.63

-0.45

Area Woody Debris

0.70

0.20

Pools > 30cm

0.40

0.09

% Pools

0.51

0.76

As indicated, fish density shows a moderately strong direct correlation with the availability of
spawning gravel, woody debris, and an inverse correlation with maximum summer temperatures,
while biomass is most strongly correlated with spawning gravel and availability of pools. These
findings suggest that the changes in population structure (both density and biomass) in LRC are
likely largely attributable to differences in habitat variables, especially spawning gravel, woody
debris, water temperature, and pool availability. Potential contributions of LA to the observed
differences in population structure cannot be determined with certainty, however, if present, they
are likely minor relative to the effects of habitat.

4.4 In Situ Lesion Studies

EPA and the BTAG determined that a comparison of the frequency and severity of external and
internal lesions in resident fish captured from OU3 to that for fish from reference streams would
provide an additional useful line of evidence for evaluating risks to fish. The study requirements
were specified in the Phase V Part B SAP/QAPP (EPA 2012a), and the results are presented in
Golder (2014a). The main elements and findings of the study are summarized below.

Study Design

Resident trout were collected by electrofishing at five reaches of lower Rainy Creek (TP-TOE-2,
LRC-1, LRC-2, LRC-3 and LRC-5), one reach on Noisy Creek (NSY-R1), and two reaches in
upper Rainy Creek (URC-1A and URC-2). Minnow traps were also set, but were not effective in
capturing fish and no fish from minnow traps were evaluated. Collection occurred from August
1 to August 6, 2012.

The goal of the study was to collect fish in each of two size (length) classes: < 65 mm, and 65-
100 mm. A total of 10 fish in each size class were sought from both lower Rainy Creek (5
reaches combined) and from NSY and upper Rainy Creek (combined). Lengths of collected fish
were measured in the field to ensure they met the size class requirements. Only cutthroat,
rainbow, and cutbow trout in the intended size classes were kept, and all other fish were released.
Collected fish were kept in cold water from their respective creek in plastic containers until

-------
FINAL

transported to a laboratory in Libby where they were humanely euthanized, weighed, and re-
measured to ensure lengths were accurate. Fish were preserved in 10% neutral buffered formalin
solution and sent for pathological examination.

All fish were examined by a board-certified pathologist for external lesions or abnormalities,
paying particular attention to the gills and lateral line. The pathologist also selected a subset of
the fish for additional histological examination. These fish were sectioned transversely at four
locations to include the head and rostral aspect of the coelom and body, such that the gills,
cranial line, lateral line, fins, and skin could be examined symmetrically for microscopic lesions,
and to evaluate the pathogenesis of any observed macroscopic lesions. Observed external and
histologic abnormalities were scored based on severity and extent as follows:

Severity

Score

Extent

Multiplier

None

Unilateral

Mild

Bilateral

Moderate

Marked

Severe

Statistical Comparisons

Data from all LRC locations were pooled into a combined Site dataset and data from URC and
NSY locations were pooled into a combined Reference dataset for analysis. The frequency of
lesions was compared using a one-tailed Fisher Exact test, while severity scores were compared
using a one-tailed Wilcoxon Rank Sum test.

Results

Number of Fish Submitted

Table 4-12 summarizes the fish that were captured and submitted for examination. As indicated,
there were 10 in each size class (20 total) submitted from LRC (note that all fish were from the
upper reaches (TP-TOE-2, LRC-1 and LRC-2), and none were collected from LRC-3 or LRC-5),
and there were 15 fish < 65 mm and 25 fish 65-100 mm (40 total) submitted from the three
reference locations.

External Lesions

The pathologist performed external examinations of all 60 fish. A summary of the frequency and
severity of the abnormalities observed is presented in Table 4-13, and the findings are discussed
below.

-------
FINAL

Lesion Frequency

Panel A summarizes the frequency of external lesions observed in site and reference fish.
External lesions were most evident as fraying of the fins, particularly the dorsal, pectoral, pelvic,
and tail fins. Statistical evaluation using a one-tailed Fisher Exact test indicated that there were
no external abnormalities that occurred more frequently in Site fish than in fish from the
reference creeks.

Lesion Severity

Panel B summarizes the mean severity scores for site and reference fish. As seen, mean values
were generally the same or higher in fish from reference streams than from site streams, except
for tail fin. However, based on a one-tailed Wilcoxon Rank Sum test, this difference was not
statistically significant.

Nature and Etiology of External Lesions

A detailed description of the nature and likely etiology of the external lesions is provided in
Appendix B of Golder (2014a). Fin lesions were mainly erosions and ulcers of the fin epidermis
which were attributed to a combination of traumas (conspecific or other aggression, collisions
with substrates or rocks, etc.). Skin lesions presented mainly as small flat patches of white
discoloration on the flanks, dorsum and head. In a few fish, the patches or plaques were present
dorsally and ventrally around the lateral line. These white patches were attributed to erosions
and ulcers in the skin likely due to the same factors causing fin erosions and ulcers. Changes due
to tissue processing and formalin fixation may also have contributed to some of the discoloration
noted. Raised plaques that could represent epidermal hyperplasia were not seen in these fish.

Gill lesions were characterized by white discoloration of the filaments in a few fish. The white
coloration in the gills was attributed to the same factors affecting the skin. No lesions were
attributed to LA.

Histological Lesions

After completing the external examinations, the pathologist identified a subset of fish with
certain external abnormalities for further histological examination. This included 5 fish from
each reference stream and 4 fish each from LRC and TP-TOE stations. The fish were selected to
include fish with gill spots and other gill issues (including flaring and reddening) as well as some
white skin discolorations and plaques. However, the histological examination included all
tissues (not just gill and skin).

-------
FINAL

Frequency and Severity

Histological lesion scores were assigned for each tissue based on the severity and extent as
follows:

• Lesion severity (inflammation, hemorrhage, edema, necrosis, etc.)

0 = no lesions, 1= mild, 2 = moderate, 3 = marked, 4 = severe

• Lesion distribution on skin and fins:

1 = dorsal, 1 = lateral, 1= ventral, 1 = operculum

• Lesion distribution on all tissues:

Multiplication factor of 1= unilateral, 2 = bilateral,

The frequency and severity data are summarized in Table 4-14. Because fish were selected to
include certain lesions rather than being a random subset of the whole, the frequency data (Panel
A) have only limited relevance.

Data on histological lesion severity (Panel B) are more meaningful because the data are based
only on the severity scores of observed lesions, not the frequency. As indicated, scores were
generally similar or higher in fish from reference stations than LRC, and statistical comparisons
based on the one-tailed Wilcoxon Rank Sum test indicated that there were no tissues with
statistically significant (p < 0.05) higher severity in site than reference fish, although a
marginally significant (0.05
-------
FINAL

Gills: The principal lesions in gills included atrophy or necrosis of the secondary
lamellae and interstitial lymphocytic inflammation. These effects were judged to be due
to irritation and/or antigenic stimulation, and possibly from post mortem autolysis. Gill
lesions associated with asbestos exposure were not seen in these fish.

Fins: Lesions of the fin were seen in fish from all locations. Lesions were mainly
erosions or ulceration of the skin, similar to that seen in the trunk and head. Other lesions
atrophic changes likely corresponding to the frayed appearance noted in the gross exams.
It was unclear if these atrophic changes were due to external irritation, toxicant exposure,
trauma or stress related damage to the epidermis, or intrinsic factors such as genetics,
nutrition, or metabolic derangements. Fin lesions associated with asbestos exposure have
not been documented.

Oral mucous membranes: Lesions were primarily lymphocytic inflammation in the
submucosa and epithelial layers, and more or less diffuse. The lesions were attributed to
antigenic stimulation, the mouth being one of the first sites of environmental antigen
exposure.

Nasal mucous membranes: Lesions in the nasal mucosa included mild inflammation,
erosions and necrosis. These lesions likely had the same pathogenesis as for skin and
cranial/lateral lines. Nasal mucosal lesions have not been described in fish
experimentally or naturally exposed to asbestos.

Corneas: Lesions were acute erosions or ulcers of the external corneal epithelial layer
and edema in the underlying corneal stroma. These lesions likely had the same
pathogenesis as those for skin, although euthanasia procedures or post mortem abrasions
may also have contributed.

Brain and skeletal muscle lesions: Acute hemorrhage was frequently detected in fish
from all groups, primarily in the facial muscles and in the meninges of the brain.
Hemorrhage was accompanied by acute rhabdomyolysis in the skeletal muscle, and
hydrocephalus in the brain. These lesions were attributed to the manner of capture
(electroshock).

.Skeleton: Some mild curvature of the spine was seen in few fish. A representative fish
examined histologically revealed no abnormalities in histogenesis of bone, bone
symmetry or degeneration of spinal cord, and the curvature was attributed to hyperflexion
associated with tissue fixation.

-------
FINAL

Additional tissues: Several different tissues were examined opportunistically in the
histologic analysis. No lesions were seen in these additional tissues.

Lesion Summary

Gross and histologic lesions were seen in all groups, primarily involving the fins, skin and gills.
Neither the frequency of occurrence nor the severities of external abnormalities were statistically
higher in Site fish compared to reference fish in any case. Histologic lesions were more
extensive in the gills and skin than were apparent from gross (external) examination, suggesting
that gross lesion assessment is not a sensitive means of identifying lesions in these fish. No
primary infectious agents or deposition materials were identified histologically that would
account for the lesions, although the light microscopy techniques used in this study would not
have been able to detect structures lower than 1 [j,m in diameter. No unique lesion morphology
was identified to suggest that asbestos was a contributing factor to lesion development in the
study creeks, and all of the lesions observed are commonly encountered in captive and wild fish
and attributed to a combination of trauma, stress, or suboptimal water quality.

4.5 Weight of Evidence Evaluation for Fish

Four lines of evidence are available to help evaluate the effects of exposure of fish to LA in site
waters, including:

• In situ toxicity studies of eyed eggs and alevins

• In situ toxicity tests of juvenile trout

• Fish population studies

• Resident fish lesion studies

The data and conclusions from these lines of evidence are summarized in Table 4-15.

-------
FINAL

Confidence in this conclusion is medium to high. The chief limitation to the in situ exposure
studies is that there is no control over environmental variables and the findings are limited to the
conditions and concentration values that occurred during the studies (about 40-45 MFL for eyed
eggs and about 10-30 MFL for juvenile trout). Consequently, if substantially higher
concentrations were to occur in other years, the consequences, if any, cannot be predicted. In
general, the chief limitation to fish population surveys is that population parameters and habitat
variable often tend to be variable between years, making it difficult to distinguish between
random and site-related differences. However, in this case, results were relatively consistent
across two years, so confidence in these studies is good.

-------
FINAL

5.0 RISKS TO BENTHIC MACROINVERTEBRATES

5.1 Reported Effects

The toxic effects of asbestos on benthic macroinvertebrates have not been extensively-studied.
Relevant studies that were located are summarized below.

• Stewart and Shurr (1980) exposed larval Artemia salina, a filter-feeding saltwater
crustacean, to suspensions of chrysotile or crocidolite asbestos. The authors reported that
both forms of asbestos caused a decrease (usually about 20%) in larval survival at
concentrations up to 400 mg/L, with no additional increases at higher concentrations. A
suspension of "short chrysotile" was judged to be more potent than "medium" or "long"
chrysotile, although all forms caused the same level of mortality at high concentrations
(400 mg/L or more). Crocidolite was found to be of similar toxicity as chrysotile when
concentrations and fiber length were similar. A concentration of 400 mg/L was estimated
to correspond to concentrations of about 40-200 MFL, depending on fiber length.

• Belanger et al. (1986b, 1986c) investigated the effects of chrysotile exposure on larval,
juvenile, and adult Asiatic clams (Corbicula sp.). Siphoning activity and shell growth of
adult clams and siphoning activity, shell growth, and weight gain of juveniles were
significantly reduced following 30 days of exposure to 0.1 MFL chrysotile. Exposure to
0.001 to 100 MFL caused a significant reduction in release of larva by brooding adults as
well as increased mortality in larva.

No studies were located on the effects of LA on any species of benthic invertebrate.

5.2 Laboratory Toxicity Tests

The EPA, working in concert with the Libby OU3 BTAG, determined that site-specific studies of
the toxicity of LA-contaminated sediment from OU3 would provide one valuable line of
evidence to evaluate risks to benthic macroinvertebrates.

5.2.1 Study Design

The overall study requirements developed by EPA and the BTAG were specified in Section 5 of
the Phase 2 Part C SAP of the RI for OU3 (EPA 2008c). In brief, the SAP specified that static
renewal lifecycle tests be performed for two species of organisms (the amphipod Hyalella azteca
and the midge Chironomus tentans), comparing the effects of exposure to site sediments to
appropriate reference and control sediments. Based on these requirements, the performing
laboratory (Parametrix, Inc.) submitted study protocols that were designed to comply with EPA

-------
FINAL

standard methods (EPA 2000) and the Phase 2 SAP. The protocols were reviewed and approved
by EPA and the BTAG, and the studies were implemented in 2009. Detailed results are
presented in Parametrix (2009b,c), and key features are summarized below.

Treatments

For each species, seven treatments were evaluated:

Category

Location

LA Concentration (%) (a)

Initial

Final

On-site

Carney Pond

Bin C (5%)

Bin C (4%)

Fleetwood Pond

Bin C (1.5%)

Bin C (3%)

Mill Pond

Bin B2 (0.2-1%)

BinBl (< 0.2%)

Tailings Pond

Bin B2 (0.2-1%)

Bin C (1.5%)

Reference

Tepee Pond

Bin A (ND)

Schrieber Lake

Bin A (ND)

Banana lake

Bin A (ND)

Bobtail Pond

Bin A (ND)

(a) As discused in Section 2, sediment is analyzed by PLM and results are
semiquantitative:

Bin A = Non-detect (ND)

Bin B1 = detected at a concentration judged to be less than 0.2%
Bin B2= Detected at a concentration judged to be between 0.2% and 1%
Bin C = 1 % or greater

Data Source: CDM Smith 2013a

Amphib Field Sediment LA_v2.xlsx

LA cone

-------
Table 6-4. Exposure Conditions in Water

Panel A. LA Concentrations Measured in Water

Group

Location

Number of

Concentration (MFL)

samples

Mean

Range

OU3

Carney Pond

7.9

0.03 - 270

Fleetwood Pond

0.09 - 110

Mill Pond

6.7

ND - 52

Tailings Pond

8.7

ND - 53

Reference

Bobtail Pond

ND - ND

Banana Lake

<0.1

ND - 0.09

Teepee Pond

ND - ND

Panel B. Water Temperature

Group

Location

Number of

Temperature (°C)

measurements

Mean

Range

OU3

Carney Pond

16.5

8.6 - 22.1

Fleetwood Pond

18.6

10.6 - 24.3

Mill Pond

15.5

7.8 - 23.9

Tailings Pond

18.0

5.7 - 26.2

Reference

Bobtail Pond

17.6

7.8 - 24.9

Banana Lake

14.4

7.1 - 20.6

Teepee Pond

19.1

8.1 - 25.5

Data Source: Golder 2014c

Amphibian field study data v2.xlsx

Water

-------
Table 6-5. Amphibians Collected During Field Study

Developmental
Stage

Target number
of specimens

OU3

5onds

Reference Ponds

Species

Carney
Pond

Fleetwood
Pond

Mill
Pond

Tailings
Pond

Bobtail
Pond

Banana
Lake

Tepee Pond

Northern Tree

Egg

Frog

Premetamorphs

Prometamorphs

Metamorphs

Columbia

Egg

Spotted Frog

Premetamorphs

Prometamorphs

Metamorphs

Western Toad

Egg

Premetamorphs

Prometamorphs

Metamorphs

Data Source: Golder 2014c

Amphibian field study data v2.xlsx

Organisms collected

-------
Table 6-6.

Metamorphs Sent for Histological Examination

Columbia

Northern

Group

Location

Spotted Frog

Tree Frog

OU3

Carney Pond

Fleetwood Pond

Tailings Pond

OU3 Total

Reference

Bobtail Pond

Banana Lake

Teepee Pond

Reference Total

Grand Total

101

Data source: Golder 2014c

Amphibian field study data v2.xlsx

Sent for Histopath

-------
Table 6-7. List of Tissues Examined Histologically

Skin

Head

Dorsum

Ventrum

Leg

Feet

Adipose

Skeletal Muscle

Bones

Flat

Long

Vertebrae

Digits

Endolymphatic sacs

Ears

Eyes

Nervous system

Brain

Spinal cord

Coelomic cavity

GI Tract

Mouth

Tongue

Esophagus

Stomach

Duodenum

Small intestine

Large intestine

Cloaca

Organs

Pancreas
Liver

Gall bladder

Resp System

Nasal cavity
Larnyx

T rachea/Bronchi

Lungs

Gills

Cardiovascular system

Heart

Large vessels
Small vessels

Renal System

Kidney

Ureter

Bladder

Reproductive Organs

Ovaries/Testes

Endocrine system

Pituitary
Adrenals
Thyroid
Parathyroid

Hematopoetic tissues

Bone marrow

Thymus

Spleen

Data Source: Golder 2014c

Amphibian field study data v2.xlsx

Target Tissues

-------
Table 6-8. Frequency of Histologic Lesions in Field-Collected Metamorphs

Panel A: Columbia Spotted Frog

Tissue

OU3

Reference

FET
(1-T)

Normal

Abnormal

Total

Normal

Abnormal

Total

Dorsum skin

0.162

Ventrum skin

0.679

Skeletal muscle

1.000

Vertebrae

0.679

Brain

0.406

Spinal cord

1.000

Coelomic cavity

0.008

Mouth

1.000

Tongue

0.784

Duodenum

0.777

Small intestine

0.981

Large intestine

1.000

Cloaca

1.000

Pancreas

1.000

Liver

1.000

Gall bladder

0.161

Nasal

0.406

Lungs

1.000

Heart

1.000

large vessels

1.000

Kidney

1.000

Bladder

0.999

Pituitary

0.553

Panel B: Northern Tree Frog

Tissue

OU3

Reference

FET
(1-T)

Normal

Abnormal

Total Organisms

Normal

Abnormal

Total Organisms

Ventrum skin

0.267

Adipose

0.900

Skeletal muscle

0.956

Vertebrae

0.523

Endolymphatic

0.523

Brain

0.523

Coelomic cavity

0.881

Tongue

1.000

Large Intestine

1.000

Liver

0.468

Lungs

0.816

Heart

0.964

Large vessels

0.523

Kidney

1.000

Bladder

1.000

Thyroid

0.389

OU3 significantly greater than Reference (p < 0.05)
OU3 marginally greater than Reference (0.05 < p < 0.20)

Data Source: Golder 2014c

Appendix l_RSv3.xlsx

Lesion frequency

-------
Table 6-9. Severity of Histologic Lesions in Field-Collected Metamorphs

Panel A: Columbia Spotted Frog

Tissue

OU3

Reference

WRSf

Value

Abnormal

Sum of Scores

Mean Score

Abnormal

Sum of Scores

Mean Score

2-tail

1-tail

Dorsum skin

1.50

Ventram skin

2.00

1.000

0.500

Skeletal muscle

3.38

Vertebrae

1.00

1.67

0.182

0.909

Brain

1.00

—

Spinal cord

—

3.00

—

Coelomic cavity

2.53

1.62

0.005

0.003

Mouth

—

4.00

—

T ongue

2.00

3.50

0.134

0.933

Duodenum

4.00

1.00

0.157

0.079

Small intestine

3.00

2.50

0.434

0.217

Large intestine

3.00

136

4.25

0.151

0.925

Cloaca

—

2.00

—

Pancreas

—

2.40

—

Liver

155

4.08

261

4.35

0.074

0.963

Gall bladder

2.00

—

Nasal

4.00

—

Lungs

2.00

3.28

0.022

0.989

Heart

—

3.92

—

large vessels

—

2.00

—

Kidney

190

4.87

274

4.57

0.282

0.141

Bladder

2.50

2.60

1.000

0.500

Pituitary

1.00

—

Panel B: Northern Tree Frog

Tissue

OU3

Reference

WRSf

Value

Abnormal

Sum of Scores

Mean Score

Abnormal

Sum of Scores

Mean Score

2-tail

1-tail

Ven tram skin

2.00

—

Adipose

2.00

1.000

0.500

Skeletal muscle

3.00

1.00

0.083

0.042

Vertebrae

2.00

—

Endolymphatic

2.00

—

Brain

2.00

Coelomic cavity

1.83

1.25

0.106

0.053

T ongue

—

3.50

—

Large Intestine

—

2.50

—

Liver

2.86

2.45

0.511

0.256

Lungs

2.75

3.00

1.000

0.500

Heart

1.50

3.00

0.105

0.948

Large vessels

2.00

—

Kidney

3.74

108

5.14

0.022

0.989

Bladder

—

3.00

—

Thyroid

1.00

OU3 significantly greater than Reference (p S 0.05)
OU3 marginally greater than Reference (0.05
-------
Table 6-10 Weight of Evidence Summary for Amphibians

Study Type

Exposure
Pathways

Endpoint

Was a Difference Observed?

Is the Difference Attributable to LA ?

Is the Difference Judged to be
Ecologically Significant ?

Conclusion

Confidence and Limitations

Survival

No. Survival was higher for organisms
exposed to OU3 sediment that for
organisms exposed to an off-site reference
sediment

Site-specific
sediment toxicity
test using
developing
tadpoles of the
southern leopard
frog

Direct contact
with sediment
and overlying
water; ingestion
of sediment and
detritus

Growth

No. Organisms exposed to OU3 sediment
were larger than organisms exposed to
either control or reference sediment

Frog larvae exposed to OU3
sediment are not impacted by
LA

Medium-High. The sediments tested were
selected to be at the high end of the range
observed on-site. Most sediments from
LRC have lower concentrations, so risk of
effect would be even lower.

Development

Yes. About half of all organisms exposed
to OU3 sediment did not complete full
metamorphosis by study termination

Unknown. Study design was intended to evaluate
potential effects of maximally exposed organisms and
does not allow assessment of dose-responsiveness.

Unlikely. Development was nearly
complete, with most lagging organisms
having reached Gosner stages 43-45.

Size and
weight

No. There is no consistent pattern of
decreases in either size or weight for
organisms collected from OU3 compared
to organisms from reference locations.

Site-specific
survey of lesion
frequency in
native species
(northern tree
frog, Columbia
spotted frog)

All pathways,
including direct
contact with
sediment.,

External
lesions

No. No lesions were observed in
organisms captured in OU3.

Native amphibian species
captured in OU3 do not have

High. Results are based on two species
(tree frog, spotted frog), although

surface water,
ingestion of
sediment and
detritus

Histological

lesion

prevalence

No. Histological lesions were not more
frequent in amphibians from OU3 than
expected based on organisms from
Reference areas

No. Nearly all of the tissue lesions observed in
organisms from both OU3 and Reference areas were

lesions attributable to LA in
water or sediment.

insufficient numbers of toads were
captured to allow evaluation.

Histological
lesion severity

No. There is no apparent tendency for
tissue lesions to be more severe in OU3
that in Reference areas

inflammatoiy in nature and were attributed to
parasitism

Amphibian WOE v3.xlsx

-------
Table 7-1 Small Mammal Species Captured

Western

Deer

jumping

Yellow-pine

Bushy tailed

Location

Trap Line

Mouse

mouse

chipmunk

woodrat

Reference

Total

OU3

Total

Data Source: Golder2010

Small mammal results v2.xlsx

Species captured

-------
Table 7-2. Size Data for Deer Mice

Trap

Body Weight (g)

Lengt

l (cm)

Location

Line

Females

Males

Females

Males

Reference

15.7

16.2

16.4

16.5

—

16.5

—

14.9

15.4

16.6

16.5

13.3

14.3

16.1

16.5

Mean

15.1

16.4

16.5

Stdev

1.4

0.7

0.2

0.1

OU3

15.9

16.2

16.4

16.2

15.0

12.2

15.6

14.7

13.5

17.6

14.8

15.9

12.8

15.6

14.5

16.2

—

17.5

—

17.5

12.6

16.3

15.6

16.4

Mean

14.0

15.9

15.4

16.2

Stdev

1.4

2.0

0.7

0.9

Stat. Signif. (t-test)

0.265

0.429

0.042

0.430

Statistically sifgnificant (p ^ 0.05)

Data Source: Golder2010

Small mammal results v2.xlsx

Body weight

-------
Table 7-3. Gender Distribution of Mice

Trap

Number

Percent

Location

Line

Females

Males

Total

Female

Male

Reference

Total

65%

35%

OU3

Total

45%

55%

Stat. Signif. (2-tail FET)

0.103

Marginally statistically significant (0.05 < p < 0.20)

Data Source: Golder 2010

Small mammal results v2.xlsx

Gender

-------
Table 7-4. Estimated Age of Mice

Trap

Estimated Age (days)

Location

Line

Females

Males

Reference

180

218

161

—

155

316

139

113

Mean

159

216

Stdev

102

OU3

165

137

214

105

136

142

163

—

226

105

186

Mean

144

159

Stdev

Stat. Signif. (t-test)

0.560

0.438

Data Source: Golder 2010

Small mammal results v2.xlsx

Age

-------
Table 7-5. Small Mammal Lesion Frequency and Severity

Frequency

Severity (a)

System

Tissue

Reference

Site

FETp

Reference

Site

WRSp

Upper airway

Larynx

15/33

45%

24/38

63%

0.104

1.33

1.75

0.021

Trachea

26/34

76%

28/38

74%

0.706

1.96

1.89

0.563

Left Mainstem Bronchus

21/32

66%

28/34

82%

0.102

1.57

1.68

0.301

Right Mainstem Bronchus

20/29

69%

22/33

67%

0.678

1.70

1.95

0.167

Lung

Left Cranial Lung

24/33

73%

30/37

81%

0.292

2.96

2.63

0.837

Left Middle Lung

23/33

70%

27/37

73%

0.484

3.17

2.93

0.784

Left Caudal Lung

29/33

88%

32/37

86%

0.700

3.03

2.88

0.705

Right Cranial Lung

29/34

85%

33/38

87%

0.558

3.07

3.24

0.330

Right Middle Lung

26/34

76%

32/38

84%

0.298

2.92

3.16

0.334

Right Caudal Lung

33/34

97%

35/38

92%

0.928

4.00

4.57

0.240

Post Caval Lung

29/33

88%

31/37

84%

0.796

4.03

4.39

0.302

Upper GI

Esophagus

2/34

3/38

0.553

1.50

1.33

0.500

Cardiac Stomach

8/34

24%

3/38

0.986

1.63

2.67

0.075

Fundus

1/34

2/38

0.542

1.00

1.50

0.500

Pylorus

5/34

15%

4/37

11%

0.802

1.00

0.500

Lower GI

Duodenum

27/34

79%

34/38

89%

0.196

1.07

1.00

0.940

Jejunum

28/34

82%

35/38

92%

0.186

1.25

1.23

0.574

Ileum

32/34

94%

35/38

92%

0.785

1.22

1.14

0.787

Cecum

25/34

74%

30/38

79%

0.396

1.24

1.10

0.912

Colon

19/34

56%

19/38

50%

0.769

1.32

1.11

0.935

Rectum

2/34

2/38

0.734

2.00

1.50

0.500

Anus

2/26

1/28

0.895

1.00

0.500

Other tissues

Adrenal

6/34

18%

5/38

13%

0.804

2.33

2.40

0.500

Thryoid

1/32

2/36

0.545

2.00

2.50

0.500

(a) Mean severity score for animals with lesions

Site statistically higher than Reference (p ^ 0.05)

Site marginally higher than Reference (0.05 < p ^ 0.20)

Data Source: Golder 2010

Small mammal lesion data v2.xlsm

Table

-------
Table 7-6 Weight of Evidence Summary for Mammals

Study Type

Exposure
Pathways

Endpoint

Was a Difference Observed?

Is the Difference
Attributed to LA ?

Is the Difference Judged to be
Ecologically Significant?

Conclusion

Confidence and Limitations

Site-specific survey of
lesion frequency in
mice

All pathways,
including inhalation
exposure while
foraging, ingestion of
LA from food or soil,
and direct contact.

External lesions

No. No deformities or other gross
abnormalities were observed in any of the
animals, and all animals appeared to be in
good health.

Small mammals residing in
the forest area of OU3 are
not impacted by exposure to
LA.

High. However,
extrapolation of this
conclusion to other
mammals is limited by
several uncertainties
including a) differences in
lifespan, b) differences in
area usage, and c)
differences between the
forest and the mine area.

Histological

lesion

prevalence

Yes. The frequency of lesions was
marginally significantly higher in animals
from the site than from the reference area for
larynx, left mainstem bronchus, duodenum,
and jejunum.

No. None of the lesions were
judged to be consistent with
asbestos exposure, but rather
were attributed to parasitism or
infectious disease.

No. None of the lesions would be
expected to affect survival or
reproduction.

Histological
lesion severity

Yes. The median severity of lesions was
significantly higher (p<0.05) for larynx, and
marginally significantly higher for right
mainstem bronchus and cardiac stomach

table 7-6.xlsx

-------
FINAL

ATTACHMENT A
WILDLIFE SPECIES THAT MAY OCCUR IN OU3

-------
Attachment A-l. Amphibian Species Occuring within the Libby OU3 Site
Page 1 of 32

Habitat Group

Observation in Lincoln,
Co., Montana

Group

Common Name
(Genus/species)

Foraging

Nesting

General Habitat Description

Feeding
Guild

Food

Migration/
Hibernation

Longevity

Size

Global
Rank

State
Rank

Oldest

Most
Recent

Number

Chorus Frogs
(Hylidae)

Pacific Treefrog

(.Pseudacris regilla)

Aquatic

Regularly found in the water only
during the breeding period in spring. In
western Montana they breed in
temporary ponds in lower elevation
forests and intermountain valleys
shortly after snowmelt. Eggs hatch in 2
to 3 weeks and tadpoles take 8 to 10

NA/NP

1946

2006

101

Family Woodland

Salamanders

(Plethodontidae)

Coeur d'Alene
Salamander (Plethodon
idahoensis)

Aquatic

Springs and seeps, waterfall spray
zones, and stream edges. More
specifically, primary habitats are
seepages and streamside talus; they also
inhabit talus far from free water (deep
talus mixed with moist soil on well-
shaded north-facing slopes). In wet w

Invertivore

When above ground, Coeur d'Alene salamanders
feed primarily on insects (11 orders documented)
and other invertebrates, including millipeds,
mites, spiders, harvestmen, snails, and segmented
worms. They appear to be opportunistic feeders
and generally rest

1962

2006

102

Tailed Frogs
(Ascaphidae)

Rocky Mountain Tailed
Frog (Ascaphus
montanus)

Aquatic

Small, swift, coldmountain streams.
Eggs are laid during late summer and
take approximately 4 weeks to hatch.
Tadpoles take 1 - 4 years to
metamorphose, depending on water
temperature. Sexual maturity in
Montana is attained at 6 or 7 years of
ase (the la

Insectivore

Larva feed almost exclusively on diatoms, though
also pollen opportunistic; forage at night. Adults
in forest near streams. Prey on invertebrates,
mainly terres. but also aquatic forms

NA/NP

1949

2006

TVue Frogs
(Ranidae)

Columbia Spotted Frog

(Rana luteiventris)

Aquatic

Spotted frogs are regularly found at
water's edge in or near forest openings.
Wetlands at or near treeline are also
used, but populations are uncommon in
large, open intermountain valleys.
Breeding takes place in lakes, ponds
(temporary and permanent), sp

Larvae: veg (Callitriche/Spirogyra) in
Yellowstone. Adults: mainly ground insects in W
MT: coleoptera 35%, hymenoptera 22%, arachnid
15%; others < 10%

1922

2007

309

TVue Salamanders
Plethodontidae)

Long-toed Salamander

(Ambystoma
macrodactylum)

Aquatic

Variety of habitats from sagebrush to
alpine. They typically breed in ponds or
lakes, usually those without fish
present.

Insectivore

Larv: ostracods/cyclops; also red water mites,
insect egg masses, algae. Adult: terres. arthropods
(mostly formicid coleop, diptera) 74%; aq. insect
larv. (mostly tri- chop) 37%

1962

2007

246

TVue Frogs
(Ranidae)

Northern Leopard Frog

(Rana pipiens)

Aquatic

Low elevation and valley bottom ponds,
spillway ponds, beaver ponds, stock
reservoirs, lakes, creeks, pools in
intermittent streams, warm water
springs, potholes, and marshes. There
is no evidence that this species in
Montana has occupied high elevation
wetlands, in contrast to Wyoming and
Colorado

Invertivore

Metamorphosed frogs eat various small
invertebrates, including various insects, spiders,
leeches, and snails obtained along the water's edge
or in nearby meadows or fields. They rarely eat
small vertebrates. Larvae eat algae, plant tissue,
organic debris, andprobably some small
invertebrates. In Montana, adults have been
documented feeding on 10 orders of insects,
spiders, mites, harvestmen, centipedes,
millipedes, snails, and newly metamorphosed
boreal toads

S1S3

1922

2006

TVue Toads
(Bufonidae)

Western Toad (Bufo
boreas)

Aquatic

Habitats used by boreal toads in
Montana are similar to those reported
for other regions, and include low
elevation beaver ponds, reservoirs,
streams, marshes, lake shores, potholes,
wet meadows, and marshes, to high
elevation ponds, fens

Insectivore

Five insect orders; spiders, daddy longlegs, and
millipeds

NA/NP

1949

2006

126

Montana Species Ranking Codes: Montana employs a standardized ranking system to denote global (G - range-wide) and state status (S) (NatureServe 2003). Species are assigned numeric ranks ranging from 1 (critically imperiled) to 5 (demonstrably secure), reflecting the relative
degree to which they are "at-risk". Rank definitions are given below. A number of factors are considered in assigning ranks - the number, size and distribution of known "occurrences" or populations, population trends (if known), habitat sensitivity, and threat.

G1 SI

At high risk because of extremely limited and potentially declining numbers, extent and/or habitat, making it highly vulnerable to global extinction or extirpation in the state.

G2 S2

At risk because of very limited and potentially declining numbers, extent and/ or habitat, making it vulnerable to global extinction or extirpation in the state.

G3 S3

Potentially at risk because of limited and potentially declining numbers, extent and/or habitat, even though it may be abundant in some areas.

G4 S4

Uncommon but not rare (although it may be rare in parts of its range), and usually widespread. Apparently not vulnerable in most of its range, but possibly cause for long-term concern.
G5 S5

Common, widespread, and abundant (although it may be rare in parts of its range). Not vulnerable in most of its range.

-------
Attachment A-2. Bird Species Occuring within the Libby OU3 Site
Page 2 of 32

Habitat Group

Observations in
Lincoln, Co.,

Common Name
(Genus/ species)

Foraging

Nesting

General Habitat Description

Feeding
Guild

Food

Migration

Longevity

Size

Home Range

Global
Rank

State Rank

Oldest

Most
Recent

Number

American Bittern
(Botaurus lentiginosus)

Riparian

Freshwater wetlands with tall, emergent vegetation. Sparsely vegetated
wetlands occasionally, tidal marshes rarely.

Aquatic
Invertivore

Mainly insects, amphibians, crayfish and small fish
and mammals.

Migratory

706 g

S4B

1991

2006

American Coot (Fulica
americana)

Riparian

Marshy borders of ponds

Herbivore

Grains, grasses, and agricultural crops on land;
however, it generally forages in or under water,
where it is almost exclusively an herbivore

Migratory

724 g

S5B

1991

2006

American Crow (Corvus
brachyrhynchos)

Scavenger

One of the most widespread of North American birds. Found in a wide variet
of habitats, particularly in open landscapes, with scattered trees and small
woodlots. Uses both natural habitats and those created by humans (logged,
areas, agricultural fields, cities, and villages). Generally avoids large areas ol
forest

Omnivore

Wide variety of invertebrates (terrestrial and
intertidal marine); amphibians; reptiles; small birds
and mammals; birds' eggs, nestlings and fledglings;
grain crops ; seeds and fruits; carrion; and discardec
human food

Migratory

316-575 g

spring-summer home range
averaged 2.6 sqkm

S5B

1992

2006

American Dipper
(Cinclus mexicanus)

Riparian

Prefers fast-moving, clear streams along with waterfalls. Species prefers
sand, pebble, or rocky stream bottoms, which provide sufficient aquatic
invertebrates. Shorelines with large boulders, fallen trees, and rubble provide
good shelter and protection from predators.

Aquatic
Invertivore

aquatic invertebrates, insects, and insect larvae. Occ

Non-Migratory

reported defense of up to 320
meters of stream in breeding
season, and from 46-820
meters in nonbreeding season.
Year-round density was 1.3 to
2.9 birds per kilometer of
stream.

1991

2005

American Goldfinch
(Carduelis tristis)

Arboreal/Shr
ub/Ground

Widely distributed in temperate North America. Common in weedy fields,
river flood plains, early second growth forest, and also cultivated lands,
roadsides, orchards and gardens, in shaded locations under canopy of leaves
or dense cluster of needles.

Grainivore

Feeds on seeds (e.g., birches, alders, conifers,
thistles, goldenrod, etc.); eats some berries and
insects. Small seeds of various trees. Insects only af
encountered.

Migratory

13g

S5B

1991

1998

American Kestrel (Fcdco
sparverius)

Ground

Arboreal/Cl
iffs/Cavity

found in nearly all habitats in Montana. Nests are often located in cavities in
trees, banks, cliffs, and buildings. They also use man-made nest boxes. They
usually hunt in open habitat. Kestrels often perch on overhead wires or posts
while looking for prey, or hover in midair. In Bozeman area, summer birds
are concentrated in the valley, but some birds are found far up mountain
canyons; wintering birds tend to frequent irrigated areas

Carnivore

During the summer, kestrels feed heavily on large
insects such as grasshoppers. Other prey includes
small birds, rodents, and snakes. During winter they
feed primarily on small birds and rodents.

Migratory

160 g

Average territory size was
109.4 ha and 129.6 ha in two
westernU.S. studies (Cade
1982); home range diameter
during the breeding season
ranged from about 0.5 to 2.4
km in different reaioni

S5B

1991

2006

American Redstart
(Setophaga ruticilla)

Arboreal

Shrub

prefers second growth, deciduous woodlands usually near water. Often fount
in shrubby areas, along with alder and willow thickets

Invertivore

mainly of insects. In late summer months, small
berries and fruits. Eats mostly forest tree insects,
also spiders and some fruits and seeds

Migratory

Less than 2 ha

S5B

1991

2005

American Robin
(Turdus migraiorius)

Ground

Arboreal/Sh
rub

iviosi widespread iNoitn Americantnrusn. frequents roresi, woodland, anc
gardens, breeding primarily where lawns and other short-grass areas are
interspersed with shrubs and trees, such as residential areas, towns,

Invertivore

Eats worms, insects, and other invertebrates (mostly
obtained on ground), and small fiuits

Migratory

W g

Territory sizes average 3.65
acres in Douglas fir forests in
western Montana.

S5B

1991

2006

828

American Three-toed
Woodpecker (Picoides
dorsalis)

Arboreal

Dead tree -
Cavity

Nesting habitat includes coniferous forests (with spruce, larch, or fir trees), o
logged areas and swamps. A cavity nest is dug by both sexes and is placed
1.5 to 15 meters (5 to 50 feet) high in a stump or other dead or dying trees,
often near water.

Invertivore

larvae of bark beetles. Also, tree sap and insects.

breeding density hit 13.5 birds
per 100 acres in lodgepole pine
during a pine beetle epidemic,
probably due to the ability of
birds to nest in lodgepole pine.
In Oregon, home ranges for 3
radioed individuals were 751,
351, and 131 acres.

S3S4

1992

2005

American Wigeon
(Anas americana)

Riparian

Breeds near shallow, freshwater wetlands: sloughs, ponds, small lakes,
marshes, and rivers. For nesting prefers areas with upland cover of
brush/grass vegetation in the vicinity of lakes or marshy sloughs.

During winter and migration almost entirely
vegetarian - stems and leafy parts of acquatic plants
leafy parts of upland grasses and leafy parts and
seeds of various agricultural crops. During breeding
season there is a shift toward a greater proportion ol
seeds and fruits and a substgantial shift toward mort
nonplant foods - insects, mollusks and crustaceans.

Migratory

792 g

S5B

1986

2005

-------
Attachment A-2. Bird Species Occuring within the Libby OU3 Site
Page 3 of 32

Habitat Group

Observations in
Lincoln, Co.,

Common Name
(Genus/ species)

Foraging

Nesting

General Habitat Description

Feeding
Guild

Food

Migration

Longevity

Size

Home Range

Global
Rank

State Rank

Oldest

Most
Recent

Number

Bald Eagle (.Haliaeetus
leucocephalus)

Riparian

Arboreal

Riparian and lacustrine habitats (forested areas along rivers and lakes),
especially during the breeding season. Important year-round habitat includes
wetlands, major water bodies, spring spawning streams, ungulate winter
ranges and open water areas. Nesting sites are generally located within large
forested areas near large lakes and rivers where nests are usually built in the
tallest, oldest, large diameter trees. Nesting site selection is dependent upon
maximum local food availability and minimum disturbance from human
activity

Piscivore

The majority of diet is comprised of fish. Important
prey for Bald Eagles are waterfowl, especially in tht
winter, salmonids, suckers, whitefish, carrion and
small mammals and birds

Non Migratory

First breeds
in 5-6 yr

5244 g

Defended territories are 11^5
hectares and average 23 ha and
territory radius around active
nests averaged 0.6 km. Feeding
home ranges 7 square
kilometers breeding home
ranges averaged 21.6 square
kilometers

1983

2005

325

Bank Swallow (Riparia
rip aria)

Riparian

Ground

Breeds primarily in lowland areas along ocean coasts, rivers, streams, lakes,
reservoirs, and wetlands. Nesting colonies also found in artificial sites such a
sand and gravel quarries and road cuts. Most rivers and streams with nesting
habitats are low-gradient, meandering waterways with eroding streamside
banks.

Aquatic
Invertivore

Takes flying or jumping insects almost exclusively
on the wing. Occasionally eats terrestrial and
aquatic insects or larvae. Rare consumption of
vegetable matter appears to be accidental.

Migratory

1 -2 yr

15g

Most foraging flights within
0.8 kilometers of colony

S5B

1993

1999

Bain Swallow (Hirundo
rustica)

Aerial

Buildings

Originally nesting primarily in caves, it has almost completely converted to
breeding under the eves of or inside artificial structures such as buildings anc
bridges. Presently found in various habitats, including agricultural areas,
cities, suburbs, and along highways. Breeding habitat usually contains open
areas (fields and meadows) for foraaina, a nest site that includes a vertical oi

Aerial
Invertivore

Flying insects. Flies over open land and water and
forages on insects; forages nearer to the ground thar
other swallows (usually not greater than 10 meters
and often less than 1 meter above the ground) Feed
opportunistically on a wide variety of flying insects:

Migratory

17-20 g

Usually forages within a few
hundred meters of nest when
breeding.

S5B

1991

2005

Barred Owl (Strix
varia)

Carnivore

Restricted to forested areas, ranging from swamps and riparian areas to
upland regions. Large, unfragmented blocks of forests preferred. Throughout
its range, found in association with mature and old growth forests, typically
of mixed deciduous-coniferous compositio

Carnivore

An opportunistic predator, consuming small mammals and
rabbits, birds up to the size of grouse, amphibians, reptiles, and
invertebrates

Non-Migratory

801 g

Home range usually is less thar
400 ha (but up to 760 ha) over
2-7 months, average 273
hectares

1995

2004

Barrow's Goldeneye
(Bucephala islandica)

Riparian

Chiefly a bird of the western montane region of North America. This species
is generally restricted to areas west of the Continental Divide. Prefers alkalin
to freshwater lakes in parkland areas; to lesser extent, subalpine and alpine
lakes, beaver ponds, and small sloughs. In summer usually found in small,
scattered groups. In winter often seen in large flocks.

Aquatic
Invertivore

Aquatic invertebrates (insects, mollusks,
crustaceans) and fish eggs. Seeds and tubers providt
a small fraction of the diet

Non Migratory

1090 g

S5B

1987

1995

Belted Kingfisher
(Megaceryle alcyon)

Riparian

Riparian -
Burrow

Inhabits streams, rivers, ponds, lakes, and estuaries or calm marine waters in
which prey are clearly visible. Availability of suitable nesting sites - earthen
banks where nesting burrows can be excavated - appears critical for the
distribution and local abundance of this species. Prefers to excavate a nesting
burrow near its fishing territory. Needs clear still waters for fishing.

Piscivore

Chiefly fish. Also mollusks, crustaceans, insects,
amphibians, reptiles, young birds, small mammals,
even berries.

Migratory

148 g

Regularly forages up to 8 km
from the nest

S5B

1991

2006

Black-backed
Woodpecker (Picoides
arc tic us)

Arboreal

Early successional, burned forest of mixed conifer, lodgepole pine, Douglas-
fir, and spruce-fir (Hutto 1995a, 1995b), although they are more numerous in
lower elevation Douglas-fir and pine forest habitats than in higher elevation
subalpine spruce forest habitats

Invertivore

Bulk of the diet is wood-boring beetle larvae
(including Monochamus spp. and Englemann spruct
beetle, Dendroctonus englamanni), but they also
feed on other insects (e.g., weevils, beetles, spiders,
ants). Occasionally they will eat fruits, nuts, sap,
and cambium, obtain food by flaking bark from
trees (usually dead conifers) and logs, sometimes bj
picking gleaning. They feed primarily on logs and
low on large-diameter tree trunks (more than 7.5
centimeter diameter at breast height; but most often
15-25 centimeter dbh)

Non Migratory

72 g

178, 307, and810 acres

1987

2005

Black-billed Magpie
{Pica hudsonia)

Ground

Arboreal

Historically, it frequently followed Native Americans and lived on the refuse
of their hunts. In breeding season will be found in thickets in riparian areas,
often associated with open meadows, grasslands, or sagebrush for foraging.
Less specific in its habitat requirements in nonbreeding season. Frequently
numerous near human habitats such as livestock feedlots, barnyards, landfills
sewage lagoons, and grain elevators. Nests are durable, domed structures of
sticks, with mud cup and anchor. Generally prefers high trees. Have been
know to nest on utility poles.

Omnivore

Ground-dwelling arthropods, seeds, and carrion

Non Migratory

189 g

1993

1998

-------
Attachment A-2. Bird Species Occuring within the Libby OU3 Site
Page 4 of 32

Habitat Group

Observations in
Lincoln, Co.,

Common Name
(Genus/ species)

Foraging

Nesting

General Habitat Description

Feeding
Guild

Food

Migration

Longevity

Size

Home Range

Global
Rank

State Rank

Oldest

Most
Recent

Number

Black-capped Chickadee
(Poecile atricapillus)

Arboreal/
Shrubs

Arboreal -
Cavity

Deciduous and mixed deciduous/coniferous woodland, open woods and park
willow thickets, and cottonwood groves. Also disturbed areas such as old
fields or suburban areas. Generally more common near edges of wooded
areas. Nests in cavities. Natural sites typically in trees, especially dead snags
or rotten branches, sometimes old woodpecker holes or even in bird boxes.

Invertivore

Eats mainly insects and other small invertebrates,
and their eggs and immature stages, and seeds and
fruits; forages mainly on woody twigs, branches,
and stems

Non Migratory

Territory size averaged about 8
9 ha

1992

2006

316

Black-chinned
Hummingbird
(Archilochus alexandri)

Shrub/ Groun
d

Riparian

In the arid western portion of range, nests in environments that often include
cottonwood, sycamore, willow, salt-cedar, sugar-berry, and oak. In most
regions, its preferred habitat is a canyon or flood-plain riparian community.
Nests typically in riparian habitats. Nest is a cup shape, primarily composed
of plant down.

Nectarivore

Main foods taken include nectar from flowers; smal
insects and spiders; sugar water from feeders
provided by humans

Migratory

S4B

1993

2006

Black-headed Grosbeak

(Pheucticus

melanocephalus)

Arboreal

Occupies diverse habitats. Cottonwood/willow groves and other riparian
habitats in desert and dry grassland; openings in mature pine forest; aspen
groves; deciduous growth especially in mountain valleys/canyons; pinyon-
juniper woodlands; oak savanna; gardens; orchards. Relatively tolerant of
human disturbance. Nests widely reported to be so thinly constructed that
eggs can be seen through bottom. Nests are generally well concealed among
foliage of branches.

Omnivore

Insects and spiders; cultivated fruit, wild fruit, weec
seeds, and grains. During breeding season, gleans
insects high in trees and in understory.

Migratory

47 g

S5B

1993

2002

Blue Jay (Cyanocitta
cristaia)

Ground

Arboreal

Primarily inhabits deciduous, coniferous, and mixed forests and woodlands.
Common in towns and residential areas, especially those having large oaks o
other mast-producing trees.

Omnivore

Arthropods, acorns and other nuts, soft fruits, seeds,
small vertebrates.

Migratory

87g

S5N

1988

2002

Blue-winged Teal (4nets
discors)

Riparian

Main habitat consists of shallow ponds with adequate supplies of aquatic
invertibrates. Prefers to nest in grass or herbaceous vegetation and rarely uses
brushy nesting cover.

Omnivore

Diet consists of aquatic inveritbrates, seeds,
vegetative parts of aquatic plants, duckweeds, algae
and occasional grains from agricultural crops.
Animal matter dominates diet of laying females.

Migratory

409 g

S5B

1992

1998

Bohemian Waxwing
(Bombycilla garrulus)

Arboreal

Prefers open coniferous or mixed-coniferous and deciduous forests. Often
found in recently burned areas or near lakes and streams, beaver ponds, and
swamps.

Frugivore,
Invertivore

Sugary fruits and insects. During spring, also tree
sap and budding flowers.

Migratory

56 g

SHB,
S5N

1920

1993

Boreal Chickadee
{Poecile hudsonica)

Arboreal

boreal coniferous and mixed forests, muskeg bogs, in the vicinity of white
cedar and hemlock swamps, birches and streamside willows. The species
nests in natural cavities or abandoned woodpecker holes, or in a cavity dug b
a pair in a rotten tree stub, usually within 1 meter of the ground (but up to 3."
m).

Omnivore

conifer and birch seeds, and the eggs, larval stages,
and adults of insects. It forages mainly on twigs and
branches of trees.

Non-Migratory

10 g

S1S2

1994

2005

Boreal Owl (Aegolius
funereus)

Carnivore

Arboreal

High elevation spruce/fir forest, with lodgepole pine sometimes present.
Mature spruce/fir forests with multilayered canopies and a highly complex
structure, at elevations greater than 1500m with a mosaic of openings or
meadows, roost at sites scattered throughout their home range, rarely in the
same stand on consecutive nights or the same tree more than 2X per year.
Roost alone, usually far from their nest and mate

Carnivore

Predominately small mammals, with a few birds ant
insects

Non-Migratory

167 g

1986

1996

Brewer's Blackbird
(Euphagus
cyanocephalus)

Ground

Open, human-modified habitats such as residential lawns, golf courses,
cemeteries, mowed urban parks and campus areas. Also found in large
clearcut forests and plowed fields

Omnivore

During breeding season, diet consists of insects and
other invertebrates, along with grains and weed
seeds. During migration and winter, diet consists of
primarily vegetarian such as waste grains, weed and
arass seeds.

Migratory

67 g

S5B

1991

2006

Brown Creeper
(Certhia americana)

Arboreal

Late successional stages of coniferous forests and mixed coniferous-
deciduous forest. Especially common in unlogged, old-growth stands. The
consistent factor appears to be the need for large trees and snags (dead trees)
for foraging and nesting microsites. Breeding season is the same as winter,
but possible no vegetable matter is eaten. Nest built in 2 parts, base and nest
cup, behind a piece of peeling bark.

Invertivore

Forages primarily on trunks of live trees. In winter
main foods taken include a variety of insects and
larvae, spiders and their eggs, ants, and
pseudeoscorpions; a small amount of seeds and
other vegetable matter.

Altitudinal

Territories ranged from 2.3 to
6.4 ha

1992

2004

225

Brown-headed Cowbird
(Molothrus ater)

Ground

Brood
parasite

Areas with low or scattered trees among grassland vegetationff'woodland
edges, brushy thickets, prairies, fields, pastures, orchards, or even residential
areas. Species is a brood parasite; nests are chosen by females, but are that ol
another species. Care given to cowbird eggs and young is provided by the ho
and reflects characteristics of that species.

Omnivore

Mainly of anthropods and seeds.

Migratory

adult male is
39-57 g,
female is
smaller

S5B

1992

2006

102

-------
Attachment A-2. Bird Species Occuring within the Libby OU3 Site
Page 5 of 32

Habitat Group

Observations in
Lincoln, Co.,

Common Name
(Genus/ species)

Foraging

Nesting

General Habitat Description

Feeding
Guild

Food

Migration

Longevity

Size

Home Range

Global
Rank

State Rank

Oldest

Most
Recent

Number

Bufflehead (.Bucephala
albeola)

Riparian

Freshwater, permanent ponds with no outlet or only seasonal outflow, and
small lakes. Large lakes are avoided except by molting flocks, habit of nestin
in the holes of the Northern Flicker. Will also nest in boxes.

Aquatic
Invertivore

Main foods taken are aquatic invertebrates (insects,
crustaceans, mollusks). Will take some seeds.

Migratory

473 g

S5B

1995

2006

Bullock's Oriole
{Icterus bullockii)

Arboreal

Prefers open woodland areas, especially riparian (river) woodlands with largt
cottonwoods, sycamores, and willows. During spring and fall migration it is
found in a variety of open woodland and urban parklands and tall shrubland.
Nests are typically pensile, often suspended from a few thin branches.

Invertivore

Mostly insects, especially butterfly and moth larvae
and pupae, grasshoppers and crickets, beetles and
other insects.

Migratory

34 g

Females foraged regularly
more than 200 meters from
nest, and up to 1 kilometer

S5B

1993

2004

California Gull (Larus
californicus)

Riparian

Prefers larger lakes, but also occurs on ponds and rivers, especially in spring
and fall. Nests varied in shape from depressions in the ground to constructec
mounds; they were located 2 to 75 feet apart

Aquatic
Invertivore

Insects, oligochaetes, crustaceans, amphibians and
birds, and plant material believed to be ingested
incidentally to consuming animals

Migratory

609 g

Breeding pairs in MT foraged
an average of 17.4 km
(maximum 61 km) from
colony. At another colony,
maximum foraging distance
was 32 km

S5B

1991

1995

Calliope Hummingbird
(Stellula calliope)

Aerial

Arboreal

Mountains; along meadows, canyons and streams. Open montane forest,
mountain meadows, and willow and alder thickets, gardens; in migration and
winter also in chaparral, lowland brushy areas, deserts. Nests in tree
(frequently conifer) at edge of meadow or in canyon or thicket along stream.
Nests <1-21 m above ground (usually low, with branch or foliage above).
Nectar supply unimportant in location of male's breeding territory In
Bozeman area occurs on thickety hillsides and in forest openings to moderatt
elevations in the mountains.

Aerial
Invertivore

Floral nectar and small insects. Like other
hummingbirds, it forages aerially for small insects.

Migratory

S5B

1991

2004

Canada Goose (Branta
canadensis)

Ground

Riparian

Various habitats near water, from temperate regions to tundra. In migration
and winter, coastal and freshwater marshes, lakes, rivers, fields, etc. Breeds i
open or forested areas near lakes, ponds, large streams, inland and coastal
marshes. The nest is built on the ground or on an elevated place (muskrat
house, abandoned heron's nest, rocky cliffs, etc.). Usually returns to nesting
territory used in previous year.

Herbivore

Grazes on marsh grasses, sprouts of winter wheat
(spring), grain (fall); eats clover, cattails, bulrushes,
algae, pond- weed, and other plants. Feeds in
shallows, marshes, fields. Also eats mollusks and
small crustaceans

Migratory

Begin
breeding at 2
years, most
by age 3
years.

4741 g

S5B

1991

2006

Canvasback (Aythya
valisineria)

Riparian

Breeds in small lakes, deep-water marshes, sheltered bays of large fresh wate
and alkali lakes, permanent and semi permanent ponds, sloughs, potholes, an
shallow river impoundments. In aspen parklands and mixed-grass prairie,
prefers wetlands bordered by dense emergent vegetation. In boreal forest,
utilizes open marshes. Nest is a large bulky structure. May be overtopped by
vegetation and may have one or more well-maintained ramps.

Omnivore

Foods vary depending upon availability. During
winter and migration, mainly plants (winter buds,
rhizomes, and tubers or aquatic plants. When plant
food is limited, may take small clams and snails.

Migratory

1248 g

S5B

Canyon Wren
(C'alherpes mexicanus)

Ground

Ground/ Clil
fs and Rock
Outcops

Limited to cliffs, steep-sided canyons, rocky outcrops, and boulder piles,
usually in arid regions. Inhabits the same territories year-round. Also
sometimes found in towns, around houses and barns, on old stone buildings.
Nests on canyon walls; may also nest around human-built structures.

Invertivore

Uses its long, decurved bill and flattened head to
probe for spiders and insects in rock crevices

Non-Migratory

39 g

1995

Cassin's Finch
(Carpodacus cassinii)

Arboreal

Prefers open coniferous forests of interior western mountains along with
mature forests of lodgepole pine. Nests in conifer, 3-25 m above ground, on
outer end of limb; may sometimes nest in deciduous tree or in shrub. May
return to same nesting area in successive years, though this may be unusual

Herbivore

Consists of mostly vegetable matter, particularly
buds, seeds, berries and other fruits, along with
some insects.

Migratory

Breeds at 1-
2yr

27 g

1990

2004

155

Cassin's Vireo (Vireo
cassinii)

Prefer dry, open forests. Occupies coniferous, mixed-coniferous/deciduous,
and deciduous forests in mountains and foothills.

Omnivore

Diet consists almost exclusively of arthropods,
spring through autumn. Winter diets consists of
fleshy fruits.

S4B

1994

2005

733

Cedar Waxwing
(Bombycilla cedrorum)

Arboreal

Habitats include deciduous, coniferous, and mixed woodlandsff'especially
open forests and riparian areas of deserts and grasslands; farms, orchards,
conifer plantations, and suburban gardens also popular.

Frugivore,
Invertivore

Diet consists of fleshy fruits and insects. Feeds
opportunistically on small fruits, in spring and
summer also various insects. May consume maple
tree sap and flower petals. Apparantly cannot
maintain positive energy balance when feeding
solely on high-sucrose fruits.

Migratory

33 g

S5B

1992

2006

-------
Attachment A-2. Bird Species Occuring within the Libby OU3 Site
Page 6 of 32

Habitat

Group

Observations in
Lincoln, Co.,

Common Name
(Genus/ species)

Foraging

Nesting

General Habitat Description

Feeding
Guild

Food

Migration

Longevity

Size

Home Range

Global
Rank

State Rank

Oldest

Most
Recent

Number

Chestnut-backed
Chickadee (Poecile
rufescens)

Arboreal

humid coastal and interior forests from southeastern Alaska to southern
California. Year-round resident throughout its range. Occurs within the
densest coniferous forests, or along edges, where temperature is even and
there is considerable shade. Nests in tree cavities and readily colonizes
available nest boxes.

Invertivore

Insects and arthropods make up approximately 65%
of the diet. Seeds and plant material make up the
rest. Eats mainly insects gleaned from twigs,
branches, and trunks of trees and shrubs; in the
breeding season, forages often on outer foliage
(needles, leaves, or buds); also eats spiders, some
fruit, conifer seeds

Non-Migratory

10 g

1991

2005

119

Chestnut-sided Warbler
(Dendroica
pensylvanica)

Arboreal

Shrub

Nesting in shrubby habitat close to the ground, sometimes deciduous trees. Ii
new, second-growth thickets of alder and other deciduous bushes growing in
scrubby clearings and brushy areas or along the margins of streams, in
orchards, pasturelands, forest edges, cut-over forests, roadsides, in open
deciduous woodlands and in powerline corridors. Becomes most common in
deciduous second growth or large forest clearings. Avoids deep woods.

Invertivore

Eats primarily the larvae and some adults of
LepidopteraandDiptera, some spiders, and some
seeds and fruit as well. Usually forages alone.
Gleans the undersurfaces of leaves at the low to
medium levels in shrubs and the lower branches of
small trees, but may feed in the upper canopy

Migratory

log

SNA

1972

1993

Chipping Sparrow
(Spizella passerina)

Ground

Arboreal

Prefers open woodlands, the borders of natural forest openings, edges o
rivers and lakes, and brushy, weedy fields. It has apreference for nesting in
open glades of coniferous forests, and for foraging in brushy open areas
making it suited to human-modified habitats. Nests in a wide variety of tree
and shrubs; has a distinct preference for conifers. Nest is a loosely woven
cup.

Herbivore

Feeds primarily on seeds of grasses and various
annual plants, infrequently supplementing this diet
with small fruits. Adds insects and other
invertebrates when breeding. Mainly forages on the
ground, but also in foliage.

Migratory

Territory sizes of 1.1 to 1.8
acres

S5B

1989

2006

969

Cinnamon Teal (4nets
cyanoptera)

Riparian

Prefers wetlands including large marsh systems, natural basins, reservoirs,
sluggish streams, ditches, and stock ponds. Well-developed basins with
emergent vegitation common habitat.

Omnivore

Seeds and aquatic vegetation, aquatic and semi-
terrestrial insects, snails, and zooplariktonFeeds on
aquatic plants in shallow water areas; especially on
rush seeds, pondweed seeds and leaves, and salt
grass seeds. Also eats small amounts of animal fooc
especially insects and mollusks

Migratory

408 g

S5B

1991

1993

Foraging 0.8 to 2.4 km from

Clark's Nutcracker
(Nucifraga eolumbiema)

Arboreal

Found in close association with ponderosa pine, Douglas fir, and white-bark
pine. Usually nests at elevations between 1800 and 2500 m. Nests on outer
end of branch of a conifer, 2-45 m above ground.

Grainivore

Fresh and stored pine seeds. Also eats insects,
acorns, berries, snails, carrion; sometimes eats eggs
and young of small birds.

Non Migratory

141 g

nest, summer home range of
1500 ha (4.4. km in diameter).
Year-round home ranges are
muchlarger: 15,000 ha in
areas of good food

1991

2005

130

Clay-colored Sparrow
{Spizella pallida)

Ground

Prefers open shrubland, thickets along edges of waterways, second-growth
areas, and forest edges and burns

Omnivore

Feeds on a wide variety of seeds; during the summe
eats insects. Forages on or near the ground. When
breeding, feeds in area separate from nesting
territory

Migratory

Nesting territories about 0.1 to
0.5 haand0.04-0.1 ha.

S4B

1995

2004

Cliff Swallow
(Petrochelidon
pyrrhonota)

Aerial

Cliffs/Eaves

Open to semiwooded habitat, cliffs, canyons, farms; near meadows, marshes,
and water. Builds bottle shaped mud nest in colonies on cliffs, eaves of
buildings, under bridges, etc. Prefers sites with overhang.

Aerial
Invertivore

Flying insects at all times of the year. Insects taken
reflect local availability.

Migratory

22 g

Forages usually within 0.5 km
of colony

S5B

1992

2005

Common Goldeneye
(Bucephala clangula)

Riparian

Breeding birds usually are found in forested wetland habitats

Aquatic
Invertivore

During breeding season, primarily insectivorous ant
prefers lakes (often Ashless) with abundant aquatic
invertebrates. Fish, crustaceans, and mollusks
become a more important part of the diet in winter.

Migratory

1000 g

1977

2006

Common Merganser
(Mergus merganser)

Riparian

Occur on large lakes and large rivers. During migration, most birds are on
lakes

Piscivore

Eats primarily small fish, but will also eat insects,
mollusks, crustaceans, worms, frogs, small
mammals, birds, and plants

Migratory

Breeds at
end of 2nd
yr

1709 g

S5B

1977

2000

Common Nighthawk
(Chordeiles minor)

Aerial

Coastal sand dunes and beaches, woodland clearings, prairies and plains, and
flat gravel rooftops of city buildings. During times of migration, habitat
includes farmlands, river valleys, marshes, and coastal dunes.

Invertivore

Diet consists solely of flying insects

Migratory

64 g

S5B

1992

2006

-------
Attachment A-2. Bird Species Occuring within the Libby OU3 Site

Page 7 of 32

Habitat Group

Observations in
Lincoln, Co.,

Common Name
(Genus/ species)

Foraging

Nesting

General Habitat Description

Feeding
Guild

Food

Migration

Longevity

Size

Home Range

Global
Rank

State Rank

Oldest

Most
Recent

Number

Common Raven
(Corvus corax)

Ground

Broad range ofhabitats: boreal, conifer, and deciduous forests; tundra;
prairies and grasslands; isolated settlements, towns, and cities; deserts; sea
coasts and islands; agricultural fields; Arctic ice floes; and the highest
mountains. It is one of the most widespread naturally occurring birds in the
world.

Omnivore

Diverse diet includes arthropods (even scorpions),
amphibians, reptiles, birds (adults, chicks, and
eggs), small mammals, carrion, grains, buds, and
berries.

Non Migratory

689-1,625 g.

Home range size of breeding
birds reported at 0.2-4.4, 6.6,
9.4 and40.5 sqkm.

1991

2006

592

Common Redpoll
('Carduelis jlammea)

Ground/Trees

Arboreal

Open subarctic, largely coniferous forest and scrub, on dry, rocky, or damp
substrates; level or steeply sloped; avoids dense forest; occurs on tundra and
above timberline only where shrubby deciduous and sometimes coniferous
vegetation occurs in hollows and sheltered places. Nests are built on loose
foundation of small twigs laid across adjacent branches out from trunk of
small spruce or in crotch of alder or willow. Forages in trees or on the
around.

Grainivore,
Invertivore

Very small seeds and other plant material
throughout the year. Also arthropods, particularly in
summer when feeding young

Migratory

13g

move up to 20 km while
foraging

S5N

1990

Common Y ellowthroat
(Geothlypis trichas)

Ground

Occupies thick vegetation in wide range ofhabitats from wetlands to prairie
to pine forest. Nests just above ground or over water, in weeds, reeds,
cattails, tules, grass tussocks, brier bushes, and similar situations; often at
base of shrub or sapling, sometimes higher in weeds or shrubs up to about 1
m.

Invertivore

Eats various small invertebrates obtained among
low plants

Migratory

lOg

S5B

1992

2006

Cooper's Hawk
(Accipiter cooperii)

Arboreal

Nest in dense deciduous and coniferous forest cover, often in draws or
riparian areas. They hunt in these areas or in adjacent open country

Carnivore

Small to medium-sized birds comprise most of the
diet of Cooper's hawks, although they also eat small
mammals

Migratory

529 g

3.2 km from nest

S4B

1991

2005

Cordilleran Flycatcher
(Empidonax
occidentalis)

Aerial/
Arboreal

Ground/Arb
oreal

Coolness, shade, and nest sites" are requisites, and this species, from Alberta
to n. Mexico, "invariably associated with water courses, and thus openings, i
the timber. Has been know to nest in rocky outcroppings near water, in
natural nest cavities in five trees (quaking aspen, Douglas fir), tree stumps,
and about mountain cabins.

Invertivore

Feeds almost exclusively on insects caught in the ai
or gleaned from foliage of trees and shrubs.

Migratory

1993

2004

Dark-eyed Junco (Junco
hyemalis)

Ground

Ground-
Cavity

Occurs across the continent from northern Alaska south to northern Mexico.
Conspicuous ground-foraging flocks are often found in suburbs (especially a
feeders), at edges of parks and similar landscaped areas, around farms, and
along rural roadsides and stream edges. Most often in small cavity on slopin
bank or rock face, under protruding rock, among roots (especially on vertical
surface of root ball of large trees topple by wind), and in sloping road cut
(especially if overhung by grass or other vegetation).

Omnivore

Seeds and arthropods; occasionally fruit and waste
grain in agricultural fields. Most food obtained fron
ground and leaf Utter

Migratory

Territory sizes form of 1.7 to
2.6 acres

S5B

1991

2006

1977

Dark-eyed Junco
(Oregon) (Junco
hyemalis oreganus)

Ground

Ground/Roc
k/Cavity

Occurs across the continent from northern Alaska south to northern Mexico.
Conspicuous ground-foraging flocks are often found in suburbs (especially a
feeders), at edges of parks and similar landscaped areas, around farms, and
along rural roadsides and stream edges. Nest site highly variable. Most often
in small cavity on sloping bank or rock face, under protruding rock, among
roots (especially on vertical surface of root ball of large trees topple by wind
and in sloping road cut (especially if overhung by grass or other vegetation).

Omnivore

Seeds and arthropods; occasionally fruit and waste
grain in agricultural fields. Most food obtained fron
ground and leaf Utter

G5T
5

SNR

1994

2000

Downy Woodpecker
(Picoides pubescens)

Arboreal

Open riparian and deciduous woodlands throughout its entire range. Also use
wooden human-made structures in urban areas. Nests mostly in hole dug by
both sexes in dead stub of tree, also in live tree (especially dead part),
fenceposts; 1-15 m above around.

Invertivore,
Frugivore

Insects, including adults, larvae, pupae, and eggs,
obtained from bark of trees; also eats berries and
nuts

Non Migratory

27 g

1991

2004

Dusky Flycatcher
{Empidonax oberholseri)

Aerial

Shrub

Open coniferous forest, mountain chaparral, aspen groves, streamside willow
thickets and brushy open areas. In MT, Nests were in small bush crotches;
the average nest height was 5 fee

Aerial
Invertivore

aerial forager - a sit and wait predator. It eats flying
insects, occasionally pounces on prey on the ground

Migratory

lOg

S5B

1993

2005

316

Dusky Grouse
(Dendragapus obscurus)

Ground

Winter at high elevations in conifer stands. In early spring, they descend to
lower altitudes, where they prefer forest edges and openings

Omnivore

In winter they eat mainly conifer needles. In summe
they eat a mixed diet of insects, green plants and
berries. The young eat mainly insects

Altitudinal

1188 g

Brood movement in summer is
generally less than 0.5 mile

1977

2006

Eared Grebe (Podiceps
nigricollis)

Riparian

Shallow lakes and ponds with vegetation and macro invertebrate communitie
rarely on ponds with fish. They prefer saline habitats at all seasons, allowing
them to escape fish predators and have an abundant of invertebrates.

Aquatic
Invertivore

large variety of aquatic prey, mainly invertebrates,
small crustations, insects, and less often small fish,
mollusks, amphibians.

Migratory

297 g

S5B

1993

1995

-------
Attachment A-2. Bird Species Occuring within the Libby OU3 Site
Page 8 of 32

Habitat Group

Observations in
Lincoln, Co.,

Common Name
(Genus/ species)

Foraging

Nesting

General Habitat Description

Feeding
Guild

Food

Migration

Longevity

Size

Home Range

Global
Rank

State Rank

Oldest

Most
Recent

Number

Eastern Kingbird
(Tyrannus tyrcmnus)

Aerial

Open environments along forest edges and fields. Also orchards and scatterei
shrubs and trees favorable.

Aerial
Invertivore

Eats mainly insects obtained by flycatching from
perch; also eats seeds and small fruits, and may pict
food from ground or water surface

Migratory

40 g

S5B

1991

2006

European Starling
(;Sturnus vulgaris)

Ground

Ground/Arb
oreal

Exotic species. Non-Native. Owing to their close association with man and
behavioral plasticity, starling inhabit a wide variety of areas if a few crucial
needs are met. They forage in open country on short, mown, or grazed fields
abundantly available in urban as well as agricultural areas. These areas also
provide the necessary food resources, nesting cavities, and water. Nests can
be found virtually anywhere a cavity can be found. Preferred sites include
cavity-like openings in buildings, nest-boxes, cavities usurped from
woodpeckers, and natural cavities in trees. Found occasionally without a
cavity in dense vegetation in trees or on the ground.

Omnivore

Extremely diverse diet that varies geographically,
with the age of individuals, and with season.
Generally will eat invertebrates when available,
fruits and berries, grains and certain seeds during
other times of the year. Most foraging time is spent
in open areas with short vegetation.

Non Migratory

85 g

SNA

1991

2006

Evening Grosbeak
(Coccothraustes
vesperiinus)

Arboreal

Common in mixed-conifer and spruce-fir forests, less common in pine-oak,
pinon, Cascadian, ponderosa pine and aspen forests. Less closely tied to
coniferous tree species than other carduelines-also uses deciduous species fo:
nesting and food. Nests primarily in trees but also in shrubs, a spare
structure, shaped like flattened saucer.

Omnivore

Invertebrates, especially spruce budworm and other
larvae; wide variety of small fruits and seeds,
especially maples

Migratory

60 g

1992

2003

154

Flammulated Owl (Otus
flammeolus)

Ground

Arboreal

Associated with mature and old-growth xeric ponderosa pine/Douglas-fir
stands and in landscapes with higher proportions of suitable forest and forest
with low to moderate canopy closure. They are absent from warm and humid
pine forests and mesic ponderosa pine/Douglas-fir. Most often nests in an
abandoned tree cavity made by Pileated Woodpecker, flicker, sapsucker or
other large primary cavity nester, at heights from 1 to 16 meters

Invertivore

Hunt at night and eat nocturnal arthropods. Feeds
on various insects (e.g., moths, beetles,
grasshoppers, crickets, caterpillars;

Migratory

47 g

Territory size about 5.2 sqkm

S3B

1992

2005

Fox Sparrow
(Passerella iliaca)

Ground

Areas of thick cover, usually around forest edges and brushy woodland edges
Also found in grown-up fields, cut-over woodland, and scrubby woods.

Omnivore

Forages on the ground for seeds (e.g., smartweed,
ragweed). Also eats berries (e.g., blueberries,
elderberries) grapes and other fruits. Diet consists
mainly of insects. Other food sources include seeds,
fruit and plant matter.

Migratory

30 g

S5B

1991

2005

192

Gadwall (Anas
strepera)

Riparian

Nest density was highest in saline lowlands, followed by dense nesting cover
panspots, and silty/ shallow clay. Nest success was highest in saline lowland'
then clay, panspots, silty sites and dense cover

Herbivore

Mainly of submerged aquatic vegitation, seeds and
aquatic invertibrates.

Migratory

990 g

S5B

1995

2006

Golden Eagle (Aquila
chrysaetos)

Carnivore

Arboreal/Cl
iffs

Nest on cliffs and in large trees (occasionally on power poles), and hunt over
prairie and open woodlands.

Carnivore

Primarily jackrabbits, ground squirrels, and carrion
(dead animals). They occasionally prey on deer and
antelope (mostly fawns), waterfowl, grouse,
weasels, skunks, and other animals.

Migratory

4,692 g

Territory size in several areas
of the western U.S. averaged
57-142 sqkm

1997

2000

Golden-crowned Kinglet
(Regulus satrapa)

Arboreal

Nests in forests with closed or open canopies, edges of clearings, or near
water

Invertivore

Feeds primarily on insects and their eggs (e.g., bark
beetles, scale insects, aphids). Also drinks tree sap
and eats some fruit and seeds (rare). Young are fed
various insects and other small arthropods and
sometimes small snails

Migratory

Territory size in northern
Minnesota was 2.1-6.2 acres
(mean 4.1 acres)

1991

2005

818

Gray Catbird
(Dumetella carolinensis)

Shrub

Throughout range found in dense shrubs or vine tangles; most abundant in
shrub-sapling-stage successional habitats. Also found in forest edges and
clearings, roadsides, fencerows, abandoned farmland and home sites, pine
plantations, streamsides, and some residential areas. Uncommon in areas
dominated bv conifers.

Omnivore

Main foods taken include insects and small fruits

Migratory

37g

S5B

1994

2005

-------
Attachment A-2. Bird Species Occuring within the Libby OU3 Site
Page 9 of 32

Habitat Group

Observations in
Lincoln, Co.,

Common Name
(Genus/ species)

Foraging

Nesting

General Habitat Description

Feeding
Guild

Food

Migration

Longevity

Size

Home Range

Global
Rank

State Rank

Oldest

Most
Recent

Number

Gray Jay (Perisoreus
canadensis)

Arboreal

A widespread resident of North America's boreal and sub-alpine coniferous
forests. Nests of low to moderate height, often 1 or 2 trees north of north
edge of open bog, road allowance, or other break in the forest.

Omnivore

Arthropods, berries, carrion, nestling birds, fungi.
Copious sticky saliva from enlarged salivary glands
is used to fasten food items in trees, food that is
used extensively by pairs throughout the winter and
even during other times of the year.

Non Migratory

71g

1991

2006

328

Gray Partridge (Perdix
perdix)

Ground

Exotic species. Non-native. Habitat consists of a mixture of cultivated and
noncultivated land; grasslands interspersed with wheat fields, weed patches,
and brushy cover. Optimum conditions are a cool, moderately dry climate an
a mixture of cultivated and noncultivated land. Grain fields and winter wheat
stubble are also used. Field edges provide escape and winter cover

Grainivore

Waste grain is a staple fall and winter food. Weed
seeds and insects are summer food. Feeds primarilj
on seeds of wheat, corn, barley, oats, smartweeds,
lambs's quarters, crabgrass, etc. Also eats leaves of
clover, alfalfa, bluegrass, dandelion, etc. Chicks
feed on insects for first few weeks of fife.

Non Migratory

398 g

In New York, home range size
was 82-672 ha, did not differ
by season

SNA

Great Blue Heron
(Ardea herodias)

Riparian

Nested primarily in cottonwoods in riparian zones, and also in drier,
coniferous sites. Nesting trees are the largest available. Active colonies are
farther from rivers than inactive colonies. The number of nests in the colony
corresponded to the
distance from roads

Piscivore

Feeds mostly in slow moving or calm freshwater.
Eats mostly fish but also amphibians, invertebrates,
reptiles, mammals, and birds.

Migratory

2,576

S3S4

1981

2006

Great Gray Owl (Strix
nebulosa)

Carnivore

Dead Trees

Use lodgepole pine/Douglas-fir in Montana. Habitat is dense coniferous and
hardwood forest, especially pine, spruce, paper birch, poplar, and second-
growth, especially near water. They forage in wet meadows, boreal forests
and spruce-tamarack bogs in the far north, and coniferous forest and
meadows in mountainous areas. Nest in the tops of large broken-off tree
trunks (especially in the south), in old nests of other large birds (e.g., hawk
nest) (especially in the north), or in debris platforms from dwarf mistletoe,
frequently near bogs or clearings.

Carnivore

Small mammals, especially rodents (i.e. voles)
dominate prey over most of the range. Pocket
gophers also dominate the diet of Great Gray Owls
in North America. They usually forage in open area;
where scattered trees or forest margins provide
suitable sites for visual searching.

Migratory

1,298

2000

Great Horned Owl
{Bubo virginianus)

Carnivore

Arboreal/Cl
iffs/Cavity

Occurs from river bottoms to timberline throughout the state.

Nests in stick nests made by other birds, broken-topped snags, hollow trees,

and cliff cavities.

Carnivore

small to medium-sized mammals and birds.

Non Migratory

1,769

Home range size varies
seasonally and geographically.
Breeding territories in
southwest Yukon 230-883 ha,
averaging 483 ha;
nonterritorial floaters averaged
725 ha

1992

2005

Green-winged Teal
(Anas crecca)

Riparian

Highest densities in wooded ponds of deciduous parklands, with additional
breeding in boreal forests, arctic deltas, and mixed prairie regions. Often
inhabits grasslands or sedge meadows with brush thickets or woodlands next
to a marsh or pond. Often inhabits beaver ponds in wooded areas. Ground
nester. Nests typically in sedge meadows, grasslands, brush thickets, or
woods near a pond. Eggs are elliptical to subelliptical.

Omnivore

Broad diet. Seeds of sedges, grasses, and aquatic
vegetation; aquatic insects and larvae, molluscs,
crustaceans

Migratory

364 g

S5B

1986

2005

Hairy Woodpecker
(Picoides villosus)

Arboreal

Arboreal -
Cavity

Primarily a forest bird; widely distributed in regions where mature woodland
prevalent. Also occurs in small woodlots, wooded parks, cemeteries, shaded
residential areas, and other urban areas with mature shade trees, but often
scarce within these habitats. Cavity nester. In western North America, more
often in large dead stubs or in some areas in aspen with fungal decay.

Omnivore

Tree surface and subsurface arthropods and a
diversity of fruits and seeds. Readily comes to
feeders

Migratory

70 g

Territory size 0.6-15 hectares;
varies with habitat quality. In
central Ontario, breeding
territories averaged 2.8
hectares, range 2.4 to 3.2 ha

1991

2005

237

Hammond's Flycatcher
(Empidonax hammondii)

Aerial

Arboreal

Inhabits cool forest and woodland, breeding primarily in dense fir, mature
coniferous or mixed forests to near timberline. nests were saddled on limbs (
mature conifers, 10.5 to 40 feet high.

Aerial
Invertivore

Diet consists of insects. The Hammond's Flycatcher
is primarily an aerial forager, capturing most of its
insect diet on the wing. On occasion it may forage
from leaf surfaces or from the ground

Migratory

lOg

Territory sizes of 1.6 to 3.2
acres in Douglas fir or
lodgepole forests in western
Montana

S4B

1992

2006

355

Harlequin Duck
(Histrionicus
histrionicus)

Riparian

Inhabit fast moving, low gradient, clear mountain streams. Overstory in
Montana does not appear to affect habitat use

Aquatic
Invertivore

95% of the material in droppings in Grand Teton
National Park consisted of Stoneflies, Mayflies, and
Caddisflies

Migratory

687 g

S2B

1972

2005

-------
Attachment A-2. Bird Species Occuring within the Libby OU3 Site
Page 10 of 32

Habitat Group

Observations in
Lincoln, Co.,

Common Name
(Genus/ species)

Foraging

Nesting

General Habitat Description

Feeding
Guild

Food

Migration

Longevity

Size

Home Range

Global
Rank

State Rank

Oldest

Most
Recent

Number

Harris's Sparrow
(Zonotrichia querula)

Ground

Frequents streams, hedgerows, shelterbelts, and brushy ravines dominated by
deciduous trees and shrubs. Feeds primarily on the ground, scratching and
kicking away ground litter with its feet; forages less frequently among
branches of trees. Nests are located on the ground, typically under a shrub
that is on top of, or next to, a hummock. May also be located beneath rock or
turf overhangs. In Northwest Territories, most nests are concealed amid dwai
birch, alder, spruce, and Labrador tea. Nest entrances are often oriented to thi
southeast, opposite the direction of prevailing storms

Omnivore

Diet consists of seeds, fruits, arthropods, and young
conifer needles.

Migratory

longest 1 lyr
8mo

39 g

Territories averaged 2 hectares
but birds foraged up to 500
meters outside territories

SNA

Hermit Thrush
(Catharus guttalus)

Ground

Species prefers interior forest edges such as margins of ponds and edges of
meadows in forested areas.

Omnivore

During breeding diet consists mostly of animal
matter, especially insects and other small
invertebrates. During migration and winter, diet
supplemented by wide variety of fruits. Forages
from around.

Non Migratory

31 g

Territory sizes of 5.1 to 5.6
acres in Douglas fir or
lodgepole pine forests in
western MT

S5B

1991

2005

355

Herring Gull (Larus
argentatus)

Riparian

Mainly islands and areas around water. Sometimes found in rocky or sandy
cliffs; occasionally on rooftops near water.

Scavenger

Diet consists of marine invertebrates, fishes, insects
other seabirds, and adults, eggs, and young of
congeners. Feeds opportunistically mostly on
various animals and garbage. Often a scavenger
around bays and harbors.

Migratory

Adult
plumage in 4
yr

1226 g

SNA

1995

Hoary Redpoll
('Carduelis hornemanni)

Ground

Open forest and scrub, extending farther onto tundra than Common Redpoll,
but still requiring shrub, at least in sheltered hollows; substrate damp or dry.
During migration and in winter, often joins with Common Redpolls. Occurs
open woodland and shrub, along field edges and week patches and in towns
and villages. Nest sites similar to Common Redpoll but may be closer to
water, often over shallow water; in willows, alder, spruce, tamarack, birch.
Where otherwise suitable sites unavailable, nests in cavities in driftwood.

Herbivore

Small seeds of various trees, shrubs, weeds and
grasses, along with other plant parts, supplemented
with invertebrates in summer

Migratory

13g

SNA

Hooded Merganser
(Lophodytes cucullaius)

Riparian

Hooded Mergansers are generally found in river areas bounded by woods ant
supporting good fish populations associated with clear water

Aquatic
Invertivore

Main foods taken are primarily aquatic insects, fish,
and crustaceans (particularly crayfish).

Migratory

First breed
at2yr

680 g

S4B

2006

Horned Grebe
(Podiceps auritus)

Riparian

Breeding Range is on shallow freshwater ponds an marshes with beds of
emergent vegetation, especially sedges, rushes and cattails. In spring and fall
the Horned Grebe is mainly on large sized bodies of water, including rivers
and small lakes. The floating nest is usually concealed in the vegetation.

Aquatic
Invertivore

Aquatic arthropods in the summer, & fish and
crustaceans in winter, especially amphipods,
crayfish, andpolychaetes.

Migratory

453 g

Horned Lark
(Eremophila alpestris)

Ground

Ground -
Cavity

Open, gerally barren country; avoids forests. Prefers bare ground to grasses
taller than a few cm. May nest on marshy soil but generally prefers,
throughout its range, bare ground such as plowed or fall-planted fields. Digs
nest cavity or may use a natural depression. Food obtained from ground.

Grainivore

In winter, mostly seeds. During the breeding season
adults eat mostly seeds but feed insects to their
young. Adults take more insects during the spring
and fall than at other times, perhaps to compensate
for the energetic demands of breeding and molt

Migratory

32 g

Territory size varies with
habitat and population density;
ranges from means of 3.5 ha in
higher latitude heath, 1.6 ha in
the agricultural Midwest to a
range of 0.3-14 ha in Colorado
shorstgrass prairie

House Finch
(Carpodacus mexicanus)

Ground

A common backyard bird throughout most of the contiuguous United States.
In its native west, this species occupies a wide range of open or semi-open
habitats from undisturbed desert to highly urbanized areas. In the east, it is
rarely found far from urban or suburban areas.

Herbivore

In all seasons, 97% of diet is vegetable matter
including buds, seeds, and fruits. Primary weed
seeds eaten include Napa thistle, black mustard,
wild mustard, Amaranth, knotweed and turkey
mullen, plus some 21 additional seed varieties. In
late summer it will eat fruits.

Non Migratory

21 g

1995

1998

House Sparrow (Passer
domesticus)

Ground

Arboreal

Exotic. Non-Native. Breeding habitat is mostly associated with human
modified environments such as farms, and residential and urban areas. Absei
from extensive woodlands, forests, grasslands, and deserts. Nest often in
enclosed spaces. If they nest in trees the nest usually is a globular structure
with a side entrance and may share a wall with a neighboring nest.

Grainivore

Have been known to eat livestock feed. Grains, wee
seeds, relatively few insects. Urban birds eat
commercial birdseed.

Non Migratory

28 g

SNA

1995

2005

-------
Attachment A-2. Bird Species Occuring within the Libby OU3 Site
Page 11 of 32

Habitat Group

Observations in
Lincoln, Co.,

Common Name
(Genus/ species)

Foraging

Nesting

General Habitat Description

Feeding
Guild

Food

Migration

Longevity

Size

Home Range

Global
Rank

State Rank

Oldest

Most
Recent

Number

House Wren
(Troglodytes aedon)

Ground/ Shru
b

Cavity

Affinity for open, shrubby woodlands, mimicked so well by small town and
suburban backyards and city parks; has a preference for human-made "bird
houses". Nests usually in cavities (natural, abandoned woodpecker holes,
bird boxes, and within various human artifacts). Male starts several nests,
female finishes nest.

Invertivore

Feeds primarily on small, terrestrial invertebrates

Migratory

S5B

1992

1998

Killdeer (Charadrius
vociferus)

Ground

Frequents open areas, especially sandbars, mudflats, heavily grazed pastures,
and such human-modified habitats as cultivated fields, athletic fields, airports
golf courses, graveled or broken-asphalt parking lots, and graveled rooftops

Invertivore

Main foods taken include terrestrial invertebrates,
especially earthworms, grasshoppers, beetles, and
snails; infrequently small vertebrates and seeds

Migratory

101 g

S5B

1992

2007

Lark Sparrow
(Chondestes
gra.mma.cus)

Ground

Ground/Arb
oreal/Cavity

Widespread in open habitats such as shrub-steppe, pinion-juniper edges,
grasslands, roadsides, farmlands, and pastures. Nests on bare ground, in
hollow depression, or in shrub or tree up to 2.75 m from ground. May use
unusual nest sites such as a natural cavity of a dead tree. Nest either on the
ground or close to the ground (within 4 meters) in woody vegetation

Omnivore

Categorized as a ground-foraging omnivore during
the breeding season, and a ground-gleaning
granivore during the nonbreeding period. In breedin
season, eats more insects than seeds. During colder
periods, when insects are less readily available,
seeds may be primary diet.

Migratory

29 g

Territories around immediate
nest site (Martin and Parrish
2000), 66-248 sq. m in extent

S5B

2004

Lazuli Bunting
(Passerina amoena)

Ground

Arboreal/Sh
rub

Arid brushy areas in canyons, riparian thickets, chaparral and open
woodland; in migration and winter also in open grassy and weedy areas Nest!
in small trees, shrubs, or vines, 0.3-3 m above around

Omnivore

Feeds on insects (grasshopper, caterpillars, beetles,
ants, etc) and seeds (wild oats, canary grass,
needlearass, etc.).

Migratory

16g

S5B

1991

2006

Least Flycatcher
(Empidonax minimus)

Aerial

Semi-open, second-growth, and mature deciduous and mixed woods;
occasionally conifer groves, burns, swamp and bog edges, orchards, and
shrubby fields. Often found near open spaces such as forest clearings and
edges, water, roads, and cottage clearings. Nest is a neat compact cup,
generally not protected or only partially protected by surrounding vegetation

Aerial
Invertivore

Feeds almost exclusively on insects caught by
hawking from the air or gleaned from foliage of
trees and shrubs. Fruits and seeds taken
occasionally.

Migratory

lOg

S5B

1994

1998

Lesser Scaup (Aythya
qffinis)

Riparian

Ground

In the Bozeman area, habitat is generally restricted to lakes and ponds.
Throughout fall and winter this species forms large flocks on rivers, lakes,
and large wetlands. Pairs and broods typically associated with fresh to
moderately brackish, seasonal and semipermanent wetlands and lakes with
emergent vegetation such as bulrush, cattail and river bulrush . builds nest o
the ground near or over water, as well as in uplands

Aquatic
Invertivore

Mainly aquatic invertebrates such as insects,
crustaceans, andmollusks. Seeds and vegetative
parts of aquatic plants are important in certain areas

Migratory

850 g

S5B

1993

1995

Lewis's Woodpecker
(Melanerpes lewis)

Aerial

Arboreal

Occur in river bottom woods and forest edge habitats. Nest in a natural
cavity, abandoned northern flicker hole, or previously used cavity, 1-52
meters above ground. Sometimes will excavate a new cavity in a soft snag
(standing dead tree), dead branch of a living tree, or rotting utility pole

Aerial
Invertivore

Adult emergent insects (e.g., ants, beetles, flies,
grasshoppers, tent caterpillars, mayflies) in summer
and ripe fruit and nuts in fall and winter. They are
opportunistic and may respond to insect outbreaks
and grasshopper swarms by increasing breeding
densities.

Migratory

116 g

S2B

1991

1995

Lincoln's Sparrow
(Melospiza lincolni i)

Ground

Found mainly in boggy, willow, sedge, and moss-dominated habitats,
particularly where shrub cover is dense. At lower elevations, also prefers
mesic willow shrubs, but can be found in mixed deciduous wood groves sucl
as aspen and cottonwoods. Nests on the ground, most often inside a low
willow shrub or mountain birch that also contains fairly dense sedge cover.

Omnivore

Winter: small seeds, terrestrial invertebrates when
available. Occasionally uses feeders. Breeding
season: mostly arthropods, also small seeds when
available. Forages on ground under grass and brush

Migratory

17g

Breeding territory about 0.4 ha

S5B

1992

1998

Long-eared Owl (4sio
otus)

Carnivore

Arboreal

Most often observed in hedgerows, woody draws, and juniper thickets,
although they do occur within the forest edge. They are predominantly open-
country hunters; however, they are seldom seen because of their nocturnal
habits. Nests in a stick nest built by other raptors, magpies, crows, or ravens

Carnivore

Depends heavily on small rodents.

Migratory

279 g

in Siberia, nesting pairs
remained in an area about 100-
300 meters in diameter

2003

MacGillivray's Warbler
(Oporomis tolmiei)

Riparian-
Ground

Shrub

Commonly found in riparian habitat and clearcuts of northern coniferous
forests along the Rocky Mountains. Forages along streams or in dense seconc
growth. Commonly found in deciduous, shrubby riparian habitats. Usually
nests low, 0.6-1.5 meters above ground, in bushes, saplings, clump of ferns,
etc.

Invertivore

Main food is insects. Feeds on or just above the
ground.

Migratory

lOg

S5B

1991

2005

488

-------
Attachment A-2. Bird Species Occuring within the Libby OU3 Site
Page 12 of 32

Habitat Group

Observations in
Lincoln, Co.,

Common Name
(Genus/ species)

Foraging

Nesting

General Habitat Description

Feeding
Guild

Food

Migration

Longevity

Size

Home Range

Global
Rank

State Rank

Oldest

Most
Recent

Number

Mallard {Anas
platyrhynchos)

Riparian

In North America, the Mallard is the most abundant duck species. Its success
in the wild reflects its adaptability to varied habitats, its hardiness in cold
climates, and tolerance of human activities. Usual nest site is in uplands clost
to water. Nests in wide variety of situations with dense cover, including
grasslands, marshes, bogs, riverine floodplains, dikes, roadside ditches,
pastures, cropland, shrubland, fence lines, rock piles, forests, and fragments
of cover around farmsteads

Omnivore

Very flexible in food choice; diet composition
depends on stage of annual cycle, hydrological
conditions, invertebrate behavior, and crop-
harvesting schedule

Migratory

1,082 g

1977

2006

Marsh Wren
(Cistothorus palustris)

Riparian

Freshwater and brackish marshes in cattails, tule, bulrush, and reeds. Nests ii
marsh vegetation; female finishes one of several nests started by male; male
may continue to build nests even after female begins incubation. Nesting
success may be greatest in marshes with relatively dense vegetation and deep
water

Aquatic
Invertivore

Eats mainly insects and other invertebrates

Migratory

12g

S5B

1991

2006

Merlin (Fcdco
columbarius)

Carnivore

Arboreal

Breeding pairs in eastern Montana usually use sparse conifer stands adjacent
to prairie habitats, but sometimes use shelterbelts and river bottom forests. In
western Montana, they use open stands of conifers and river bottom forests.
Merlins sometimes nest in urban areas

Carnivore

Bulk of diet usually consists of small to medium-
sized birds, often flocking species. Large flying
insects (e.g., dragonflies) may be important for
young learning to hunt. Also eats toads, reptiles, anc
mammals

Migratory

244 g

Mountain Bluebird
(Sialia currucoides)

Ground

Arboreal

Subalpine meadows, grasslands, shrub-steppe, savanna, and pinyon-juniper
woodland; in south usually at elevations above 1500 m. In winter and
migration also inhabits desert, brushy areas and agricultural lands. Nests are
built in natural tree cavities, or abandoned woodpecker holes. May also use
bird box, old swallow's nest, rock crevice, or old mammal burrow.

Invertivore/O
mnivore

Insectivorous. Feeds on beetles, ants, bees, wasps,
caterpillars, grasshoppers, etc. Also consumes some
berries and grapes seasonally. Hovers and drops to
ground while foraging or darts out from a low perch
to catch prey.

Migratory

28 g

S5B

1991

2006

147

Mountain Chickadee
(Poecile gambeli)

Shrub

Ground/Arb
oreal

Year round resident of montane coniferous forests of west North America,
primarily in areas dominated by pine, spruce-fir and pinon juniper. Occurs in
mixed coniferous-deciduous forests. Nests in a natural tree cavity,
woodpecker hole, hole in the ground, or under a rock in a bank. Nest height
usually is low, but may be up to 25 m.

Invertivore

Insects during warm seasons augmented with
spiders. Conifer seeds during cool seasons.

Non Migratory

12 g

Mean territory size 1.5 ha in
Arizona;

1991

2006

875

Mourning Dove
(Zenaida macroura)

Ground

tremendous adaptability. Generally shuns deep woods or extensive forest anc
selects more open woodlands and edges between forest and prairie biomes fo
nesting. Human alteration of original vegetations is generally beneficial for
this species, with creation of opening in extensive forest and plowing of
grasslands for cereal-grain production. Additional habitat created with
planting of trees and shrubs in cities, towns, and suburbs. Nests primarily at
woodland or grassland edge, usually in trees but readily on ground in absenci
of suitable trees or shrubs.

Grainivore

Mostly seeds (99%). Insignificant amounts of
animal matter and green forage may be acquired
incidentally. Principal food items vary by region anc
immediate locale. Feeds almost entirely on ground

Migratory

123 g

Average home range in
Missouri was 3200 ha, but
most activity was within 1.6
kilometers

S5B

1993

2006

Myrtle Warbler
(Dendroica coronata
auduboni)

G5T
5

S5B

1994

2000

Nashville Warbler
(Vermivora ruficapilla)

Ground/Arbo
real

Ground

Forest-bordered bogs, second growth, open deciduous and coniferous
woodland, forest edge and undergrowth, cutover or burned areas; in migratio
and winter in various woodland, scrub, and thicket habitats. Nests on ground
at base ofbush, small tree, sapling, or clump of grass, or in hollow in moss.

Invertivore

Eats insects; forages from ground to treetop, but
mainly low in trees and thickets at edge of forest

Migratory

S5B

1991

2005

Northern Flicker
(Colaptes auraius)

Ground

Arboreal

A common, primarily ground-foraging woodpecker that occurs in most
wooded regions of North America. Prefers forest edge and open woodlands.
Yellow-shafted Flickers reported nesting in most tree species in the wide
range of woodlands it inhabits. Red-shafted Flickers are particularly commoi
in quaking aspen stands and cottonwoods in riparian woodlands and in burne
woodlands. Cavities excavated by flickers are used by many species of
secondary cavity users.

Invertivore

Insects, primarily ants; fruits and seeds, especially
in winter. Feeds on the ground or catches insects in
the air.

Migratory

142 g

1991

2006

572

-------
Attachment A-2. Bird Species Occuring within the Libby OU3 Site
Page 13 of 32

Habitat Group

Observations in
Lincoln, Co.,

Common Name
(Genus/ species)

Foraging

Nesting

General Habitat Description

Feeding
Guild

Food

Migration

Longevity

Size

Home Range

Global
Rank

State Rank

Oldest

Most
Recent

Number

Northern Flicker (Red-
shafted) (Colaptes
auraius cafer)

Ground

Arboreal

Invertivore

Insects, primarily ants; fruits and seeds, especially
in winter. Feeds on the ground or catches insects in
the air.

Migratory

142 g

G5T
5

SNR
B

1994

2000

Northern Goshawk
(Accipiter gentilis)

Goshawks in Montana tend to nest predominately in mature large-tract
conifer forests with a high canopy cover (69%), relatively steep slope (21%)
and little to sparse undergrowth. Nests were constructed an average 10.9
meters above the ground and were usually located near water (232 m) or a
clearing (85 m)

Carnivore

Forage during short flights alternating with brief
prey searches from perches. They also hunt by
flying rapidly along forest edges, across openings,
and through dense vegetation. An opportunistic
hunter, Northern Goshawks prey on a wide variety
of vertebrates and, occasionally, insects. Prey is
taken on the ground, in vegetation, or in the air.

Non Migratory

Breed at 1-2
yr

1137 g

1924

2005

153

Northern Pintail (Anas
acuta)

Riparian

prefer large lakes . Breeders favor shallow wetlands interspersed throughout
prairie grasslands or arctic tundra. An early fall migrant, the species arrives
on wintering areas beginning in August, after wing molt, often forming large
roosting and feeding flocks on open, shallow wetlands and flooded
agricultural fields

Grainivore

Grain (rice, wheat, corn, barley), moist-soil and
aquatic plant seeds, pond weeds, aquatic insects,
crustaceans, and snails

Migratory

1035 g

S5B

1995

2006

Northern Pygmy-owl
(Glaucidium gnoma)

Carnivore

most often seen in mixed fir forests, but can be found form river bottoms to
timberline.

Carnivore

Small birds, mammals, insects, and probably a few
reptiles and amphibians. Small birds may be an
important part of its diet.

Non Migratory

73 g

1994

2005

Northern Rough-winged
Swallow
(Stelgidopte ryx
serripennis )

Aerial

Ground

Long-distance migrant in the U.S. and Canada. Breeding populations from th
lowlands and central interior of Mexico southward are generally sedentary,
though they may make local elevational migrations to coastal areas in winter

Invertivore

Flys through air and catches insects (e.g., flies,
wasps, bees, beetles). Swoops low over open grounc
or water. Occasionally may scavenge on ground.

Migratory

16g

S5B

1991

2006

Northern Saw-whet Owl
(Aegolius acadicus)

Carnivore

Arboreal

Most common in coniferous forests; however, they can be found in deciduou
trees along watercourses. Nests in woodpecker holes and possibly natural
cavities.

Carnivore

Eats mainly small mammals sometimes birds and
insects.

Non
Migratory/Eleva
tional

91 g

1994

2005

Olive-sided Flycatcher
(Contopus cooperi)

Aerial

Ground

Generally breeds in the montane and boreal forests in the mountains o
western North America, highly adapted to the dynamics of a landscape
frequently altered by fire. They are more often associated with post-fire
habitat than any other major habitat type, but may also be found in other
forest openings (clear cuts and other disturbed forested habitat), open forests
with a low percentage of canopy cover, and forest edges near natural
meadows, wetlands, or canyons. Nests are placed most often in conifers
(Harrison 1978, 1979), on horizontal limbs from two to 15 meters from the
around.

Invertivore

hovering or sallying forth, concentrating on prey
available via aerial attack. They generally launch
these aerial attacks from a high, exposed perch atop
a tree or snag. Like others in the flycatching guild,
this bird is a passive searcher, looking for easy to
find prey, but is also an active pursuer, attacking
prey difficult to capture

Migratory

32 g

S3B

1992

2005

332

Orange-crowned Warbler
(Vermivora celata)

Arboreal

Ground

Prefers habitats with shrubs and low vegetation, often in aspen forest or in
riparian or chaparral areas which provide cover for its nest. Nests well
concealed, often on or near ground or in small crevices or depression in
ground/rock, along shady hillside, on slopes or steep banks, sheltered by
overhanging vegetation. Also found in shrubby bushes, ferns, vines. Nest is i
small open cup.

Invertivore

Gleans insects from leaves, blossoms, and the tips o
boughs, but also eats some berries and fruit and is
attracted to suet feeders in the winter.

Migratory

S5B

1992

2004

608

Pileated Woodpecker
(Dryocopus pileaius)

Arboreal

Late success ional stages of coniferous or deciduous forest, but also younger
forests that have scattered, large dead trees. Dead trees provide favored sites
in which to excavate nest cavities. Only large- diameter trees have enough
girth to contain nest.

Invertivore

Diet consists primarily of wood-dwelling ants and
beetles that are extracted from down woody materia
and from standing five and dead trees. Fruit and
mast of wild nuts when available.

Non-Migratory

9 years

308 g

1991

2005

256

Pine Grosbeak
(Pinicola enucleator)

Arboreal

Open coniferous forests of north-western mountain ranges and in coastal and
island rain forests of Alaska and British Columbia. Always most common in
places where forest is open.

Omnivore

During most of the year 99% of diet is vegetable
matter, especially buds, seeds and fruits. Feeds
young a diet of mainly insects and spiders often
mixed with vegetable matter

Migratory

56 g

1988

2004

-------
Attachment A-2. Bird Species Occuring within the Libby OU3 Site
Page 14 of 32

Habitat Group

Observations in
Lincoln, Co.,

Common Name
(Genus/ species)

Foraging

Nesting

General Habitat Description

Feeding
Guild

Food

Migration

Longevity

Size

Home Range

Global
Rank

State Rank

Oldest

Most
Recent

Number

Pine Siskin (Carduelis
pinus)

Ground

Forests and woodlands, parks, gardens and yards in suburban areas; in
migration and winter in a variety of woodland and forest habitats, partly opei
situations with scattered trees, open fields, pastures and savanna.

Herbivore

Forages in trees and on the ground for seeds (e.g., o
alder, birches, pines, maples, thistles) and insects.
Also eats flower buds of elms, drinks nectar from
eucalyptus blossoms and sap from sapsucker's holes

Migratory

15g

1991

2006

1243

Pygmy Nuthatch (Sitta
pygmaea)

Arboreal

long-needled pine forests - principally ponderosa pines. Reaches its highest
densities in mature pine forests little affected by logging, firewood collection
and snag removal. A cavity nester, can excavate own cavity, but will use
woodpecker holes and natural cavities

Invertivoe

Feeds mainly on weevils and leaf and bark beetles,
but also eats pine seed. At feeders, eats suet and
sunflower seeds

Non-Migratory

11 S

1993

2004

Red Crossbill (Loxia
curvirostra)

Arboreal

Coniferous and mixed coniferous-deciduous forests; also pine savanna and
pine-oak habitat. In migration and winter may also occur in deciduous forest,
and more open scrubby areas. Nests in conifers, 1.5-25 m above ground,
toward outer end of branch

Omnivore

Eats seeds, buds, and insects. Forages in trees; also
picks up seeds from the ground.

Non-Migratory

37 g

1989

2004

692

Red-breasted Nuthatch
(Sitta canadensis)

Arboreal

Prefers forests that have a strong fir and spruce component. May also breed
in mixed woodland when a strong coniferous component is associated with
deciduous trees such as aspen, oak and poplar. The nests are open and built
up from a variety of grasses, strips of bark and pine needles.

Invertivore

Eats mainly arboreal arthropods during the breeding
season and alarge number of conifer seeds outside
the breeding season.

Migratory

10 g

1991

2005

1724

Red-eyed Vireo <^ireo
olivaceus)

Arboreal

Breeds in deciduous and mixed deciduous-coniferous forest. Absent from siti
where understory shrubs are sparse or lacking. Often found near small
openings in forest canopy. Can occur in residential areas, city parks, and
cemeteries where large trees grow. During spring and fell migration uses a
greater variety of forested habitats than during breeding season, but still
prefers deciduous woodland over conifers. Winter range finds them present i
various forested habitats from sea level up to 3000 m elevation.

Invertivore

Consumes mostly insects, particularly caterpillars.
During breeding season most often observed
foraging in canopy vegetation. Also eats various
small fruits, most frequently in late summer and fall
In winter, mostly frugivorous

Migratory

17g

S5B

1993

2000

Red-naped Sapsucker
(Sphyrapicus nuchalis)

Arboreal

nesting in broken-top larch; optimum habitat is old-growth larch, particularly
near wet areas. Nest cavities made in dead trees or dead portions of live
trees. Pure white, moderately glossy eggs are ovate to elhptical-ovate or
rounded-ovate.

Herbivore

Sap wells in the bark of woody plants and feed on
sap that appears there. Often drill sap wells in the
xylem of conifers and aspens. Once the temperature
increase and sap begins to flow, theses birds switch
to phloem wellls in aspen or willow, if available.
Insects, also bast (inner bark), fruit, and seeds.

Migratory

S5B

1992

2006

189

Red-tailed Hawk (Buteo
jamaicensis)

Carnivore

Arboreal/Cl
iffs

nest in trees and on cliffs, and hunt over grasslands, open woodlands, and
agricultural areas.

Carnivore

primarily ground squirrels and other small rodents,
but also feed on awide variety of other animals. Ret
tailed hawks often eat snakes, including rattlesnakes

Migratory

1,224 g

S5B

1989

2006

Red-winged Blackbird
(Agelaius phoeniceus)

Breeds in a variety of wetland and upland habitats. Wetland habitats include
freshwater marsh, saltwater marsh, and rice paddies. Upland breeding
habitats commonly include sedge meadows, alfalfa fields and other crop lane
and old fields. Roosts in habitats with dense cover.

Omnivore

During the nonbreeding season, diet is primarily
plant matter. During breeding season, diet is
primarily animal matter with some plant matter.

Migratory

64 g

S5B

1993

2006

Ring-billed Gull (Laras-
delawarensis)

Riparian

Spring and fall migration prefers fresh water (lakes, river marshes, reservoirs
irigation and agricultural areas). Occurs inland more often than other species
of gulls - near landfill sites, golf courses, farm fields. Winter range mostly oi
or near coast. Common around docks, wharves, harbors; scarce in pelagic
waters; inland on reservoirs, lakes, ponds and streams, landfill sites, and
shopping malls in large metropolitan centers.

Invertivore

fish, insects, earthworms, rodents, and grain.. At
Freezeout Lake, stomach contents included insects,
oligochaetes, crustaceans, birds and mammals, and
plant material believed to be consumed incidentally
to consuming animals

Migratory

566 g

S5B

1991

2006

Ring-necked Duck
(Aythya collaris)

Riparian

Freshwater wetlands, especially marshes, fens, and bogs that are generally
shallow with fringes of flooded or floating emergents, predominantly sedges
interspersed with other vegetation and shrubs; also open water zones
vegetated with abundant submerged or floating aquatic plants (Hohman and
Eberhardt 1998). In the Bozeman area, habitat is restricted to lakes and
ponds.

Omnivore

Moist-soil and aquatic plant seeds and tubers;
aquatic invertebrates

Migratory

730 g

S5B

1992

2006

Rock Wren (
-------
Attachment A-2. Bird Species Occuring within the Libby OU3 Site
Page 15 of 32

Habitat Group

Observations in
Lincoln, Co.,

Common Name
(Genus/ species)

Foraging

Nesting

General Habitat Description

Feeding
Guild

Food

Migration

Longevity

Size

Home Range

Global
Rank

State Rank

Oldest

Most
Recent

Number

Ruby-crowned Kinglet
(Regulus calendula)

Arboreal

In the west, nests in spruce-fir, lodgepole pine and Douglas-fir forests. Spring
and fall migration includes a broad range of habitats: coniferous and
deciduous forests, floodplain forests, willow shrubs, abandoned homesteads
rangeland, old fields, and suburban yards. Nest globular or elongated,
usually pensile but may be placed on limb. In all cases nests protected from
above by overhanging foliage.

Invertivore

Winter: spiders and their eggs, a variety of insects
and their eggs, psuedoscorpions, small amounts of
fruit, seeds and other vegetable matter. Breeding
season: same as winter except no vegetable matter
eaten

Migratory

S5B

1992

2006

500

Ruddy Duck (Oxyura
jamaicensis)

Riparian

Breeding is usually on overgrown, shallow marshes with abundant emergent
vegetation and some open water. Non-breeding birds are found on large,
generally deeper waters with silty/muddy bottoms

Invertivore

primarily aquatic insects, crustaceans, zooplankton,
and other invertebrates. Typically consumes small
amount of aquatic vegetation and seeds. Forage
almost exclusively by diving but occasionally forag
by "skimming" water surface, straining food from
water

Migratory

590 g

S5B

1992

1993

Ruffed Grouse (Bonasa
umbellus)

Ground

Arboreal/Sh
rub

found in dense, brushy, mixed-conifer and deciduous tree cover, often along
stream bottoms. In the Bozeman area they are mostly in deciduous thickets ii
the foothills and mountains; also in riparian areas to the lowest elevation says
they inhabit the denser cover of mixed conifer and deciduous trees and brush
and are often along stream bottoms.

Omnivore

In winter deciduous tree buds and shrubs. In
summer, a mixed diet of insects, green plants and
berries, with young birds eating primarily insect

Migratory

1977

2006

148

Rufous Hummingbird
(Selasphorus rufus)

Riparian

primarily aquatic insects, crustaceans, zooplankton, and other invertebrates.
Typically consumes small amount of aquatic vegetation and seeds. Forage
almost exclusively by diving but occasionally forage by "skimming" water
surface, straining food from water

Invertivore

Migratory

S5B

1991

2007

Savannah Sparrow
(Passerculus
sanchvichensis)

Ground

Arboreal

widespread and abundant in open habitats throughout North America. During
the breeding season its persistent buzzy song can be heard in agricultural
fields, meadows, marshes, coastal grasslands, and tundra. During spring and
fall migration it can be found in open fields, roadsides, dune vegetation,
coastal marshes, edges of sewage ponds and other ponds in open country.

Omnivore

The main foods taken in winter include small seeds,
fruits, and insects when available. During breeding
season they eat adult insects, larval insects, insect
eggs, small spiders, millipedes, isopods, amphipods
decapods, mites, small mollusks, seeds, and fruits.

Migratory

25 g

Territories are small, ranging
from 0.05 to 1.25 hectares

S5B

1992

2004

Say's Phoebe (Sayornis
soya)

Aerial

Open country, prairie ranches, sagebrush plains, badlands, dry barren
foothills, canyons, and borders of deserts

Invertivore

Primarily flying or terrestrial insects, most
frequently wild bees and wasps but also flies,
beetles, and grasshoppers. Little vegetable matter

Migratory

21 g

S5B

Sharp-shinned Hawk
(Accipiter striatus)

Carnivore

commonly use heavy timber, especially even-aged stands of conifers, but
sometimes hunt in open areas

Carnivores

almost entirely on songbirds, although they
occasionally take small mammals and insects

Non-Migratory

174 g

S4B

1991

2003

Solitary Vireo (Vireo
solitarius)

Arboreal

Mixed coniferous-deciduous woodland, humid montane forest; in migration
and winter also in "a variety of wooded habitats, but favors tall woodland
with five oaks and pines in the temperate zone.

Invertivore

Eats mostly insects, some spiders and small fiuits;
forages among foliage and branches of trees and
shrubs. Eats fruits and insects in about equal
proportions

Migratory

17g

SNR

1993

1994

Song Sparrow
(Melospiza melodia)

Arboreal

Wide range of forest, shrub, and riparian habitats, but limited to those
adjacent to fresh water more often in arid environments

Omnivore

In nonbreeding period, primarily seeds, fruits, and
invertebrates, as available. During breeding,
primarily insects and other small invertebrates; som
seeds and fruit

Migratory

21 g

S5B

1991

2006

Sora (Porzana Carolina)

Riparian

Primarily shallow freshwater emergent wetlands (e.g., marshes of cattail,
sedge, blue-joint, or bulrush), less frequently in bogs, fens, wet meadows, an
flooded fields, sometimes foraging on open mudflats adjacent to marshy
habitat.

Omnivore

Eats mollusks, insects, seeds of marsh plants,
duckweed

Migratory

S5B

1991

2000

Spotted Sandpiper
(Actitis macularius )

Riparian

Shores of lakes, ponds, and streams, sometimes in marshes; prefers shores
with rocks, wood, or debris; also mangrove edges in Caribbean. Nests near
freshwater in both open and wooded areas, less frequently in open grassy
areas away from water; on ground in growing herbage or low shrubby
growth or against log or plant tuft

Aquatic
Invertivore

Eats mainly small invertebrates obtained from
surface or by probing along shores or some distance
inland if insects are abundant there

Migratory

40 g

S5B

1992

2006

-------
Attachment A-2. Bird Species Occuring within the Libby OU3 Site
Page 16 of 32

Habitat Group

Observations in
Lincoln, Co.,

Common Name
(Genus/ species)

Foraging

Nesting

General Habitat Description

Feeding
Guild

Food

Migration

Longevity

Size

Home Range

Global
Rank

State Rank

Oldest

Most
Recent

Number

Spotted Towhee (.Pipilo
maculatus)

Ground

Uses a wide variety of shrubby habitats characterized by deep litter and
humus on ground, and sheltering vegetation overhead. Undergrowth of open
woodland, forest edge, second growth, brushy areas, chaparral, riparian
thickets, woodland

Invertivore

Forages on the ground beneath shrubs and
undergrowth, using a two-footed scratching
maneuver to find food among loose debris. Eats
various invertebrates, seeds, small fruits, some smal
vertebrates

Migratory

42 g

S5B

1991

2006

Spruce Grouse
(Falcipennis canadensis)

Ground

dense forest types such as alpine fir, Engelmann spruce, or lodgepole pine.
Winter home ranges northeast of Missoula are covered by Douglas fir,
ponderosa pine, lodgepole pine and larch. Douglas fir provided the most
important cover, the average size being 24.1 hectars

Herbivore

Conifer needles (larch, ponderosa pine, lodgepole
pine) were the main food in late fall through early
spring. In summer, herbaceous vegetation and
insects were utilized.

Migratory

492 g

1992

2004

Steller's Jay
(Cyanocitta steller i)

Ground

Arboreal

Coniferous and mixed coniferous-deciduous forest, open woodland, orchards
and gardens including humid coniferous forest in nw. North America.
Habituates readily to humans and is well known at feeders, picnic areas, and
campgrounds. Nests typically placed on horizontal branches close to trunk,
often close to top of tree. When nesting close to human habitation, frequently
nests close to a window, building, or path, above ground in bushes or trees.

Omnivore

Consumes wide variety of animal and plant food
including arthropods, nuts, seeds, berries, fruits,
small vertebrates, and eggs and young of smaller
birds. At feeders, picnic areas and campgrounds,
consumes wide variety of foods such as suet,
sunflower seeds, peanuts, meat, cheese, bread, and
cookies

Non Migratory

106 g

1987

2005

Swainsoris Thrush
(Cathams ustulaius)

Arboreal

Coniferous and mixed coniferous-deciduous forest, open woodland, orchards
and gardens including humid coniferous forest in nw. North America.
Habituates readily to humans and is well known at feeders, picnic areas, and
campgrounds. Nests usually in small tree, close to trunk, often 2 m or less
above ground; often in conifer, sometimes deciduous tree or shrub.

Omnivore

Berries and insects. Breeding and spring migrating
populations tend to be insectivorous; fall migrating
and wintering populations more frugivorous

Non Migratory/
Altitudinal

23-45 g

Territory sizes of 1.7 to 3.3
acres

S5B

1991

2005

1387

Tennessee Warbler
(Vermivora peregrina)

Arboreal

Openings of northern woodland, edges of dense spruce forests, cleared balsai
tamarack bogs, grassy places of open aspen and pines, alder and willow
thickets, open deciduous second growth. In migration and winter generally ii
single species flocks in tops of trees of various woodland types—not typically
in continuous mature forest; in winter prefers semi-open, second growth,
coffee plantations, gardens. Nests in hollow of moss in bog, or on higher levt
ground or hillside, in thickets or in open at base of grass or shrub

Invertivore
I

Eats insects and spiders, seeds, fruit juices; forages
over terminal twigs and leaves of trees and in dense
patches of weeds

Migratory

lOg

S2S4
B

1991

2000

Townsend's Solitaire
(Myadestes townsendi)

Ground

Open woodland, pinyon-juniper association, chaparral, desert and riparian
woodland nest sites were in cutbanks and 2 were in open woodlands

Invertivore

In Missoula, insects were the primary summer food,
obtained primarily by ground predation. Rocky
Mountain juniper cones were the primary food
during late winter. Feeds oninsects (e.g.,
caterpillars, beetles, wasps, ants, bugs) and fruit
(e.g., juniper berries, and berries of rose, cedar
mistletoe, madrona); also pine seeds. Flies out from
a perch and catches insects in the air.

Migratory

34 g

1991

2004

515

Townsend's Warbler
(Dendroica townsendi)

Arboreal

Tall coniferous and mixed coniferous-deciduous forest at various elevations,
from wet coastal forest at sea level to dry subalpine forest. Most abundant in
unlogged, old-growth forest, but also common in late successional stages.
Uncommon in logged forest. Appears to prefer conifers; may nest 2.7^.5 m
above ground, maybe higher

Invertivore

Insects. Honeydew excreted by scale insects in low-
latitude cloud forests. Winter: gleans small insects
and caterpillars in foliage at all heights, occasionally
hovers and plucks them from undersides of leaves;
hawks flying insects

Non Migratory

S5B

1991

2005

1306

Tree Swallow
(Tachycineta bicolor)

Aerial

Arboreal

Open fields, meadows, marshes, beaver ponds, lakeshores and other wetland
margins. Uses trees only for nesting and occasional roosting.

Invertivore

Mostly flying insects, though vegetable matter is
eaten during unfavorable weather conditions. Forag
over open water, marshes, ponds, and fields, as well
as in shrubby habitat.

Migratory

20 g

S5B

1992

2006

Turkey Vulture
(Cathartes aura)

Carnivore

Cliffs

Turkey vultures forage in a variety of habitats, including grasslands,
badlands, open woodlands, and farmlands. Nesting in the northern Rockies if
usually done on cliff ledges under overhangs, or in rock crevices, often in
river valleys

Carnivore

Carrion is the primary food, but they sometimes
prey on small mammals.

Migratory

1467 g

S4B

1992

2006

-------
Attachment A-2. Bird Species Occuring within the Libby OU3 Site
Page 17 of 32

Habitat Group

Observations in
Lincoln, Co.,

Common Name
(Genus/ species)

Foraging

Nesting

General Habitat Description

Feeding
Guild

Food

Migration

Longevity

Size

Home Range

Global
Rank

State Rank

Oldest

Most
Recent

Number

Varied Thrush (.Ixoreus
naevius)

Ground

Arboreal

Humid coastal and interior montane coniferous forest, deciduous forest with
dense understory, and tall shrubs (especially alder); in migration and winter
also open woodland and chaparral. Usually nests in a small conifer,
sometimes a deciduous tree, 3^.5 m above ground

Omnivore

Feeds in trees or forages on the ground for insects,
earthworms, seeds, and berries.

Non Migratory

78 g

S5B

1990

2005

619

Vaux's Swift (Chaetura
vauxi)

Aerial

Arboreal

During breeding prefer late stages of coniferous forests and deciduous forests
mixed with coniferous. More common in old-growth forests than in younger
stands. During spring and fall migrations prefer forests and open areas; roost
trees and chimneys important as they allow swifts to avoid exposure and
conserve body heat. Hollow trees are its favored nesting and roosting sites.
Nest in hollow trees in the forest; less commonly in chimneys.

Invertivore

Almost entirely insects and spiders. Catches its prey
from the air.

Non Migratory

17g

S4B

1991

2002

Veery (Calharus
fuscescens)

Ground

Riparian

Generally inhabits damp, deciduous forests. Has a strong preference for
riparian habitats in several regions, including the Great Plains. Prefers
disturbed forest, probably because denser understory is not found in
undisturbed forests. Breeds in early-successional, damp, deciduous forests,
often nesting near streamside thickets or swamps. Nest are typically on or
near the ground, most often elevated in or at the base of a bush or small tree.

Omnivore

Primarily a ground forager, with a diet fairly evenly
divided between insects and fruit. Roughly 60%
insects, 40% fruit, feeding primarily on insects as
breeders and on fruit late summer and fall.

Migratory

31g

S4B

1994

1995

Vesper Sparrow
(Pooecetes gramineus)

Ground

In central Montana they nest on the ground under big sagebrush, but
concealment of the nest is not greatly important. They are found in areas
where vegetation was short and dense, with a high percentage of cover

Omnivore

In central Montana, 70-90% of food was animal
(mostly Coleopterans), while 3 to 23% was plant
(mostly grass seeds)

Migratory

27 g

S5B

1991

2006

Violet-green Swallow
(Tachycineta thalassina)

Aerial

Arboreal

Occurs principally in montane coniferous forests. Breeding range includes
open deciduous, coniferous, and mixed woodlands. Often perches on wires
and exposed tree branches.

Invertivore

Flying insects exclusively. Not known to feed on
seeds or berries.

Migratory

14 g

S5B

1991

2006

Warbling Vireo (Vireo
gilvus)

Ground

Arboreal

Throughout range, shows a strong association with mature mixed deciduous
woodlands especially along streams, ponds, marshes, and lakes but sometime
in upland areas away from water. Also found in young deciduous stands that
emerge after a clear-cut. In general, overall habitat structure consists of large
trees with semi-open canopy. Other habitats include urban parks and gardens
orchards, form fencerows, campgrounds, deciduous patches in pine forests,
mixed hardwood forests, and rarely, pure coniferous forests. Usually nests at
end of branch in a deciduous tree, 9-18 m above ground, or 1-3.5 m above
ground, in shrub or orchard tree

Invertivore

Insects, throughout the year. Some fruit in winter

Migratory

12g

Territory sizes of 3.4 to 5.6
acres

S5B

1992

2006

435

Western Bluebird
(Sialia mexicana)

Ground

Can usually be found in open coniferous and deciduous woodlands, parklike
forests, edge habitats, burned areas and where moderate amounts of logging
have occurred, provided a sufficient number of larger trees and snags remain
to provide nest sites and perches. Nests usually found in rotted or previously
excavated cavities in trees and snags, or between trunk and bark.

Invertivore

Insects during the warmer months, but forages
primarily on berries and fruits through the winter.
Forages by flycatching and by dropping from perch
to ground.

Non Migratory

29 g

averaged 0.43 hectares and
0.56 hectares

S4B

1991

2003

Western Grebe
(Aechmophorus
occidentalism

Riparian-
Oppoitunist

Lives on fresh water lakes and marshes which have large areas of open water
and vegetation around it.

Piscivore,
invertivore

Feeds mainly on fish, but will also eat salamanders,
crustaceans, polychaete worms, and insects. They
tend to be opportunists.

Migratory

1477 g

20 hectares or more open water

S4B

1987

1991

Western Kingbird
(Tyrannus verticalis)

Aerial/Groun
d

Arboreal/
Shrub

Open and partly open country, especially savanna, agricultural lands, and
areas with scattered trees, also desert.

Invertivore

Primarily insectivorous; feeds on wasps, beetles,
moths, caterpillars, grasshoppers, true bugs. Also
eats spiders, millipedes, and some fruit. May
occasionally take tree froas

Migratory

40 g

Foraging range at least 400
meters from nest

S5B

1991

2006

Western Meadowlark
(Sturnella neglecta)

Ground

Most common in native grasslands and pastures, but also in hay and alfalfa
fields, weedy borders of croplands, roadsides, orchards, or other open areas;
occasionally desert grassland. Preference shown for habitats with good grass
and litter cover.

Grainivore,
Invertivore

Grain and weed seeds, and insects. Favorite insect
foods include beetles, weevils, wireworms,
cutworms, grasshoppers, and crickets. Seasonal
differences: grain during winter and early spring,
insects late spring and summer, weed seeds in fall.

Migratory

106 g

4-13 hectares

S5B

1992

2006

-------
Attachment A-2. Bird Species Occuring within the Libby OU3 Site
Page 18 of 32

Habitat Group

Observations in
Lincoln, Co.,

Common Name
(Genus/ species)

Foraging

Nesting

General Habitat Description

Feeding
Guild

Food

Migration

Longevity

Size

Home Range

Global
Rank

State Rank

Oldest

Most
Recent

Number

Western Tanager
(Piranga ludoviciana)

Arboreal

Favors open woodlands, but occasionally extends into fairly dense forests.
During migration, frequents a wide variety of forest, woodland, scrub and
partly open habitats and various human-made environments such as orchards
stands of trees in suburban areas, parks, and gardens.

Frugivore,
Invertivore

Feeds predominantly on insects during the breeding
season, but it also incorporates fruits and berries in
its diet whenever it can

Migratory

28 g

S5B

1991

2006

1158

Western Wood-pewee
(Contopus sordidulus)

Aerial

Arboreal

Seen wherever there are clearings or groves of deciduous trees along the rive
valleys

Invertivore

Flying insects, especially flies, ants, bees, wasps,
and beetles, moths and bugs.

Migratory

13g

S5B

1992

2006

White-breasted Nuthatch
(Sitta carolinensis)

Arboreal

A common resident of deciduous forests in North America. Also in mixed
deciduous and coniferous forests. Favors woodland edges over more central
locations, prefering open areas. Over much of its range the presence of some
oaks seems to be a requirement.

Grainivore,
Invertivore

Feeds on a variety of insects and plant matter
(acorns, nuts, etc).

Migratory

21g

10-20 hectares feeding territor}

1992

2006

White-crowned Sparrow
(Zonotrichia leucophrys)

Ground

Ground/Shr
ub/Arboreal

Necessary habitat features of breeding territories include grass, either pure or
mixed with other plants; bare ground for foraging; dense shrubs or small
conifers thick enough to provide a roost and conceal a nest; standing or
running water on or near territory; and tall coniferous trees, generally on
periphery of territory.

Grainivore,
Invertivore

Main foods taken in winter include seeds, buds,
grass, fruits, and arthropods, when available.
During breeding season arthropods (principally
insects) and seeds are taken.

Migratory

29 g

S5B

1989

2003

White-throated Sparrow
(Zonotrichia albicollis)

Ground

Coniferous and mixed forest, forest edge, clearings, bogs, brush, thickets,
open woodland. In migration and winter also in deciduous forest and
woodland, scrub, shrubbery, gardens, parks, cattail marshes.

Frugivore,
Granivore,
Invertivore

Eats mostly weeds seeds, also small fruits, buds,
and insects.

Migratory

26 g

SNA

1994

White-winged Crossbill
(Loxia leucoptera)

Arboreal

Coniferous forest (especially spruce, fir or larch), mixed c oniferous-
deciduous woodland, and forest edge; in migration and winter also may occu
in deciduous forest and woodland

Granivore,
Invertivore

Eats seeds (e.g., of conifers, birches, grasses,
junipers, etc.) and insects; mainly conifer seeds,
which also comprise diet of nestlings

Non-Migratory

28 g

1991

2000

Wild Turkey (Meleagris
gallopavo)

Ground

Open ponderosa pine forest in rugged terrain, interspersed with grassland ant
brushy draws is the preferred habitat (FWP). Open ponderosa pine-grassland
cover types are most widely used in the Longpine Hills during summer and
early fell; canyon bottoms at lower elevations, grain fields and livestock
feeding areas are utilized in late fall and winter.

Frugivore,
Granivore,
Herbivore,
Invertivore

Summer foods include insects (primarily
grasshoppers), bearberry, snowberry and
skunkbrush sumac fruits, grass leaves and stems,
and Carex seeds; winter foods are grains, hawthorn
and snowberry fruits, and grass leaves, stems and
heads.

Non-Migratory

7400 g

260 to 520 hectares

SNA

1994

2005

Williamson's Sapsucker
(Sphyrapicus thyroideus)

Arboreal

Coniferous forest, especially fir and Lodgepole Pine; in migration and winter
also in lowland forest.

Invertivore

Drills holes in trees and consumes sap, cambium
and insects. Ants may comprise 86% of its animal
food; also eats wood-boring larvae, moths of spruce
budworms, etc.

Migratory

48 g

Reported territory sizes vary
from 4 hectares to 6-7 hectares

S3S4
B

1991

2002

Willow Flycatcher
(Empidonax traillii)

Aerial

Riparian

Strongly tied to brushy areas of willow (SAL IX spp.) and similar shrubs.
Found in thickets, open second growth with brush, swamps, wetlands,
streamsides, and open woodland. Common in mountain meadows and along
streams; also in brushy upland pastures (especially hawthorn) and orchards.
The presence of water (running water, pools, or saturated soils) and willow,
alder (ALNUS spp), or other deciduous riparian shrubs are essential habitat
elements.

Invertivore

Eats mainly insects and occasionally berries, 96
percent of diet is animal matter, most of which is
flying insects.

Migratory

14 g

0.1 to 0.9 hectares

S5B

1991

2006

Wilson's Phalarope
(Phalaropus tricolor)

Riparian

Riparian -
ground

During spring, the species is widespread in the valley in lakes, ponds and
flooded fields. Summer birds are restricted to marshy borders of lakes and
ponds

Invertivore

Small aquatic invertebrates in freshwater or
hypersaline environments; also some terestrial
invertebrates.

Migratory

68 g

Usually nests less than 100
meters from shoreline

S4B

1995

Wilson's Snipe
(Gallinago delicata)

Ground

During summer birds are widely distributed in the valley in moist meadows.
In winter, they occur along warm, bog-bordered streams in the valley.
Requires soft organic soil rich in food organisms just below surface, with
clumps of vegetation offering both cover and good view of approaching
predators. Avoids marshes with tall, dense vegetation (cattails, reeds, etc.).

Invertivore

Eats mostly larval insects, but also takes
crustaceans, earthworms, andmollusks. Stomachs
contain as much as 66% plant material, but
probably little or no energy is obtained from plants

Migratory

128 g

Common Snipes breed
throughout the state. Most
wintering records are for
western Montana.

1991

2006

Wilson's Warbler
(Wilsonia pusilla)

Arboreal/ Aeri
al

Ground

Breeding territories are usually located in riparian habitat or wet meadows
with extensive deciduous shrub thickets. Likes edeges of beaver ponds, lakes
bogs and overgrown clear-cuts of montane and boreal zones.

Invertivore

Bees, flies, mayflies, spiders, beetles and
caterpillars. Occasionally eats berries.

Migratory

Ranges from about 0.2 to 2.0
hectares.

S5B

1991

2005

349

-------
Attachment A-2. Bird Species Occuring within the Libby OU3 Site
Page 19 of 32

Habitat Group

Observations in
Lincoln, Co.,

Common Name
(Genus/ species)

Foraging

Nesting

General Habitat Description

Feeding
Guild

Food

Migration

Longevity

Size

Home Range

Global
Rank

State Rank

Oldest

Most
Recent

Number

Winter Wren
{Troglodytes
troglodytes)

Ground/Shra
bs

Arboreal -
Cavity

Coniferous forest, primarily with dense understory and near water, and in
open areas with low cover along rocky coasts, cliffs, islands, or high mtn.
areas, logged areas with large amounts of slash; in winter and migration also
in deciduous woods with understory, thickets, brushy fields.

Invertivore

Eats almost entirely insects (beetles, Diptera,
caterpillars) and spiders.

Migratory

1991

2005

487

Wood Duck (Aix
sponsa)

Riparian/Gro
und

Arboreal -
Cavity

Wide variety of habitats: creeks, rivers, overflow, bottomlands, swamps,
marshes, beaver and farm ponds.

Omnivore

Omnivore with a broad diet. Seeds, fruits and
aquatic and terrestrial invertebrates are main foods
taken.

Migratory

681 g

Home ranges of of fledged
broods range up to 12.8
kilometers.

S5B

1996

2006

Yellow Warbler
(Dendroica petechia)

Arboreal/ Aeri
al

Arboreal/Sh
rub

Found throughout much of North America in habitats categorized as wet,
deciduous thickets. Found especially in those dominated by willows.

Invertivore

Main foods include insects and other arthropods.
May take wild ftuits occasionally.

Migratory

10 g

Breeding territories are as
small as 0.16 hectares.

S5B

1991

2006

Yellow-breasted Chat
(Icteria virens)

Arboreal

Arboreal/Sh
rub

Found in low, dense vegetation without a closed tree canopy, including
shrubby habitat along stream, swamp, and pond margins; forest edges,
regenerating burned-over forest, and logged areas; and fencerows and uplanc
thickets of recently abandoned farmland

Frugivore,
Invertivore

Adults feed on small invertebrates (mainly insects
and spiders), fruit and berries when available.

Migratory

26 g

Territory size averages 1.24
hectares.

S5B

1991

1993

Y ellow-headed Blackbird

(Xcmthocephalus
xa.nthocepha.lus)

Ground

Riparian

Primarily prairie wetlands, but also common in wetlands associated with
quaking aspen parklands, mountain meadows, and arid regions. Scattered
colonies occur on forest edges and on larger lakes in mixed-wood boreal
forest.

Granivore,
Invertivore

During breeding season specializes in "aquatic"
prey; feeds aquatic insects to nestlings. Consumes
primarily cultivated grains and weed seeds during
the postbreeding season.

Migratory

80 g

Forages up to 1.6 kilometers
from nesting area.

S5B

1993

2006

Yellow-ramped Warbler
{Dendroica coronata)

Arboreal/Aeri
al/Ground

Arboreal

Nests in forests or open woodlands. In migration and winter found in open
forests, woodlands, savanna, roadsides, pastures, and scrub habitat.

Invertivore

Feeds on insects (ants, wasps, flys, beetles,
mosquitoes, etc.), spiders, some berries and seeds.

Migratory

13 g

S5B

1991

2005

1716

-------
Attachment A-3. Mammalian Species Occuring within the Libby OU3 Site

Page 20 of 32

Habitat Group

Observation in
Lincoln, Co.,

Common Name
(Genus/species)

Foraging

Breeding,
Resting

General Habitat Description

Feeding
Guild

Food

Migration/
Hibernation

Longevity

Size

Home Range

Global Rank

State Rank

Oldest

Most Recent

Number

Beaver (Castor
canadensis)

Riparian

Ponds, small lakes, meandering streams, and rivers.
Requires water and associated woody vegetation.

Herbivore

variety ofwoody and herbaceous species.
Willows, mountain alder, and aspen

Non-migratory

11 years in
wild

Adults 16-
23 kg (35-
50 pounds),
Kits 0.5 kg
or less (1
pound) at
birth, when
they are
about 38
cm (15
inches)
long

1947

2006

Black Bear (Ursus
americanus)

Ground/ Shr
ub/Arborea
1

Ground

Dense forests; riparian areas; open slopes or
avalanche chutes during spring green-up (FWP).
Habitat use tied to seasonal food avail./plant
phenology. Dry mtn meadows in early spring, snow
slides,stream bottoms, wet meadows early & mid-
summer. May concentrate in berry & whitebark pine
areas in fall. Sympatric with grizzly bear but more
prone to occupying closed canopy areas. Natural cub
and adult mortality low, sub-adult mortality higher.
Dens beneath downed trees, hollow trees, roots or
other shelter.

Omnivore

Grasses, sedges, berries, fruits, inner bark of trees,
insects, honey, eggs, carrion, rodents, occassional
ungulates (especially young and domestic), and
(where available) garbage. Varies. Spring-
primarily vegetation (grasses, umbels, &
horsetails). Summer—herbaceous & fruits. Fall-
berries & nuts, some begetation. Insects a frequent
bomponent of diet. Also mammals, birds, &
carrion

Non-

migratory/Semi-
hibernates in
winter

90 - 240+
kg

1917

2006

Bobcat {Lynx rufus)

Carnivore

Utilizes wide variety of habitats; known to be an
animal of "patchy" country. Prefers rimrock and
grassland/shrubland areas. Often found in areas with
dense understory vegetation and high prey densities.
Natural rocky areas are preferred den sites May be
active during all hours but is primarily nocturnal.
Solitary animal that is difficult to observe in the wild.
In Central MT selected for cover types (52+% canopy
cover) corrected with high prey densities. In W. MT
den sites within caves, btwn boulders, in hollow logs,
or abandon mine shafts.

Carnivore

Snowshoe hares and jackrabbits are the most
common prey. Also feeds heavily on medium-
sized rodents. Will eat carrion.

Non-migratory/

6.7 -15.7
kg

In LA about 5 sq
km for males
and 1 sq km for
females. In
Idaho, home
ranges averaged
42 sq km for
males and 19 sq
km for females

1997

365

Bushy-tailed Woodrat
(.Neotoma cinerea)

Ground

Dens - rock
crevices,
logs

Occurs in crevices where there are large amounts of
sticks, leaves & other debris used to build nest.
Rockslides, rocky slopes, abandoned homesites,
badlands. Occas. lodges nest in tree forks high above
ground

Herbivore

Not selective in its diet of foliage, fruits and seeds
of shrubs & forbs, conifer & fungi.

Non-

migratory/NA

1975

2006

Columbian Ground
Squirrel (Spermophilus
columbianus)

Ground

Intermontane valleys, open woodland, subalpine
meadows, even alpine tundra . Subalpine basins,
clearcuts, and other disturbed areas. At high
elevations, may use rockslides/forage in meadows.
Prefers s-lands & sedges.

Herbivore

Grasses, leafy vegetation, and bulbs. May increase
use of fruits and seeds as season progresses. Uses
a small amount of animal matter: insects, fish,
carrion.

Non-migratory/
Dormacy

340-812 g

1922

2006

-------
Attachment A-3. Mammalian Species Occuring within the Libby OU3 Site

Page 21 of 32

Habitat Group

Observation in
Lincoln, Co.,

Common Name
(Genus/species)

Foraging

Breeding,
Resting

General Habitat Description

Feeding
Guild

Food

Migration/
Hibernation

Longevity

Size

Home Range

Global Rank

State Rank

Oldest

Most Recent

Number

Coyote (Canis latrans)

Scavenger

Utilizes almost any habitat, including urban areas,
where prey is readily available. Prefers prairies, open
woodlands, brushy or boulder-strewn areas. Coyote
abundance is tied to food availability. Mainly
nocturnal, true scavenger, territorial. Occupies
diverse habitats.

Omnivore

Will eat almost anything, plant or animal.
Emphasizes small mammals, fawns, plants, birds,
and invertebrates. During winter, often preys on
deer. Commonly preys on domestic sheep.
Rodents & rabbits imp. year round. Grasshoppers,
crickets, fruits may be used in summer & fall.
Food habits vaiy bet- ween seasons & areas. May
take adult deer in winter. Young deer, elk, &
pronghorn in spring.

Non-migratory

/NA

9 - 22 kg

1999

2006

Deer Mouse
(.Peromyscus
maniculatus)

Ground

Ground-
Burrows

In virtually all habitats - sagebrush desert, grasslands,
riparian areas, montane, subalpine coniferous forests
& alpine tundra. Usually not seen in wetlands. In
forest areas densities peak about 2-5 years after clear-
cutting, then decline as succession advances. 15 yrs.
after cut, uncut & cut densities similar. On prarie
production may be linked to precipitation. Nests in
burrow in ground in trees, stumps and buildings

Omnivore

Omnivorous diet although dentition is adapted for
seed eating. Invertebrates important in warm
months, green plant material a minor but important
component. Stores some food in burrow

Non-

migratory/No
hibernation

Rarely lives
more than 2
years in
wild and
from 5-8
years in
captivity

18 - 35 g

1895

2006

Dusky or Montane
Shrew (Sorex
monticolus)

Ground

Ground -
Beneath
stumps, logs,
trees

High altitude spruce-fir forest, alpine tundra. Non-
breeders territorial. Breeders apparently not territorial.
First-year animals may not be reproductively active.
Nests in stumps, logs, beneath trees.

Invertivore

Similar to other long-tailed shrews: eats mostly
invertebrates

Non-

migratory/NA

2006

Elk (Cervus canadensis)

Ground/Gr
azer

Mainly coniferous forests interspersed with natural or
man-made openings (mountain meadows, grasslands,
burns, and logged areas) (FWP). Varies btwn pops. &
areas. Basic habitat components: securi ty, shelter
(may use to maintain thermal equil.) & forage prod.
Moist sites preferred in sum.

Herbivore

Grasses, sedges, forbs, deciduous shrubs
(especially williow and serviceberry) and young
trees (especially chokecherry and maple), some
conifers (FWP). Varies between ranges.

Migratory in
some areas
(Sun River,
North

Yellowstone)

moving

between

seasonal

ranges, non-

migratory in

others.

14 years in
the wild (25
years in
captivity)

Males (315
450 kg);
Females

(225 - 270
kg)

1977

2006

Fisher (Martes
pennanti)

Carnivore

Ground/Arb
oreal

Although they are primarily terrestrial, fishers are well
adapted for climbing. When inactive, they occupy
dens in tree hollows, under logs, or in ground or rocky
crevices, or they rest in branches of conifers (in the
warmer months). Fishers occur primarily in dense
coniferous or mixed forests, including early
successional forests with dense overhead cover. Dens
in hollow tree or on ground

Carnivore

Mammals (small rodents, shrews, squirrels, hares,
muskrat, beaver, porcupine, raccoon, deer
carrion); also birds and fruit. Snowshoe hares are
an important dietary item for fishers in Montana,
as is deer carrion, known for their skill at killing
porcupines

Fishers are non-
migratory, but
may make
extensive
movements up
to a maximum
of 40

kilometers in 3
days / NA

More than 9
years in
captivity

Males (2.7 -
5-4 kg);
Females
(1.4-3.2

kg)

1965

1992

-------
Attachment A-3. Mammalian Species Occuring within the Libby OU3 Site

Page 22 of 32

Habitat Group

Observation in
Lincoln, Co.,

Common Name
(Genus/species)

Foraging

Breeding,
Resting

General Habitat Description

Feeding
Guild

Food

Migration/
Hibernation

Longevity

Size

Home Range

Global Rank

State Rank

Oldest

Most Recent

Number

Golden-mantled Ground
Squirrel (Spermophilus
lateralis)

Ground

Ground-
Burrows

Occurs throughout the montane and subalpine forests,
where- ever the rocky habitat it dwells in (outcrops
and talus slopes) is present. It will range above
timberline and even (in summer at least) into alpine
tundra. Short, simple, concealed burrows—entrance
near rock, stump, log, or bush

Omnivore

Seeds, fruits, insects, eggs, meat (Burt and
Grossenheider, 1952)

Non-migratory/
Hibernates

170 - 276 g

1966

Gray Wolf (Canis
lupus)

Carnivore

No particular habitat preference except for the
presence of native ungulates within its territory on a
year round basis. Wolves establishing new packs in
Montana have demonstrated greater tolerance of
human presence and disturbance than previously
thought characteristic of this species. They have
established territories where prey are more abundant
at lower elevations than expected, especially in
winter.

Carnivore

Opportunistic carnivores that predominantly prey
on large ungulates. Main prey in Montana include
deer, elk, and moose. Also alternative prey, such
as rodents, vegetation and carrion. Hunt in packs,
but lone wolves and pairs are able to kill prey as
large as adult moose.

Not migratory
but may move
seasonally
following
migrating
ungulates
within its
territory.

31.5-54 kg

1974

2000

Grizzly Bear (Ursus
arctos horribilis)

Ground/Shr
ub

In Montana, grizzlies primarily use meadows, seeps,
riparian zones, mixed shrub fields, closed timber,
open timber, sidehill parks, snow chutes, and alpine
slabrock habitats. Habitat use is highly variable
between areas, seasons, local populations, and
individuals

Omnivore

large vegetative component (more than half) to
their diet and have evolved longer claws for
digging and larger molar surface area to better
exploit vegetative food sources

No true

migration

occurs,

although

grizzly bears

often exhibit

discrete

elevational

movements

from spring to

fall, following

seasonal food

availability/

Hibernates

25 years or
more in
captivity

146-282
kg

S2S3

1912

2003

Heather Vole
(Phenacomys
intermedius)

Ground

Ground-
Burrows

Most common in subalpine spruce-fir forest w/
evergreen shrub ground cover, also in timberline
krummholz, alpine tundra. Sometimes in montane
yellowpine-doug fir forests w/ bearberry-twinflower
understory. Winter nest is a hollow sphere of twigs &
lichens about 6 inches diam., above ground in
protected spot. Summer nest 4-10 in. underground
(Banfield 1974). Does not tend to construct runways.

Herbivore

Twigs, berries

Non-

migratory/NA

1948

2006

Hoary Marmot
(Marmota caligata)

Ground

Talus slopes, alpine meadows, high in mountains near
timberline

Herbivore

herbs, grasses, sedges

Hibernates

3.6-9 kg

S3S4

1949

2006

Long-tailed Vole
(Microtus longicaudus)

Ground

Ground-
Burrows

Riparian valley bottoms to alpine tundra, sagebrush-
grassland semi-desert to subalpine coniferous forests.
In forested areas may not make runways. Subordinate
to other species of voles. Streambanks and
occasionally in dry situations. Nests above ground in
winter and in burrows in summer.

Herbivore

Grasses, bulbs, bark of small twigs.

NA/NA

37 - 57 g

1895

1993

-------
Attachment A-3. Mammalian Species Occuring within the Libby OU3 Site

Page 23 of 32

Habitat Group

Observation in
Lincoln, Co.,

Common Name
(Genus/species)

Foraging

Breeding,
Resting

General Habitat Description

Feeding
Guild

Food

Migration/
Hibernation

Longevity

Size

Home Range

Global Rank

State Rank

Oldest

Most Recent

Number

Long-tailed Weasel
(Mustelafrenata)

Carnivore

Ground-
Burrows

Found in almost all land habitats near water. Has the
broadest ecological and geographical range of the
North American weasels. Prefers areas with abundant
prey. Avoids dense forest, most abundant in late serai
ecotones. Primarily nocturnal, but sometimes active
during the day. Quite fearless and curious. Mainly
terrestrial but can climb and swim well. Nests in old
burrows of other animals . Occupies a diverse range of
habitats. More prone to open country and forest
openings than M. erminea . Common in intermontane
valleys and open foresets where M. erminea is absent.
May occur up to alpine tundra

Carnivore

More of a generalist than the short-tailed and least
weasels. Feeds mostly on small mammals up to
rabbit-sized, but eats birds and other animals as
well

Non-

migratory/No
hibernation

Males (198
340 g);
Females
(85- 198 g)

1940

1992

Lynx {Lynx canadensis)

Carnivore

Subalpine forests between 1,220 and 2,150 meters in
stands composed of pure lodgepole pine but also
mixed stands of subalpine fir, lodgepole pine, Douglas
fir, grand fir, western larch and hardwoods. In
extreme northwestern Montana, primary vegetation
may include cedar-hemlock habitat types

Carnivore

The primary winter food for lynx throughout their
range is the snowshoe hare, comprising 35 to 97%
of their diet. Red squirrels are also an important
prey item, particularly when snowshoe hare
populations are reduced. Summer diets are not as
well known but are probably more varied. Lynx in
Montana probably prey on a wider variety of
species throughout the year because of generally
lower snowshoe hare densities and available
alternate prey

Non-migratory,
but movements
of 90 to 125
miles have been
recorded
between
Montana and
Canada /NA

6.7 -13.5

1941

2005

215

Marten (Martes
americana)

Carnivore

Primarily a boreal animal preferring mature conifer or
mixed wood forests. Severe forest disturbance can
significantly reduce habitat value. Uses deadfall and
snags as den sites. Spends much time in trees but will
also forage on the ground.

Carnivore

Opportunistic feeder that primarily feeds on small
mammals. Meadow voles and red-backed voles
were staples in Glacier NP. Also used Cricetidae,
jumping mice, shrews, ground squirrels, and
snowshoe hares. Use of birds, insects, and fruit
variable by season.

Non-

migratory/NA

17 years in
captivity

Males (754
1248 g);
Females
(681 -851
g)

1945

1966

Masked Shrew (Sorex
cinereus )

Ground

Coniferous forest. In western Montana, where S.
vagrans also occurs, S. cinereus is usually restricted
to drier coniferous forest habitat. Moist situations in
forests, open country, brushland. Nest of dry leaves
or grasses, in stumps or under logs or piles of brush.

Invertivore

Invertebrates, salamanders, small mice. In winter,
seeds may be main item in diet.

Non-

migratory/NA

3 - 6 g

1966

2006

Meadow Vole (.Microtus
pennsylvanicus)

Ground

Ground-
Burrows

Wet grassland habitat but not above timberline in
grassy alpine tundra. Where M. montanus not present,
M. pennsyvanicus may inhabit drier grasslands.

Makes extensive runways. In E MT mean home range
was 0.13 ac. for females, 0.14 ac. for lactating
females, 0.23 ac. for males (McCann 1976). Low
longevity, high juvenile mortality.

Herbivore

Grasses, sedges & herbaceous plants. May use
fungi, particularly endogone. Will use insects.
Occasionally will use carrion. Reported to feed on
apple trees (bark and vascular tissues of lower
trunk and roots)

Non-

migratory/NA

1 to 3 years
in wild

28 - 70 g

1895

2006

-------
Attachment A-3. Mammalian Species Occuring within the Libby OU3 Site

Page 24 of 32

Habitat Group

Observation in
Lincoln, Co.,

Common Name
(Genus/species)

Foraging

Breeding,
Resting

General Habitat Description

Feeding
Guild

Food

Migration/
Hibernation

Longevity

Size

Home Range

Global Rank

State Rank

Oldest

Most Recent

Number

Mink (Mustela vison)

Riparian

Ground

Usually found along streams and lakes. Commonly
occurs in marshes and beaver ponds. Permanence of
water and dependable source of food are most
important habitat components. Often uses den sites of
other animals and is commonly found in association
withmuskrats. Semi-aquatic forager. Can kill prey
larger than itself. Chiefly nocturnal, territorial, and
secretive. Dens underneath piles of brush or
driftwood, under rocks, in hollow logs, and in houses
or dens abandoned by beavers or muskrats.

Piscivore

Preys primarily on small mammals, birds, eggs,
frogs, and fish. Its diet is almost entirely animal.
During summer preys on waterfowl. Order of
importance varies.

Non-migratory.
Males make
extensive
movements and
juveniles
disperse / NA

Males (681
1362 g);
Females

(567 - 1089
g)

1939

1943

Moose (Alces alces)

Ground/Gr
azer

Variable; in summer, mountain meadows, river
valleys, swampy areas, clearcuts; in winter, willow
flats or mature coniferous forests; best ability of any
Montana ungulate to negotiate deep snow

Herbivore

Browse, including large saplings; aquatic
vegetation (FWP). Varies btwn ranges. Winter:
willow, servicebry, chokecherry & redosier
dogwood. Spring/sum—incr. forb use (up to70% of
diet). Some pop.s use aquat. veg. overall

Often uses
separate
summer/winter
ranges.
Movements
prompted by
temperature &
snow depth/ No
hibernation

20 or more
years in the
wild

Males
(382.5 -
531 kg);
Females
(270 -360
kg)

1977

2006

Mountain Cottontail
(Sylvilagus nuttallii)

Ground

Primarily dense shrubby undergroth, riparian areas in
Cen- tral and Eastern MT. In mountains, it uses
shrubby sulleys, and forest edges.

Herbivore

Sagebrush may be a principal food. Grasses also a
preferred food. Juniper sometimes used. May
prefer grasses in spring and summer

Non-

migratory/No
hibernation

0.7 - 1.3 kg

Mountain lion {Puma
concolor)

Carnivore

Mostly mountains and foothills, but any habitat with
sufficient food, cover and room to avoid humans. In
W MT spring-fall ranges at higher elev than winter
areas. Cover types in winter: 42% pole stands, 30%
selectively logged (pole or mature), 18% serai
brushfields

Carnivore

Deer, elk, and pocupines most important in
Montana, but may take prey ranging in size from
grasshoppers to moose (FWP).

Non-

migratory/NA

36 - 90 kg

1975

2007

182

Mule deer (Odocoileus
hemionus)

Ground/Gr
azer

Grasslands interspersed with brushy coulees or
breaks; riparian habitat along prairie rivers; open to
dense montane and subalpine coniferous forests,
aspen groves (FWP). Varies between areas & seasons.

Herbivore

Bitterbush, mountain mahogany, chokecherry,
serviceberry, grasses and forbs

Migratory in
mountain-
foothill
habitats/No
hibernation

Normal in
wild 16
years

Males (56.2
-180 kg)
Females

(45-67.5
kg)

1977

1978

Muskrat (Ondatra
zibethicus)

Riparian

Marshes, edges of ponds, lakes, streams, cattails, and
rushes are typical habitats. An essential habitat
ingredient is water of sufficient depth or velocity to
prevent freezing. The presence of herbaceous
vegetation, both aquatic and terrestrial, is another
essential ingredient. In general, has very flexible
habitat requirements and often coexists in habitats
used by beavers (FWP). Lentic or slightly lotic water
containing vegetation. Typha spp. (cattails) & Scirpus
spp. (bulrushes) usually present. Constructs bank
dens, lodges, feeding huts, platforms, pushups &
canals

Herbivore

Primarily herbivorous and will eat virtually any
vegetable matter. Utilizes shoots, roots, bulbs, and
leaves of aquatic plants. Cattails and bulrush are
preferred foods. Will also consume cultivated
crops. On occasion will eat animal matter. Food is
stored in the burrow or den and during winter may
even eat part of its own lodge

Non-

migratory/NA

908 -1,816
g

1940

2006

-------
Attachment A-3. Mammalian Species Occuring within the Libby OU3 Site

Page 25 of 32

Habitat Group

Observation in
Lincoln, Co.,

Common Name
(Genus/species)

Foraging

Breeding,
Resting

General Habitat Description

Feeding
Guild

Food

Migration/
Hibernation

Longevity

Size

Home Range

Global Rank

State Rank

Oldest

Most Recent

Number

North American
Wolverine (Gulo gulo
luscus)

Carnivore

Caves/Cavity
/Ground/Roc
k

Wolverines are limited to alpine tundra, and boreal
and mountain forests (primarily coniferous) in the
western mountains, especially large wilderness areas.
They are usually in areas with snow on the ground in
winter. Riparian areas may be important winter
habitat. When inactive, wolverines occupy dens in
caves, rock crevices, under fallen trees, in thickets, or
similar sites. Wolverines are primarily terrestrial but
may climb trees. In Montana, most wolverine use in
medium to scattered timber, while areas of dense,
young timber were used least.

Omnivore

Wolverines are opportunistic. They feed on a wide
variety of roots, berries, small mammals, birds'
eggs and young, fledglings, and fish. They may
attack moose, caribou, and deer hampered by deep
snow. Small and medium size rodents and carrion
(especially ungulate carcasses) often make up a
large percentage of the diet. Prey is captured by
pursuit, ambush, digging out dens, or climbing
into trees. They may cache prey in the fork of tree
branches or under snow

Wolverines in
northwestern
Montana and
Alaska tend to
occupy higher
elevations in
summer and
lower

elevations in
winter / NA

More than
15 years in
captivity

7-32 kg

1938

1995

Northern Flying Squirrel
{Glaucomys sabrinus)

Arboreal

Montane and subalpine coniferous forests. Also in
riparian Cottonwood forests. Nests are constructed
either within natural cavities or abandoned
woodpecker holes in dead standing trees, or they are
built over limbs or within witches' brooms

Omnivore

Seeds, fruits, flowers, insects, tree sap, fungus.
Perhaps eggs and meat.

Non-migratoiy

113-185 g

1941

1969

Northern Pocket Gopher
(!Thomomys talpoides)

Ground

Ground-
Burrows

Cultivated fields and prairie to alpine meadows.
Avoids dense forests, shollaow rocky soils and areas
with poor snow cover.

Herbivore

underground plant parts

Non-migratory

18 to 24
months
average in
wild

1966

Pika (Ochotona
princeps)

Ground

Talus slides, boulder fields, rock rubble (with
interstitial spaces adeq. for habitation) near meadows.
Usually at high elevation but mid elevation possible if
suitable rock cover and food plants present

Herbivore

Animals feed on hay individually, stored in small
clumps under rocks, boulders.

Non-

migratory/No
hibernation

Maximum 7

113 -180 g

0.3-0.5 ha and
mean 0.26 ha

1949

2006

Porcupine (Erethizon
dorsatum)

Ground/
Shrub

Dens - rock
crevices,
trees

Common in montane forests of Western Montana,
also occurs in brushy badlands, sagebrush semi-desert
and alon streams and rivers. Rockfall caves, ledge
caves, hollow trees, or brushpiles for dens,

Herbivore

In winter uses cambium, phloem, & foliage of
woody shrubs & trees—Ponderosa Pine, Lodgepole
Pine, perhaps spruce & fir. In spring & summer
uses reprod. parts & foliage of aspen, forbs,
grasses, sedges & succulent wetland vegetation

Non-migratory.
In mountainous
areas seasonal
alti- tudinal
migration may
occur

4.5-12.7

1917

1966

Pygmy Shrew (Sorex
hovi)

Ground

Ground/Cavi
ty

Dry, open coniferous forests (ponderosa pine, western
larch)

Invertivore

Primarily on invertebrates

Non-

misratory/NA

3 - 4 g

1978

2006

Raccoon (.Procyon
lotor)

Riparian

Inhabits stream and lake borders near wooded areas or
rocky cliffs. Most abundant in riparian and wetland
habitats. Uses hollow logs, trees, and rock crevices as
den sites. Forested riparian habitat—river & stream
valleys. Although tree dens are most common,
burrows & crevices, etc. also used.

Omnivore

Carrion, mammals, birds, reptiles, insects,
amphibians, grains, nuts, and fruits.

Non-migratory

/No

hibernation

900-1130
g

-------
Attachment A-3. Mammalian Species Occuring within the Libby OU3 Site

Page 26 of 32

Habitat Group

Observation in
Lincoln, Co.,

Common Name
(Genus/species)

Foraging

Breeding,
Resting

General Habitat Description

Feeding
Guild

Food

Migration/
Hibernation

Longevity

Size

Home Range

Global Rank

State Rank

Oldest

Most Recent

Number

Red fox (Vulpes vulpes)

Carnivore

Ground

Wide range of habitats. Often associated with
agricultural areas. Prefers mixture of forest and open
country near water. Uses dens for shelter during
severe weather and when pups are being reared.
Usually uses dens made by other animals. Seldom
found far from permanent water. Thrive in bushy
successional area where small mammals are most
abundant. Occupies diverse habitats. In forest
situations uses edge. Burrow den-sites comprised of
sub-dens (10-40 holes). Some dens in open and some
in brush.

Carnivore

Opportunistic predator that sometimes eats
carrion. Preys on small mammals, birds, eggs,
game birds. Varies according to avail, in W. MT.
During spring: microtus spp., birds, muskrats,
rabbits, grnd squirrels, deer carrion (in decreasing
order of importance). In winter microtus spp.,
birds, N. pocket gophers. Also uses vegetation.

Non-migratory

/NA

18-31.5 kg

Red Squirrel
('Tamiasciurus
hudsonicus)

Ground

Most common in Montane (Yellow Pine and Douglas
Fir) and subalpine (subalpine fir—Englemann Spruce)
forests in W. MT. Annual fluctuations in density are
large. Correlated with size of seed and cone crops

Herbivore

Conifer cone crops, including serotinous cones.
Opportun- istic. Uses terminal buds, seeds, sap,
berries, bark of a variety of plants. Also uses
fungi. Occasionally carnivorous

Non-

migratory/No
hibernation

198 -250 g

1945

2006

Red-tailed Chipmunk
{Tamias ruficaudus)

Arboreal

Coniferus forests, talus slides, mountains up to
timberline. Most abundant in edge openings.
Sometimes ranges into alpine

Herbivore

Primarily seeds and fruits. Leaves and flowers in
spring, less so in summer. Occasionally uses
arthropods

Non-migratory

1949

1978

Short-tailed Weasel
(Mustela erminea)

Carnivore

Ground-
Burrows

Inhabits brushy or wooded areas, usually not far from
water. Tends to avoid dense forests. Prefers areas with
high densities of small mammals. Most abundant in
ecotones. Mostly nocturnal but will hunt during the
day. Active throughout the year. Dens in ground
burrows, under stumps, rock piles, or old buildings. In
Montana apparently prone to montane forest
associations.

Carnivore

Weasels prey on a variety of small mammals and
birds, they specialize in hunting voles. Mostly
small warm-blooded vertebrates, primarily
cricetidae. Hunts under snow in winter. Females
generally eat smaller prey. May use invertebrates.

Non-

migratory/No
hibernation

Males (71 -
170 g);
Females
(28 - 85 g)

1939

1969

Snowshoe Hare (.Lepus
americanus)

Ground

In W. MT, apparently preferred fairly dense stands of
young pole-sized timber with some use of more open
stands, openings, and edges.

Herbivore

Spring and summer: forbs and grasses. Fall and
winter: more shrubs and sometimes conifer
needles. Occasionally reingests feces. Sometimes
eats sand

Non-

migratory/No
hibernation

Few live
more than 3
years in the
wild.

0.9- 1.8 kg

1986

Southern Red-backed
Vole (Clethrionomys
gapperi)

Ground

Common in dense subalpine forests, also occurs in
more open forest types, even alpine tundra. A favored
prey of marten inNW MT. Populations fluctuate.
Typically does not construct runways. Simple globular
nests (75-100 mm. diam.), lined w/ grass, stems,
leaves or moss.

Herbivore

Vegetative portions of plants, nuts, seeds, berries,
mosses, lichens, ferns, fungi & arthropods

Non-

migratory/NA

14 - 40 g

1949

2006

Striped Skunk (Mephitis
mephitis)

Ground

Ground/Cavi

Variety of habitats including semi-open country,
mixed woods, brushland, and open prairie. Most
abundant in agricultural areas where there is ample
food and cover. Usually absent where water table is
too high for making ground dens. Forest edges, open
woodland, brushy grassland, riparian vegetation,
cultivated lands. Dens in ground burrows, beneath
abandoned buildings, boulders, or wood, or rock piles.

Omnivore

Omnivorous, eating more animal than plant
matter. Propor- tional composition of diet varies.
Small mammals, reptiles, amphibians, berries,
fruit, garbage, cariion, bird eggs, & arthropods.

Non-migratory

/No

hibernation

2.7-6.3 kg

1895

1999

Vagrant Shrew (Sorex
vagrans)

Ground

At elevations below 5000 ft, usually Doug. Fir,
Lodgepole Pine, W. Larch, Grand Fir, W. Red Cedar
forests. Often found in moist sites. Marshes, bogs,
wet meadows, and along streams in forests. Uses
echolocation to orient in darkness.

Carnivore

Insects, annelida, shrews, vegetable matter, insect
larvae. Also uses plant seeds, carrion, and some
mushrooms

Non-

migratory/NA

Few live
more than
16 months.

1895

2006

-------
Attachment A-3. Mammalian Species Occuring within the Libby OU3 Site

Page 27 of 32

Habitat Group

Observation in
Lincoln, Co.,

Common Name
(Genus/species)

Foraging

Breeding,
Resting

General Habitat Description

Feeding
Guild

Food

Migration/
Hibernation

Longevity

Size

Home Range

Global Rank

State Rank

Oldest

Most Recent

Number

Water Shrew (Sorex
palustris)

Riparian

Ground

Streamside habitat in coniferous forests, particularly
in or under overhanging banks or crevices—good
cover. However, also found in seasonal streams and
small seeps. Also above timberline. Nests of dried
sticks and leaves.

Invertivore

Aquatic insect larvae, also some vegetable matter,
oligo- chaetes, other shrews, arachnids, and small
fish

Non-

migratory/NA

9 - 14g

1966

1992

Water Vole (.Microtus
richardsoni)

Riparian

Ground-
Burrows

Semi-aquatic. Near streams & lakes in subalpine and
alpine zones. Normally above 5000 ft. in western
mountains. Moist grass & sedge areas, streamside
hummocks overhung w/ willows. Burrows, runways
& cuttings are conspicuous in summer

Omnivore

Possible heavy use of graminoids. Composite data
from a variety of areas suggest forbs & willows
also eaten. Use of vaccinium, erythronium bulbs,
conifer seeds, insects

Non-

migratory/NA

71 - lOOg

Western Jumping Mouse
(Zapus princeps)

Ground

tall grass along streams, with or without a brush or
tree canopy. Also dry grasslands inN. Central MT.
Mesic forests with sparse understory herbage in W.
MT. From valley floors to timberline & alpine wet
sedge meadows. Nests are in mounds or banks
elevated above surrounding ground (well-drained)
usually 2 feet underground, shredded vegetation
insulative core.

Herbivore

Seeds

Non-migratory/
Hibernates

As long as 6
years in
wild if
survive first
hibernation
(half of all
juveniles
die during
first

hibernation)

18 to 37
grams

1949

2006

White-tailed deer
(Odocoileus
virginianus)

Ground/Gr
azer

River and creek bottoms; dense vegetation at higher
elevations; sometimes open bitterbush hillsides in
winter (FWP). In W MT mature subclimax coniferous
forest, cool sites, diversity & moist sites important in
summer (Leach 1982). In winter prefer dense canopy
classes, moist habitat types, uncut areas & low snow
depths (Berner 1985).

Herbivore

Leaves, twigs, fruits, and berries of browse plants
such as chokecherry, serviceberry, snowberry, and
dogwood; some forbs during summer (FWP).
Browse most imp. statewide - yr. round,
particularly so in winter. Graminoiduse increases
in spring, forb use in late spring & sometimes in
fall.

Uses summer
range, winter
range in W MT
may be 8.69-15
mi. apart.

Up to 16.5
years in the
wild.

Males (33.7
-180 kg);
Females
(22.5 -
112.5 kg)

1978

2006

Yellow pine chipmunk
{Tamias amoenus)

Ground

Ground-
Burrows

Open stands of ponderosa pine and Douglas fir. Nest
chamber in burrow averaging 11 inches below surface.
Open coniferous forests, chaparral, rocky areas with
brush or scattered bines, burned over areas.

Herbivore

Fruits and seeds and a few insects

Non-migratory/
Hibernates

5 years or
more in the
wild

38 - 71 gran

1860

2006

Yellow-bellied Marmot
{Marmota flaviventris)

Ground/Ro
ck Slopes

Dens - Talus
slopes, rock
outcrops

Semi-fossorial. Inhabits talus slopes or rock outcrops
in meadows. Abundant herbaceous & grassy plants
nearby. Rocks support burrows & serve as sunning &
observ. posts. Avoids dense forests. Rarely in holl riv
bot fid pin c-wood trees. Occurs from valley bottoms
to alpine tundra where suitable habitat exists. Where
Marmota caligata occurs, M. flavi - ventris is
restricted to lower elevations.

Herbivore

Grasses, flowers, forbs—in late summer eats seeds.
Mode- rate grazing by ungulates may favor
marmots. Likes alfalfa

Non-migratory,

although

dispersal

movements

may be

observed/

Hibernates

2.2-4.5 kg

1949

-------
Attachment A-4. Fish Species Occuring within the Libby OU3 Site
Page 28 of 32

Observation in Lincoln,
Co., Montana

Common Name
(Genus/species)

General Habitat Description

Food Habits

Global
Rank

State
Rank

Oldest

Most
Recent

Number

Black Bullhead
(Ameirurus melas )

Turbid, mud bottomed lakes and ponds; also pools and
backwaters of streams. Tolerates high water temperatures and
low levels of dissolved oxygen.

Omnivorous. Mostly aquatic insects, crustaceans,
mollusks, fish, and vegetation matter. Young feed during
day, while adults feed at night.

SNA

1996

Brook Trout
(Salvelinusfontinalis )

Prefers small spring fed streams and ponds with sand or gravel
bottom and vegetation. Clear, cool water . Spawns over gravel in
either streams or lakes with percolation;spring areas in lakes.

Feed mainly on aquatic insects and other small aquatic
invertebrates throughout life. Larger individuals may eat
small fish

SNA

1960

2006

Brown Trout (Salmo
trutta)

Valley portions of larger rivers where gradients are low and
Summer temperatures range from 60-70 degrees F. Also
reservoirs and lakes at similar elevation with suitable spawning
trib.

Feeds largely on underwater aquatic insects. Also uses
many other small organisms available and large
individuals eat many small fish

SNA

2006

Bull Trout
(Salvelinus
conjluentus )

Sub-adult and adult fluvial bull trout reside in larger streams and
rivers and spawn in smaller tributary streams, whereas adfluvial
bull trout reside in lakes and spawn in tributaries. They spawn in
headwater streams with clear gravel or rubble bottom.

Young feed on aquatic insects. The adults are piscivorous.

1960

2004

Burbot (Lota lota)

Large rivers and cold, deep lakes and reservoirs. Spawn in
shallow water, usually in rocky areas.

Young feed on aquatic invertebrates. Adults are
piscivorous

SNA

1993

Channel Catfish
(Ictalurus punctatus )

Prefers large rivers and lowland lakes. Thrives at water
temperatures above 70 degrees. Tolerates turbid water.

Omnivorous feeder. Uses almost any living or dead
organisms available.

2006

Common Carp
('Cyprinus carpio )

Primarily lakes and reservoirs, moderately warm water and
shallows. Also rivers, pools and backwaters. Congregates in
areas of organic enrichment. Tderates turbid water and low
dissolved oxygen; avoids cold and swift, rocky streams. Spawns
in shallow weedy areas

An omnivorous feeder with vegetation and detritus
making up bulk of diet. May feed on any available aquatic
organism including eggs.

SNA

2006

Fathead Minnow
(Pimephales
promelas )

Habitat is highly variable but found mostly in small turbid creeks
and shallow ponds of flatlands. Very tolerant of extreme
conditions found in a prairie environment (turbid water, high
temperature, and low dissolved oxygen).

Variety of minute aquatic plants and animals.

S4S5

1998

Kokanee Salmon
(Oncorhynckus
nerka)

Cold, clear lakes and reservoirs and Kokanee Salmon are found
at all depths. They spawn over loose rubble, gravel, and sand in
lower portions of tributary streams or along lake shores

The diet consists mostly of plankton. Micro-crustacea are
most important, but midges and other aquatic insects are
often taken

SNA

2002

Largescale Sucker
(Catostomus
macrocheilus )

Found in both streams and lakes. Spawns in gravel riffles with
strong current or along lake margins

Almost any available organism found on the substrate

1993

2003

Longnose Dace
(Rhinichthys
cataractae")

Habitat variable. Found in lakes, streams, springs. Preferred
habitat is riffles with a rocky substrate

Eats mostly immature aquatic insects picked off the rocks.
Small amounts of algae and a few fish eggs are also eaten

2000

2006

Longnose Sucker
(Catostomus
catostomus )

Cold, clear streams and lakes; sometimes moderately warm
waters and turbid waters. Spawns over loose gravel beds in riffle
areas.

Considerable algae, midge larvae, and most aquatic
invertebrates

1996

2006

Mottled Sculpin
(Cottus bairdi)

Prefer riffle areas of fast-flowing streams that are clear and have
rocky bottoms.

Variety of immature aquatic organisms, but midge and
acddis larvae are by far the most important. A study in
southwest Montana showed bottom-dwelling aquatic
insects comprising 99.7% of the diet.

1953

1991

Mountain Whitefish
(Prosopium
williamsoni)

Medium to large cold mountain streams. Also found in lakes and
reservoirs. Normally a stream spawner in riffles over gravel or
small rubble but has been seen spawning along lake shorelines.

Mostly on aquatic insects but also takes terrestrial insects
which fall into water. May eat fish eggs, but rarely fishes
Feeds actively in Winter. Zooplankton important in lakes.

1969

2006

-------
Attachment A-4. Fish Species Occuring within the Libby OU3 Site
Page 29 of 32

Observation in Lincoln,
Co., Montana

Common Name
(Genus/species)

General Habitat Description

Food Habits

Global
Rank

State
Rank

Oldest

Most
Recent

Number

Northern Pikeminnow
(Ptychocheilus
oregonensis )

Prefers lakes and slow - flowing streams of moderate size.
Young usually school in shallow water near lake shores and in
quiet backwaters of streams

Most kinds of aquatic invertebrates. Adults frequently eat
small fish. Considered a serious predator on young salmon
and trout

1952

2006

Peamouth
(Mylocheilus
caurirms )

Shallow weedy zones of lakes or rivers.

Young feed mainly on micro-crustaceans. Adults eat
micro-crustaceans, snails, adult aquatic and terrestrial
insects. Occasionally small fish.

2006

Rainbow Trout
(Oncorhynckus
my kiss )

Cool clean streams, lakes, res., farm ponds. Able to withstand
wider range of temperatures than most trout. Spawns in streams
over gravel beds.

Feed mainly on aquatic insects but eat what is available to
them. Large adults also eat fish. River populations mostly
insect eaters while zooplankton and forage fish are
important in Lake Koocanusa.

1976

2006

Redside Shiner
(Richardsonius
bcdteatus )

Lakes, ponds, and larger rivers where current is weak or lacking.

Young feed mainly on plankton and adults eat mostly
aquatic insects and snails.

2002

2006

River Carpsucker
(Carpiodes carpio )

Reservoirs and the pools and backwaters of rivers. Spawn in
larger streams with backwater areas.

Mostly diatoms, desmids, and filamentous algae. Also
aquatic invertebrate larvae.

2006

Slimy Sculpin
(Cottus cognatus )

Rocky riffles of cold, clear streams, but it is sometimes found
along the rubble beaches of lakes, especially near the mouths of
inlet streams

Mostly immature aquatic insects and invertebrates, but
also includes any small fish available

1950

2006

Smallmouth Bass
(Micropterus
dolomieu )

Prefers clear cool water and rocky substrates in both rivers and
lakes. In streams, it prefers riffle areas with clean bottoms. In
lakes, it prefers rocky shorelines, reefs, out- croppings, gravel
bars, etc.

Feeds on most available item. Fry feed on zooplankton
and small mayflies. Adults feed heavily on fish, frogs, and
aquatic invertebrates. Seems to prefer crayfish, if
available.

SNA

2006

Torrent Sculpin
(Cottus rhotheus )

Riffles of cold, clear streams, but are also taken in lakes. They
hide near stones on the bottom.

The fry eat mostly plankton. Adults feed mainly on aquatic
insects and a variety of invertebrates, but also include
plankton. Larger individuals often eat small fish.

1950

2006

Westslope Cutthroat
Trout

(Oncorhynckus
clarkii lewisi)

Spawning and rearing streams tend to be cold and nutrient poor.
Seek gravel substrate in riffles and pool crests for spawning.
Sensitive to fine sediment. Require cold water. Thrive in streams
with more pool habitat and cover than uniform, simple habitat.
Juveniles overwinter in the interstitial spaces of large stream
substrate. Adult need deep, slow moving pools that do not fill
with anchor ice in order to survive the winter.

G4T3

1960

2006

White Sturgeon -

Acipenser

transmontanus

Data are taken from: http://fieldguide.mt.gov/

Montana Species Ranking Codes: Montana employs a standardized ranking system to denote global (G - range-wide) and state status (S) (NatureServe 2003). Species are assigned numeric ranks
ranging from 1 (critically imperiled) to 5 (demonstrably secure), reflecting the relative degree to which they are "at-risk". Rank definitions are given below. A number of factors are considered in
assigning ranks - the number, size and distribution of known "occurrences" or populations, population trends (if known), habitat sensitivity, and threat.

G1 SI

At high risk because of extremely limited and potentially declining numbers, extent and/or habitat, making it highly vulnerable to global extinction or extirpation in the state.

G2 S2

At risk because of very limited and potentially declining numbers, extent and/or habitat, making it vulnerable to global extinction or extirpation in the state.

G3 S3

Potentially at risk because of limited and potentially declining numbers, extent and/or habitat, even though it may be abundant in some areas.

G4 S4

Uncommon but not rare (although it may be rare in parts of its range), and usually widespread. Apparently not vulnerable in most of its range, but possibly cause for long-term concern.

G5 S5

Common, widespread, and abundant (although it maybe rare in parts of its range). Not vulnerable inmost of its range.

-------
Attachment A-5. Reptile Species Occuring within the Libby OU3 Site

Page 30 of 32

Observation in Lincoln,
Co., Montana

Common Name
(Genus/species)

General Habitat Description

Food Habits

Global
Rank

State
Rank

Oldest

Most
Recent

Number

Common Gartersnake
(Thamnophis sirtalis)

Found in nearly all habitats, but most commonly at lower
elevations around water. Prefer moist habitats and are found
most often along the borders of streams, ponds and lakes. They
may travel long distances (4 to 17 kilometers) from hibernacula
to forage in preferred habitat

Variety of vertebrates and invertebrates.

1954

2006

Eastern Racer
('Coluber constrictor)

Associated with relatively open habitats either in shortgrass
prairie or forested areas. Very fast and active, prey on insects and
small vertebrates such as mice and frogs. Females lay a clutch of
three to seven eggs in summer. In the NW racers generally
absent from dense forest/hi mtns.

Orthopterans can form a major part of diet and have been
re- ported as food in NC MT. Small mammals, lizards,
orthopterans, anurans are all major components of diet.

1991

Gophers nake
(.Pituophis catenifer)

Dry habitats, including open pine forests. Occasionally climb
trees.

Rodents, rabbits, ground-dwelling birds, and to a lesser
extent lizards.

1993

1994

Northern Alligator
Lizard (.Elgaria
coerulea)

Little specific information on habitat associations in Montana.
South-facing slopes in fine to course talus, sometimes in the
open, but often with some canopy cover of Douglas-fir,
ponderosa pine, a variety of shrubby species (serviceberry,
ninebark, mock orange), and a litter layer of dried leaves and
conifer needles .

An invertivore, northern alligator lizards feed on insects,
ticks, spiders, centipedes, millipedes, slugs and snails.

1949

2006

Painted Turtle
(Chrysemys picta)

NA (web page not available)

1955

2006

Rubber Boa
(Charina bottae)

Usually found under logs and rocks in either moist or dry forest
habitats. They are primarily nocturnal, but occasionally may be
observed sunning on roads, trails, or in open areas.

Feed primarily on small mice but also take shrews,
salamanders, snakes, and lizards.

1980

2004

Terrestrial Gartersnake

(Thamnophis
elegans)

Found in nearly all habitats, but most commonly at lower
elevations around water. Common near water but also found
away from water. At high elev. common on rocky cliffs/ brushy
talus .

They eat a variety of vertebrates and invertebrates.

1952

2006

Data are taken from: http://fieldguide.mt.gov/

Montana Species Ranking Codes: Montana employs a standardized ranking system to denote global (G - range-wide) and state status (S) (NatureServe 2003). Species are assigned numeric ranks ranging
from 1 (critically imperiled) to 5 (demonstrably secure), reflecting the relative degree to which they are "at-risk". Rank definitions are given below. A number of factors are considered in assigning ranks -
the number, size and distribution of known "occurrences" or populations, population trends (if known), habitat sensitivity, and threat.

G1 SI

At high risk because of extremely limited and potentially declining numbers, extent and/or habitat, making it highly vulnerable to global extinction or extirpation in the state.

G2 S2

At risk because of very limited and potentially declining numbers, extent and/or habitat, making it vulnerable to global extinction or extirpation in the state.

G3 S3

Potentially at risk because of limited and potentially declining numbers, extent and/or habitat, even though it may be abundant in some areas.

G4 S4

Uncommon but not rare (although it may be rare in parts of its range), and usually widespread. Apparently not vulnerable in most of its range, but possibly cause for long-term concern.

G5 S5

Common, widespread, and abundant (although it may be rare in parts of its range). Not vulnerable in most of its range.

-------
Attachment A-6. Invertebrate Species Occuring within the Libby OU3 Site

Page 31 of 32

Observation in Lincoln,
Co., Montana

Common Name
(Genus/species)

General Habitat Description

Global
Rank

State
Rank

Oldest

Most
Recent

Number

Freshwater Sponge
(Heteromeyenia baileyi)

Aquatic

S1S3

1997

Stonefly (Utacapnia
columbiana)

Aquatic

The larvae occur on the upper surfaces and sides of cobbles and boulders in moderate gradient, fast
flowing, foothills to mountain streams. Inhabits streams with moreintermediate characteristics between the
higher elevation, cold mountain streams (more likely to find Glossosoma & Anagapetus), and the large
warmer transitional rivers downstream (more likely to find Prototila). Generally the riparian canopy of the
occupied streams is mostly (>50%) open, and less shaded than mountain streams. In clear streams and
rivers during low flows, it is typical to be able to locate & identify Agapetus larvae on the tops of rocks. In
relation to trophic status, A. montanus larvae scrape, graze and digest algae and diatoms from the surfaces
of rocks.

Banded Tigersnail
(Anguispira kochi)

Terrestrial

SNR

2005

2007

Blue Glass (Nesovitrea
binneyana)

Terrestrial

SNR

2007

Brown Hive (.Euconulus
fulvus)

Terrestrial

SNR

2005

2007

Coeur d'Alene Oregonian
(Cryptomastix mullani)

Terrestrial

SNR

2005

2007

Land Snail, Cross Vertigo
(Vertigo modesta)

Terrestrial

SNR

2006

2007

Land Snail, Fir Pinwheel
(Radiodiscus abietum)

Terrestrial

S2S3

1959

2007

Land Snail, Forest Disc
(.Discus whitneyi)

Terrestrial

SNR

2005

2007

Slug, Giant Gardenslug
(L imax maximus)

Terrestrial

Common in gardens and buildings, and margins of native forests, does not seem to penetrate far into
undistrubed forests, although it can be abundant in modified forest remnants and secondary forests. This
nocturnal slug feeds primarily on decaying plant material and fungi, but because it shows aggresive
behavior towards other slugs, it is often erroneously regarded as a predator

SNA

2005

Slug, Gray Fieldslug
(Deroceras reticulatum)

Terrestrial

SNA

2007

Land snail, Hedgehog
Arion (Arion
intermedius)

Terrestrial

Often locally abundant in pastures, hedgerows, plantation forests, and in native forests. It can penetrate
deep into undisturbed forest from areas disturbed by humans

SNR

2007

Land snail, Idaho
Forestsnail (.Allogona
ptychophora)

Terrestrial

SNR

2005

2007

Slug, Magnum Mantleslug
(.Magnipelta mycophaga)

Terrestrial

Low- to mid-elevation sites, often with water in the general vicinity. Moist, cool sites in relatively
undisturbed forest with an intact duff layer, such as are found in moist valleys, ravines, and talus areas, are
preferred. Forest canopy composition at sites includes Picea engelmannii, Pseudotsuga menziesii, Pinus
ponderosa, Pinus albicaulis, Larix occidentalis, Abies lasiocarpa , and Abies grandis, often with Alnus
present; spruce-fir appears to be the most frequent forest association. Often found on the ground under
pieces of loose bark, logs, loose stones, and in rotted wood; surface active on cool (10-16wet and overcast
days, probably most active at night.

SI S3

2005

2007

Slug, Meadow Slug
(Deroceras laeve)

Terrestrial

Cliff, Cropland/hedgerow, Forest - Conifer, Forest - Hardwood, Forest - Mixed, Forest Edge,
Forest/Woodland, Grassland/herbaceous, Old field, Savanna, Shrubland/chaparral, Suburban/orchard,
Urban/edificarian, Woodland - Conifer, Woodland - Hardwood, Woodland - Mixed

SNA

2005

2007

Land snail, Multirib
Vallonia (Vallonia
gracilicosta)

Terrestrial

G5Q

SNR

2007

Land snail, Orange-
banded Arion {Arion
fasciatus)

Terrestrial

Damp areas and wet meadows adjacent to streams

GNR

SNR

2007

-------
Attachment A-6. Invertebrate Species Occuring within the Libby OU3 Site

Page 32 of 32

Observation in Lincoln,
Co., Montana

Common Name
(Genus/species)

General Habitat Description

Global
Rank

State
Rank

Oldest

Most
Recent

Number

Darner damselfly, Paddle-
tailed Darner (Aeshna
palmata)

Terrestrial

Found in most habitats, including warm springs; found far from water

1994

Slug, Pale Jumping-slug
(.Hemphillia camelus)

Terrestrial

S1S3

2005

2007

Slug, Pygmy Slug
(Kootenaia burkei)

Terrestrial

Forest - Mixed, Fallen log/debris, forested and adjacent to a perennial water body. Found on forest floor
mostly, either on or under woody debris, mats of moss, or deciduous tree leaves; two specimens collected
0.2 m aboveground on moss-covered tree trunk along stream edge

S1S2

2005

2007

Land Snail, Quick Gloss
(.Zonitoides arboreus)

Terrestrial

SNR

2005

2007

Land Snail, Robust
Lancetooth (Haplotrema
vancouverense)

Terrestrial

S1S2

2006

Land Snail, Rocky
Mountainsnail
(Oreohelix strigosa)

Terrestrial

Composition of the plant community appears to be of little importance, dominant plant species ranges from
sagebrush to a wide variety of deciduous shrubs and trees and a similarly wide variety of coniferous shrubs
and trees. Substrate, however, is of great importance, the presence of exposed limestone being almost
critical for occurrence; exceptions, however, are well known, there being documented occurrences on
sandstone, and occurrences on other substrates probably exist. Slope, too, has been considered to be of
importance. Herbivorous.

SNR

2005

2006

Slug, Sheathed Slug
(Zacoleus idahoensis)

Terrestrial

Moist microsites in relatively intact Pseudotsuga menziesii, Pinus ponderosa, and Picea engelmannii
forests in moist valleys, ravines, and talus on both north- and south-facing slopes. Meadows and cedar
swamps, white pine stands, spruce valleys, rockslides, and near springs.

G3G4

S2S3

1959

2007

Land Snail, Smoky
Taildropper (Prophysaon
humile)

Terrestrial

S1S3

2005

2007

Land Snail, Spruce Snail
(Microphysula ingersolli)

Terrestrial

G4G5

SNR

2005

2007

Land Snail, Striate Disc
(.Discus shimekii)

Terrestrial

Found most often in litter in rich lowland forest, generally on shaded, north-facing slope bases, often
bordering or ranging slightly onto stream floodplain. Usually on limestone soils. Species will crawl on
downed wood and is sometimes seen on rock surfaces. Primarily feeds on partially decayed deciduous
tree leaves and degraded herbaceous vegetation.

1959

Land Snail, Subalpine
Mountainsnail
(Oreohelix subrudis)

Terrestrial

SNR

2007

Western Pearlshell
(Margaritiferafalcata)

Aquatic

Cool-coldwater running streams that are generally wider than 4 m, perferrable habitat is stable sand or
gravel substrates. Found in hard as well as soft water. This species occurs in sand, gravel and even among
cobble and boulders in low to moderate gradient streams up to larger rivers.

S2S4

1992

1996

Data are taken from: http://fieldguide.mt.gov/
Inc

G1 SI

At high risk because of extremely limited and potentially declining numbers, extent and/or habitat, making it highly vulnerable to global extinction or extirpation in the state.

G2 S2

At risk because of very limited and potentially declining numbers, extent and/or habitat, making it vulnerable to global extinction or extirpation in the state.

G3 S3

Potentially at risk because of limited and potentially declining numbers, extent and/or habitat, even though it may be abundant in some areas.

G4 S4

Uncommon but not rare (although it may be rare in parts of its range), and usually widespread. Apparently not vulnerable in most of its range, but possibly cause for long-term concern.

G5 S5

Common, widespread, and abundant (although it may be rare in parts of its range). Not vulnerable in most of its range.

-------
FINAL

ATTACHMENT B
SUMMARY OF SITE-SPECIFIC SURFACE WATER TOXICITY TESTS

-------
ATTACHMENT B
SITE-SPECIFIC TOXICITY TESTS IN FISH

1.0 OVERVIEW

As discussed in Section 3 of the main text, site-specific toxicity studies are often a useful line of
evidence in ecological risk assessment. At OU3, EPA, working in concert with the Libby OU3
BTAG, determined that site-specific studies of the toxicity of LA-contaminated water would
provide one valuable line of evidence to evaluate risks to fish in OU3. Several alternative study
designs were pursued, as described below.

2.0 EXPOSURE OF FISH TO SITE WATER

The first study that was implemented to evaluate risks to fish from LA in water involved
exposure of rainbow trout fry to water collected directly from the site. The study is described in
detail in Parametrix (2009a). A summary is provided below.

Study Design

The study design was specified in the Phase II Part A Sampling and Analysis Plan (SAP) of the
RI for OU3 (EPA 2008c). The water sample used for testing was collected from the tailings
impoundment in OU3. Triplicate analysis of LA in this sample (measured before the toxicity test
began) showed that the concentration was about 21 MFL. This concentration is in the middle to
upper end of the range of LA concentrations that have been observed in surface water samples
from OU3.

The test was conducted with newly hatched larval (sac fry) rainbow trout (Oncorhynchus mykiss)
under static renewal conditions for an exposure duration of 6 weeks. Organisms were exposed to
the undiluted site water (21 MFL) as well as five serial 1:10 dilutions of the site water. A control
group (no LA) was also evaluated. During the test, the water was renewed every ten days during
the sac-fry exposure (days 0-20) and every three days following swim-up of the organisms (days
20-42). Survival, behavior, and growth were observed during the exposure period. At the end of
the test, fish were sacrificed and examined for the occurrence of pathological lesions.

Results from this study showed no significant change in any measure of effect in fish exposed to
site water when compared to controls (Parametrix 2009a). However, analysis of water samples
taken from the test aquaria during the study revealed that asbestos concentrations were
significantly lower than expected. For example, the concentration of LA in the aquaria
containing undiluted site water at the end of the first exposure cycle (day 10) had fallen from the
expected value of 21 MFL to below the analytical detection level (0.05 MFL). Further
investigations (detailed in Parametrix 2009a) indicated that the most likely reason for the low
concentrations was that LA in the water tended to become clumped with organic material in the

B-l

-------
water, and that a substantial fraction of the LA became bound to the walls of the aquaria and/or
the stock bottle. Based on this, EPA concluded that the exposure of the fish to LA in these
toxicity tests could not be reliably quantified, and therefore the results of this study could not be
used to draw reliable conclusions about risks to fish exposed to LA in site waters.

3.0 EXPOSURE OF FISH TO WATER SPIKED WITH LA

EPA and the BTAG then considered performing toxicity tests using LA added to laboratory
water, rather than using site water. The hope was that laboratory water would contain lower
levels of the organic material and microbial organisms that likely were responsible for the losses
observed in the site water studies. An initial pilot study was performed by Oregon State
University (OSU 2011) to evaluate the maximum duration that LA fibers added to laboratory
water could remain in a free (un-bound) state before fiber "loss" due to clumping, binding,
settling, etc. occurred. Rainbow trout fry were exposed in four different LA asbestos
concentrations, plus a dilution water control, for a period of 3 days. The nominal test
concentrations were 10 billion LA fibers per liter (BFL), 1 BFL, 0.1 BFL, 0.01 BFL and the
control. Samples for both total LA and free-fiber LA analyses were sampled from each
concentration and each replicate on each day of the test. Subsequent analysis of some of the
samples indicated that concentration were substantially lower than the expected nominal
concentrations (OSU 2011, SRC 2011). Based on this, EPA and the BTAG concluded that
spiking studies with normal laboratory water were subject to the same problems as studies with
site waters.

EPA and the BTAG next evaluated an alternative study design in which exposure would occur to
ozonated laboratory water spiked with LA. Ozonation is known to destroy living organisms and
biological materials in water, and helps improve the precision of analyses of asbestos in water
(EPA 1994). The logic was that if LA was added to sterile water that was entirely free from
living organisms and organic material, the problems of clumping and binding of LA could be
minimized. However, the design of such a study is complicated by two key issues, as discussed
below.

Issue 1: Form of LA in Site Water

Examination of site waters indicates that LA may occur in both a free form (individual
fibers), and as "clumps" in which multiple LA fibers exist bound to an organic material.
This was first recognized by TEM analyses of site waters in which occasional clumps of
LA were observed on the filters. The presence of clumps in site waters was further
demonstrated by noting that treatment of site waters with ozone in accord with EPA
Method 100.1 tended to increase the apparent concentration by several fold (EPA 2013b).
Consequently, if a study was successfully implemented with exposure to "free" (un-
clumped) fibers, this might or might not provide a useful basis for estimation of hazards
to fish exposed to a mixture of free and clumped fibers in site waters.

B-2

-------
Issue 2: Potential Loss of Fibers During Laboratory Tests

The second factor that complicated the design of a spiked water toxicity test was a
concern that LA spiked into laboratory water might still be subject to clumping and
binding due to growth of bio-films in bottles and tubing and on aquaria walls as the study
progressed. If uncontrolled, this could lead to a tendency for decreased exposure levels
to LA as the bio-films formed and grew, similar to the problem encountered in the first
study. If so, this could make it difficult to interpret the results of such a study.

EPA and the BTAG met several times to discuss the best approach for measuring free and
clumped fibers in water samples, and for designing a toxicity study using LA-spiked ozonated
laboratory water. With regard to the first issue, the BTAG decided that, if it were possible to
evaluate the toxicity of free fibers, those data could be used to provide a bounding estimate of
risks from site water by assuming that the toxicity of free and clumped fibers was equal.
However, before committing to the implementation of such a study, EPA and the BTAG decided
to perform a series of pilot tests to evaluate the second issue and determine if exposures to
controlled levels of free fibers could be achieved in ozonated water.

The pilot studies that were performed are summarized in SRC (2011). In brief, these studies
demonstrated that even when water was treated by ozonation to provide initially sterile
conditions, decreases in LA concentrations still occurred during subsequent storage and dilution
of the water, and that LA was also lost over time when the water was placed into aquaria. Based
on this, EPA and the BTAG decided that implementation of a study using spiked ozonated water
would be unlikely to provide reliable data, and the effort was not pursued further.

4.0 CONCLUSION

Based on the studies described above, EPA and the BTAG concluded that exposure of fish to LA
under laboratory conditions, using either site water or laboratory water spiked with LA, was
subject to technical difficulties that precluded the ability to reliably control and maintain the
exposure levels. Consequently, this approach was not used at OU3.

B-3

-------
FINAL

ATTACHMENT C

AVIAN RESPIRATORY SYSTEM
Overview of Anatomy and Function as Related to Particulate Inhalation

Report prepared for EPA by

Robert F. Wideman, Jr., Ph.D.

-------
1

Avian Respiration Synopsis as related to the Libby Amphibole; EPA 2011 RFW

AVIAN RESPIRATORY SYSTEM:

Overview of Anatomy and Function as Related to Particulate Inhalation

Robert F. Wideman, Jr., Ph.D.
rwideman@uark.edu
479-575-4397

INTRODUCTION

The avian respiratory system performs the following functions: gas exchange; thermoregulation;
phonation; olfaction; air filtration/cleansing; blood filtration; regulation of acid-base balance;
and, production and metabolism of blood-borne molecules. This summary will focus first on the
macroscopic and microscopic anatomy of the extra- and intra-pulmonary airways and their
connections to the air sacs. Patterns of air flow during inspiration and expiration then can be
summarized. Finally the defense mechanisms that protect the respiratory system from inhaled
particulates and the evidence pertinent to avian particulate inhalation will be reviewed. Extensive
reviews of avian respiratory structure and function have been published elsewhere (Jukes, 1971;
King and Molony, 1971; Duncker, 1974; Nickel et al., 1977; McLelland and Molony, 1983;
King and McLelland, 1984; Fedde, 1986, 1998; Brackenbury, 1987; Scheid and Piiper, 1987;
King, 1993; Brown et al., 1997). Animated images of air flow patterns through the lungs and air
sacs can be found at: http://people.eku.edu/ritchisong/birdrespiration.html. The descriptions
contained in the present overview pertain primarily to the respiratory system of the domestic
fowl.

ANATOMY

Nasal Passages: Depending on the species, the external nasal apertures (nares) at the base of the
upper beak may be protected by opercula (partial or complete flaps) or cere and ricti (ridges of
skin). Feathers arising from the cere may cover the nares. The nasal cavities contain turbinate
bodies consisting of convoluted mucosa-covered cartilage. The nasal cavities open through the
choana (medial fissure in the "hard" palate) into the pharynx (common passageway for food,
water and air). The slit-like glottis guards the opening from the pharynx into the larynx, and
prevents non-aerosol foreign matter (e.g., food and water) from entering the trachea.

Conducting Airways: the trachea conducts air into the thoracic cavity and bifurcates at the
syrinx (the avian organ of phonation) to form the right and left extrapulmonary primary
bronchi. These bronchi penetrate the respective lungs to become the intrapulmonary primary
bronchi (Figure I). The conducting airways up to this point are reinforced externally with
cartilage rings that maintain flexibility while preventing airway collapse. The unilobar lungs are
located lateral to the vertebral column in the dorsal thorax. The dorsal-lateral border of each lung
interdigitates between 5 ribs, thus approximately 25% of the total lung volume is encased
between the ribs (Figures 2 and 3). Within the lungs of domestic fowl, the medioventral (4
each), mediodorsal (8 each), lateroventral (8 each), and laterodorsal secondary bronchi (23-
30 each) branch from the intrapulmonary primary bronchus (Figures I a ). These secondary
bronchi are not supported by external cartilage rings.

-------
47

Avian Respiration Synopsis as related to the Libby Amphibole; EPA 2011 RFW

Gas Exchange Airways: Arching between the medioventral and mediodorsal secondary bronchi,
arcades of long cylindrical paleopulmonic parabronchi (tertiary bronchi) (Figures 2 and 4) are
layered adjacent to one another in a roughly hexagonal array (when viewed in cross section;
Stearns et al., 1987). Individual parabronchi are separated from each other by a thin
interparabronchial connective tissue septum containing interparabronchial arteries and veins
(FIgii 6). Approximately 500 paleopulmonic parabronchi are found in each lung of

domestic fowl. They measure up to 4 cm long, have a uniform outside diameter of 1.5-2 mm and
a lumen diameter of 0.5 mm. Between 100 and 300 freely anastomosing neopulmonic
parabronchi connect the lateroventral and laterodorsal secondary bronchi (Figure 4).
Neopulmonic parabronchi measure up to 1 cm long and comprise 20-25% of the total
parabronchial volume.

A simple squamous epithelium lines the parabronchial lumen, but this epithelium is not the site
of gas exchange. Instead, as shown in Figures 5 and 6 thousands of atria 100-200|im in
diameter form pockets projecting 50|im into the luminal wall. The epithelial cells lining the atria
produce surfactant, which coats the inner surfaces of conducting airways and gas exchange
membranes. Spiral bands of innervated smooth muscle underlie the parabronchial luminal
epithelium and encircle the opening to each atrium (atrial muscle. Figure 6). Elastic fibers
encase the walls (septa) and floor of the atria, presumably serving a support function. One or
more funnel-shaped infundibula penetrate from the atrial floor into the parabronchial wall, with
multiple freely anastomosing air capillaries originating from each infundibulum (Figures 5 and
6). The air capillaries average 8 to 15 |im in diameter and penetrate outward from the
infundibulum, extending 200-500 |im to the outer periphery of the parabronchial wall adjacent to
the interparabronchial septum (Figure 6). Each air capillary is surrounded by a profusion of
blood capillaries derived from intraparabronchial arterioles that branch inward into the
parabronchial wall from the interparabronchial arteries. Gas exchange occurs at the blood-gas
barrier, at the interface between blood capillaries and air capillaries (Figure 7).

Air Sacs: Air enters and exits the air sacs via ostea that connect with the intrapulmonary primary
bronchi, branches of the secondary bronchi, and terminal neopulmonic parabronchi (Figures I
ai ). Domestic fowl possess eight air sacs, including one clavicular, one cervical, two cranial
thoracic, two caudal thoracic, and two abdominal sacs (Figures 1 and 3). The thin, transparent
nonstratified squamous epithelium of the air sacs is poorly vascularized and plays essentially no
role in the gas exchange process. The air sac membrane contains small islands of ciliated and
secretory cells, and is supported by diffuse elastin fibers (McLelland, 1989). Functionally, the air
sacs serve as elastic, inflatable internal reservoirs for "fresh" and "stale" air. In conjunction with
the thoracic and abdominal musculature, the air sacs also act in a bellows-like fashion to propel
air through the parabronchi. The extensive penetration of air sacs throughout the thorax,
abdomen and skeleton accounts for serious concerns regarding carcass contamination that arise
when air sacculitis is detected during inspection of poultry at processing plants (King and
McLelland, 1984). To simplify further discussion, it is convenient to group the clavicular,
cervical and cranial thoracic sacs in the category of cranial air sacs, and the caudal thoracic and
abdominal sacs in the category of caudal air sacs.

-------
93

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

Avian Respiration Synopsis as related to the Libby Amphibole; EPA 2011 RFW

AIR FLOW DURING INSPIRATION AND EXPIRATION

Avian lungs remain essentially fixed in volume throughout the respiratory cycle, and thus the
lungs neither appreciably inflate during inspiration nor deflate during expiration. The current
consensus is that all intrapulmonary air channels remain open and relatively fixed in volume
throughout the respiratory cycle. Consequently, air must be forced to flow through the
intrapulmonary conducting airways by the bellows-like action of the air sacs. A saccopleural
membrane is anchored by skeletal muscle (costoseptal muscle) to the internal thoracic wall and
covers the ventral lung surface. This membranous structure is penetrated by the ostea to the
caudal air sacs and, unlike the mammalian diaphragm, the avian saccopleural membrane does not
contribute to the development of a negative intrathoracic pressure. The costoseptal muscles
apparently contract during expiration to hold the ostea open (King and McLelland, 1984). Thus
birds lack a functional diaphragm and must depend entirely on the contraction and relaxation of
thoracic and abdominal muscles during inspiration and expiration.

During inspiration the rib cage and sternum expand to more cranial and ventral positions,
increasing the thoracic volume and generating a negative intrathoracic pressure (suction).
Simultaneous relaxation of the abdominal muscles coupled with the forward excursion of the
sternum and gravitational pull on the visceral organs increases the volume of the abdominal
cavity. The resulting negative thoraco-abdominal pressures (-1 cm H20) serve to inflate (draw air
into) the cranial and caudal air sacs simultaneously (Figure 8, upper panel). "Fresh" air enters
the trachea and is drawn through the extra- and intra-pulmonary primary bronchi toward the
caudal air sacs. This incoming air does not enter the medioventral secondary parabronchi due to
their acute caudally-directed angle of insertion along the intrapulmonary primary bronchus.
Instead, the incoming fresh air is drawn caudally to: (a) mix with and carry end expiratory stale
air from the trachea and primary bronchus, through the neopulmonic parabronchi and into the
caudal air sacs; (b) supply the neopulmonic parabronchi and caudal air sacs with fresh air; and,
(c) flow through the mediodorsal secondary bronchi, pushing the resident stale air out of the
paleopulmonic parabronchi, through the medioventral secondary bronchi and into the cranial air
sacs. Thus the caudal air sacs are inflated mainly with fresh air, and the cranial air sacs are
inflated mainly with stale air from the paleopulmonic parabronchi (Figure 8, upper panel).
Throughout the respiratory cycle, ongoing gas exchange occurs between the blood capillaries
and air capillaries. Consequently, with the cessation of fresh air inflow at the end of inspiration,
parabronchial air once again becomes stale (PCO2 increases, PO2 decreases).

During expiration the rib cage and sternum are drawn inward to more caudal and dorsal
positions, reducing the thoracic volume and generating a positive intrathoracic pressure.
Simultaneous contractions of the abdominal wall muscles reduce the volume of the abdominal
cavity. The resulting positive thoraco-abdominal pressures (+1 cm H2O) partially deflate the
cranial and caudal air sacs (Figure 8, lower panel). The stale air from the cranial air sacs flows
through the medioventral secondary bronchi, into the primary bronchus and then cranially out
through the trachea. The relatively fresh air in the caudal air sacs is forced cranially, and due to
aerodynamic valving most of the air exiting the caudal air sacs first perfuses the neopulmonic
parabronchi and then flows through the mediodorsal secondary bronchi. After entering the
mediodorsal secondary bronchi, the relatively fresh air flows through the paleopulmonic
parabronchi. The stale air that is displaced from the paleopulmonic parabronchi flows, along

-------
139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

Avian Respiration Synopsis as related to the Libby Amphibole; EPA 2011 RFW

with stale air from the cranial air sacs, through the medioventral secondary bronchi into the
primary bronchus and out through the trachea (Figure 8, lower panel). Aerodynamic valving
within the conducting airways insures that the cranial air sacs always serve as a reservoir for
stale air exiting the parabronchi during inspiration, whereas the caudal air sacs mainly serve as a
reservoir for fresh air to supply the parabronchi during expiration. This flow of "fresh" air during
inspiration and expiration always is unidirectional in the paleopulmonic parabronchi
(mediodorsal secondary bronchus to medioventral secondary bronchus), but is bidirectional in
the neopulmonic parabronchi (e.g., air flow cessation and reversal occur in the neopulmonic
parabronchi during each respiratory cycle, as well as in all air sacs).

As shown in Figures 6 and 7, each parabronchus can be modeled as a long tube with air
capillaries (resembling the bristles of a bottle brush) radiating outward at right angles from the
parabronchial lumen. During inspiration and expiration, rapid convective air flow occurs along
the lumen of the parabronchus. Convective air flow may carry air as deep as the infundibula
(Stearns et al, 1987). However, O2 must move through the gas exchange region of the
parabronchus by the relatively slow process of diffusion from the infundibulum to the periphery
of the air capillaries, across the blood-gas barrier1, through the plasma, and into the red blood
cells (Powell, 1982; Scheid and Piiper, 1987). Blood capillaries carry deoxygenated blood
inward (convective blood flow) following the air capillaries back to their junction with the
infundibulum near the parabronchus lumen. Because convective air flow occurs longitudinally
down the lumen of the parabronchus, whereas blood flow and gas exchange occur in a transverse
path across the radius of the parabronchial wall, the pattern of blood flow and air flow in avian
lungs has been labeled a cross-current exchange system. When compared with mammalian
respiratory systems, the cross-current avian respiratory system permits a higher degree of
removal of O2 from respiratory air, and provides exceptional advantages at low atmospheric
pressure (low PO2), as confirmed by the exceptional tolerance of birds to high altitude. Sparrows
are able to fly at an atmospheric pressure of 349 mmHg, corresponding to an altitude of 6100 m,
while mice are comatose and nearly unable to crawl under identical conditions (Schmidt-Nielsen,
1975).

RESPIRATORY SYSTEM DEFENSES

Nasal Passages: Feathers covering the nares serve to coarsely filter the incoming air. Turbulent
air flow within the nasal passageways forces the inhaled air to swirl over the mucosal surfaces of
the turbinate bodies. The air becomes humidified (fully saturated with water vapor), warmed to
the bird's body temperature, and cleansed of larger particulates that adhere to the mucus.
Additional particulate entrapment is likely to occur as the inhaled air flows through the moist,
narrow choanal slit in the hard palate and flows over the moist surfaces of the pharynx and
glottis (Hayter and Besch, 1974; Fedde, 1998; Brown et al., 1997).

Conducting Airways: The avian trachea, primary bronchi, and initial roots of secondary bronchi
are lined with a mucociliary epithelium (a pseudostratified, longitudinally folded ciliated
epithelium with mucous-secreting goblet cells). Pathogens and airborne particles become trapped

1 The blood-gas barrier is composed of the blood capillary endothelium and its basal lamina, the thin air capillary
epithelium, and a thin layer of surfactant. In chickens, the endothelium comprises 67% of the barrier thickness, the
basal lamina comprises 21%, and the epithelium plus surfactant comprise only 12% of the barrier thickness.

-------
182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

Avian Respiration Synopsis as related to the Libby Amphibole; EPA 2011 RFW

in the mucus, and ciliary action sweeps the mucous cranially (at a rate of 10 mm/min; Fedde,
1998) to the oral cavity where it is swallowed or expectorated (King and Molony, 1971; King
and McLelland, 1984). In addition to mucus, the fluids lining avian conducting airways contain
antioxidants and surfactant binding proteins that assist in binding and neutralizing inhaled
pathogens and antigens (Bottje et al., 1998; Zeng et al., 1998; Johnston et al., 2000). When
mammals and birds of similar sizes are compared, the avian trachea is approximately 2.7X
longer and has a 1.3X larger radius, which yields a 4X greater tracheal volume. (King and
McLelland, 1984). Accordingly, the mucociliary escalator has a substantially enhanced
opportunity to trap pathogens and particulates in birds when compared with mammals. The
mucociliary escalator is an active and highly important line of defense in birds, preventing many
aerosol particulates and pathogens from entering the gas exchange parenchyma. For example,
poultry reared on floor litter are chronically challenged with air-borne dust, bacteria, and potent
antigens (Anderson et al., 1966; Hayter and Besch, 1974; Gross, 1990; Whyte, 1993; Brown et
al., 1997; Zucker et al., 2000; Bakutis et al., 2004; Lai et al., 2009). Only modest changes in
respiratory function can be detected when broiler chickens (meat-type chickens bred for
extremely fast growth and breast muscle accretion) reared on floor litter are compared with
broilers reared in much cleaner environments (Bottje et al., 1998; Wang et al., 2002; Lorenzoni
and Wideman, 2008). Commercial poultry populations reared on floor litter typically grow
rapidly, thrive and reproduce while exhibiting minimal mortality levels. Furthermore, necropsies
of clinically healthy broilers reared on floor litter overwhelmingly reveal healthy tracheas,
almost pristine air sacs (e.g., uniformly clear and transparent membranes), and macroscopically
unremarkable lungs (Wideman et al., 2011).

In commercial poultry the respiratory system becomes dramatically more susceptible to damage
if mucociliary transport is inhibited by exposure to noxious gasses (e.g., ammonia) and
pathogens such as infectious bronchitis virus (IBV), infectious laryngotracheitis (ILT), avian
influenza (Al), Newcastle disease virus (ND), and Mycoplasma gallisepticum. For example, IBV
causes ciliostasis and distinctive symptoms of upper airway distress (gasping, coughing,
gurgling) attributable to obstruction of the trachea by mucus accumulation. Inhibition of the
mucociliary escalator in combination with distressed patterns of breathing apparently allow
pathogenic bacteria and aerosolized respirable particles to penetrate more readily into the lung
parenchyma and air sacs. The ensuing pulmonary inflammation and air sacculitis (infection of
the air sacs) are profoundly deleterious (Gross, 1961, 1990; Tottori et al., 1997; Yamaguchi et
al., 2000).

Bronchus-associated lymphoid tissues (BALT) constitutively develop in the bronchial mucosa at
the junctions of primary and secondary bronchi, and at the ostea to the air sacs of clinically
healthy birds (Reese et al., 2006). BALT contain lymphocytes (B cells and T cells), lymphoid
nodules, and epithelial cells. The mucosal BALT tissues may functionally compensate for the
absence of fully formed lymph nodes in birds, although their specific role remains to be
elucidated (Reese et al., 2006).

Gas Exchange Airways and Air Sacs: Whereas the overwhelming majority of airborne particles
exceeding 5 |im in diameter are trapped in the nasal cavities and trachea, some of the smaller
respirable particles averaging <5 |im in diameter do reach the avian parabronchi and abdominal
air sacs (Hayter and Besch, 1974; Mensah and Brain, 1982; Stearns et al., 1987; Fulton et al.,

-------
228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

Avian Respiration Synopsis as related to the Libby Amphibole; EPA 2011 RFW

1990). Respirable particles can be heavily contaminated with a wide range of immunogenic
substances including pathogens and toxins (Bakutis et al., 2004). Macrophages and neutrophils
play a central role in the mammalian responses to aerosolized particulates, and intra-alveolar
macrophages serve as a first line of defense at mammalian gas exchange surfaces. In contrast,
healthy birds do not appear to maintain large populations of resident macrophages or other
resident leukocytes at their gas exchange surfaces (air capillaries) or within their air sacs,
although some macrophages have been detected in the atria and infundibula of the parabronchi,
as well as in the larger conducting airways (Maina and Cowley, 1998; Nganpiep and Maina,
2002). The primary phagocytic function within avian parabronchi apparently resides within the
epithelial cells lining the atria and infundibula (the same cells that secrete surfactant). These
phagocytic endothelial cells engulf particles encountered on their luminal (air space) surface.
The internalized particles then may be degraded/digested intracellularly, or they undergo
exocytosis to the underlying interstitium. There they are engulfed by resident macrophages
located in the spaces between the atrial and infundibular epithelial cells (Stearns et al., 1987;
Brown et al., 1997; Reese et al., 2006). Large numbers of macrophages can be induced to enter
the air sacs by injecting appropriate antigens or pathogens into the air sac lumen (Fedde, 1998;
Reese et al., 2006). During respiratory infection or aspiration of particulates, phagocytic
macrophages and heterophils (analogous to mammalian neutrophils) can be found in lavage fluid
from the avian respiratory tract, indicating mechanisms do exist that allow substantial
populations of phagocytic leukocytes to enter the gas filled spaces when necessary (Ficken et al.,
1986; Toth and Siegel, 1986; Toth et al., 1987, 1988; Qureshi et al., 1993; Klika et al., 1996;
Lorenzoni et al., 2009; Maina and Cowley, 1998; Nganpiep and Maina, 2002). Intratracheal
instillation of C. parvum or E. coli effectively increased the number of phagocytes collected by
lung lavage within 24 h (Toth et al., 1987). Additionally, macrophages have been reported to
migrate into air capillaries in a variety of infectious diseases, including toxoplasmisis, fatal viral
hydropericardium syndrome, highly pathogenic infectious bursal disease and highly pathogenic
avian influenza (Hower, 1985; Abe et al., 1998; Nakamura et al., 2001). Pathways by which
macrophages that have engulfed pathogens or foreign particles are cleared from the lung
parenchyma and air sacs remain to be elucidated. Phagocytosed materials may be transported and
presented to the local BALT, or they may be transported to peripheral lymphoid organs (e.g., the
spleen) (Fedde, 1998; Reese et al., 2006).

Vascular Defenses: Blood-borne particulates and antigens also trigger intrapulmonary immune
responses. In addition to particles or pathogens entering the blood stream directly, materials
engulfed by lymphatic capillaries subsequently flow through major lymph trunks that empty into
the vena cava. Thus the lungs perform the important function of filtering and clearing the
returning venous blood of micro- and macro-particulates including bacteria and thrombi, as well
as other potent antigens translocated from pathogens resident in the intestine or from sites of
infection (Weidner and Lancaster, 1999). In some mammalian species blood-borne antigens are
primarily removed from the blood stream by pulmonary intravascular macrophages (PIMs),
which are large mature macrophages bound to the pulmonary capillary endothelium. However,
resident PIMs are not present in chickens (Lund et al., 1921; Winkler, 1988; Staub, 1994;
Warner et al., 1994; Brain et al., 1999; Weidner and Lancaster, 1999). The absence of PIMs does
not leave chicken's lungs immunologically unresponsive to blood-borne antigens because the
entire blood volume and thus all of the circulating leukocytes flow through the lungs (e.g., the
lungs receive 100% of the cardiac output via the pulmonary circulation). For example,

-------
274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

Avian Respiration Synopsis as related to the Libby Amphibole; EPA 2011 RFW

intravenously injected cellulose microparticles (30|im diameter) become entrapped in inter- and
intra-parabronchial pulmonary arterioles of broiler lungs. Within 20 minutes post-injection the
microparticles trigger marked pulmonary inflammatory responses, including perivascular
infiltration of mononuclear cells in combination with luminal accumulations of macrophages.
During the ensuing 48 hours occlusive particles are surrounded by granulomatous tissue
consisting primarily of macrophages, giant cells, and fibrous tissue. Subsequently virtually all of
the microparticles are cleared from the lungs within approximately 3 weeks post-injection, the
inflammatory response subsides, and the lung parenchyma again returns to an entirely normal
(e.g., non-inflamed, unobstructed) histological appearance (Wideman et al., 2002, 2007, 201 la,b;
Wang et al., 2003; Hamal et al., 2008, 2010). Avian lungs possess an impressive ability to
eliminate (digest), clear (remove), or segregate (wall off) offending particulates.

DISTRIBUTION, DEPOSITION AND CLEARANCE OF INHALED PARTICULATES:

RELEVANT RESEARCH SYNOPSIS

Peacock and Peacock (1965) injected finely ground asbestos fibers suspended in tributyrin (a
triglyceride ester of glycerol and butyric acid) into the clavicular air sacs of adult White Leghorn
chickens. The injected material spread throughout the air sac and entered the lung parenchyma.
Immediate responses were inflammatory, with macrophages engulfing the asbestos fibers and
clearing them from the air sacs (presumably into sub-epithelial spaces). Neoplastic and
granulomatous tumors formed near the site of injection in 4 out of 30 injected birds. The
granulomatous tumor contained asbestos fibers. Evidently the majority of injected birds lived for
>3 years. Necropsies conducted 4 years post-injection revealed asbestos fibers remaining in the
lung parenchyma, and "asbestos bodies" (asbestos fibers engulfed by macrophages or encased in
mineralized connective tissue) were indentified in the "interalveolar septa" (presumably the
interatrial septa where clusters of resident macrophages have been demonstrated in chickens by
Reese et al., 2006).

Hayter and Besch (1974) evaluated the distribution of aerosolized spherical particles in
spontaneously breathing adult roosters. Larger particles (>3.7|im diameter) primarily were
deposited in the nasal passageways and cranial segment of the trachea, although a portion of
these particles also entered the caudal air sacs. Smaller particles (<1.1 nm diameter) tended to
avoid entrapment in the upper airways and instead were distributed to the lungs and caudal air
sacs. Particles were considered to accumulate preferentially at locations where branching of the
conducting airways (e.g., rapid amplification of the cumulative luminal cross-sectional area
caudal to the syrinx) caused abrupt reductions in air flow velocities, or where reversal of air flow
occurred (e.g., in the caudal air sacs) (Hayter and Besch, 1974).

Brambilla et al. (1979) retrospectively evaluated pulmonary lesions in tissues saved during
routine necropsies of 11 mammalian and 8 avian species that had chronically inhaled air
containing high levels of silicate particles (1 to 10|im in length) while residing at the San Diego
Zoo. All of the avian species exhibited severe silicate dust deposition in the tertiary bronchi
(parabronchi), accompanied in some individuals by the formation of large granulomas composed
of crystal laden macrophages. Fibrosis and necrosis were absent, and none of the birds had been
reported to have respiratory problems. Particles deposited in the conducting airways evidently

-------
320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

Avian Respiration Synopsis as related to the Libby Amphibole; EPA 2011 RFW

were effectively cleared by mucociliary escalator, whereas those engulfed by parabronchial
epithelial cells or macrophages were much more difficult to clear and, consequently, triggered
ongoing immunological responses. When compared with mammals, all of the avian species
evaluated in this study appeared to be more susceptible to parenchymal silicate dust retention and
granuloma formation (birds were less capable of clearing particulates reaching the non-ciliated
secondary and tertiary bronchi), but birds were significantly less susceptible to pulmonary
fibrosis (Brambilla et al., 1979).

Mensah and Brain (1982) evaluated the deposition and clearance rates for aerosolized particles
(< 0.8|im diameter) in unanesthetized spontaneously breathing hens. Particles of this size were
only sparsely deposited in the trachea but considerable deposition was detected in both lungs.
More particles accumulated in caudal than cranial portions of the lungs, presumably reflecting
preferential particle deposition in the neopulmonic parabronchi where air flow velocities
decrease and then abruptly reverse direction. Almost half of the particles had been cleared from
the lungs within 1 hour post-inhalation, and 65% of the particles were cleared from the lungs
within 12 hours. This rapid phase of clearance presumably reflects the activity of the mucociliary
escalator, which appears to be considerably more vigorous in birds than the more sluggish
clearance rate for similarly sized particles deposited in mammalian lungs. As particles were
cleared from the lungs they accumulated in the gastrointestinal tract (presumably after the
tracheal mucus was swallowed) and were eliminated in the feces. Approximately 35% of the
particles persisted in the lung parenchyma through the end of the study (36 hours), presumably
reflecting the proportion engulfed by parabronchial epithelial cells and interstitial macrophages.
Particles also accumulated in pneumatized bones that are penetrated by cranial air sacs,
indicating significant numbers of particles streamlined completely through the paleopulmonic
parabronchi and thus were dispersed into the cervical and clavicular air sacs (Mensah and Brain,
1982).

Nakaue, Pierson and Heifer (1982) and Bland, Nakue, Goeger and Heifer (1985) evaluated the
performance and health responses of broiler chickens exposed to Mount St. Helen's volcanic ash
(VA; particles ranging from 0.5 to 10 |im diameter). The VA was applied directly to the wood
shavings litter on the pen floor, or was blown daily (for 20 consecutive days) into pens with
resident birds. When compared with unexposed control birds, none of the modes of VA exposure
altered any of the routine indices of broiler performance, including final body weights, feed
conversion, carcass quality, and cumulative mortality. Litter moisture and ammonia levels also
were unaffected by VA, suggesting the absence of significant damage to the kidneys and
gastrointestinal tract. Aerosol induction of VA did not alter the histological appearance of the
turbinate bodies or the trachea, but pathological changes within the lungs were detected in a
portion of the birds beginning 4 days post-exposure. Macrophages initially phagocytized the VA
dust within secondary and tertiary bronchi. More chronically, a mild lymphoid hyperplasia
developed, including the formation of granulomas containing giant cells surrounding
phagocytized crystalline material (Nakaue et al., 1982; Bland et al., 1985).

Stearns et al. (1987) exposed spontaneously breathing adult female ducks to aerosolized iron
oxide (0,18jam diameter). The ducks were euthanized 24 hours post-exposure, and transmission
electron microscopy was used to evaluate particle deposition within the parabronchial
parenchyma. Particle clearance from the parabronchial lumen followed a distinctive sequence:

-------
366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

Avian Respiration Synopsis as related to the Libby Amphibole; EPA 2011 RFW

(a) entrapment in the relatively thick layer of surfactant; (b) phagocytosis by the luminal surface
membranes of atrial and infundibular epithelial cells (the same cells that secrete surfactant); (c)
movement of the phagosome to the basal-lateral surfaces of the epithelial cells; (d) exocytosis of
the particles into the interstitial spaces; and, (e) phagocytosis of the particle by atrial and
infundibular interstitial macrophages (macrophages were not seen on the epithelial/luminal
surface). The disposition of the particles after their phagocytosis by interstitial macrophages was
not assessed. Relatively few particles were observed in the air capillaries per sc\ leading to the
interpretation that aerosolized particles were distributed to the atria and infundibula primarily by
convective air flow (Stearns et al., 1987).

Brown et al. (1997) reviewed the structure and function of avian respiratory system in relation to
its susceptibility to damage by inspired particles and toxins. Deposition patterns for aerosolized
particles of different sizes and shapes were predicted based on the anatomy of the airways and
the physical forces acting on the particles (e.g., inertial forces, gravitational sedimentation, and
Brownian diffusion). Inertial impaction was predicted to clear larger particles primarily in the
nasal passageways, pharynx, larynx, trachea, syrinx, and points where secondary bronchi branch
from intrapulmonary primary bronchi. Gravitational sedimentation and Brownian diffusion were
predicted to occur where air velocities are low and particle residence time is prolonged,
particularly within the air sacs and parabronchi (Brown et al., 1997).

SYNTHESIS FROM THE AVAILABLE INFORMATION

1. Particle size distributions for the Libby Amphibole (LA) in duff (Figure 9) indicate that, if
suitably aerosolized, well over half of these particles are small enough to be distributed
throughout the avian respiratory system, including to the level of the parabronchial atria and
infundibula.

• Ground foraging birds are likely to stir up the duff and kick LA particles into the air; the
worst case scenario is created by dust-bathing birds.

• The LA particles may not be easily aerosolized during foraging or dust bathing, but some
of the smallest particles may adhere to other inspirable "dust" that more readily becomes
suspended as a colloid in the air when the duff is disturbed.

2. Over a period of months or years some of the LA particles are likely to be inspired by ground
dwelling/foraging birds.

• Particles trapped in the protective mucus of the nasal passageways, pharynx and ciliated
conducting airways will have little biological impact on those structures, and will be
cleared rapidly by the mucociliary escalator. Mucus containing particles cleared from the
upper airways will be swallowed, enter the gastrointestinal tract, and excreted in the
feces. Evaluation of LA content within the core matrix of avian fecal pellets collected
within the zone of contamination may constitute the simplest way to directly quantify the
possibility that a threat exists.

• Particles deposited in the parabronchi will be phagocytized predominately by epithelial
cells that line the atria and infundibula, but also by resident macrophages in the lumen
and interstitial macrophages. Engulfed particulates composed of substances that cannot
be degraded or digested intracellularly by the epithelial cells and interstitial macrophages

-------
412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

Avian Respiration Synopsis as related to the Libby Amphibole; EPA 2011 RFW

appear to pose a specific problem for birds: the epithelial cells (and apparently the
interstitial macrophages) remain in situ, presumably emitting modulators (cytokines and
chemokines) that provoke ongoing focal inflammatory reactions. The result in some birds
appears to be granuloma and giant cell formation at sites where engulfed particulates
cannot be cleared from the secondary and tertiary bronchi.

• The pattern of response to embedded particulates does not include fibrosis in birds; mild
focal fibrosis would have little functional impact on the non-inflating avian lung, but
fibrosis might modestly increase respiratory effort if the air sacs are affected.

• Particles deposited in air sacs are likely to be engulfed by macrophages and cleared from
the air sacs. The fate of the responding macrophages, and thus sites to which they might
redistribute the LA particles, is not known.

3. There is no evidence that the lungs of wild avian species are anatomically, physiologically or
immunologically more susceptible to inhaled particulates than mammalian lungs.

• Published assertions that "avian" lungs are more susceptible to particulate or pathogen
damage than mammalian lungs consistently cite examples of the susceptibility of poultry
(particularly broiler chickens and modern hybrid turkeys) to respiratory pathogens or to
extremely challenging air quality when commercial growout facilities are poorly
managed. Indeed, chickens bred for extremely rapid growth and meat production (broiler
chickens) provide an excellent model of genetically-imposed cardio-pulmonary and
immunological inadequacies. Broiler chicks typically hatch at a weight of 40 g and grow
to 4 kg within 8 weeks. Thus in two months a broiler's body weight doubles and
redoubles almost 7 times. If human infants grew at the same rate, their body weight
would increase from 3 kg (6.6 lb) at birth to 310 kg (690 lb) by 2 months of age. The
consequences are obvious: extremely rapid early growth in broilers imposes proportional
challenges to their developmentally immature pulmonary, cardiovascular and
immunological systems. Rapid growth triggers a suite of "metabolic diseases"
attributable primarily to "outgrowing cardio-pulmonary capacities" or "impaired
immuno-competency". Wild birds and the progenitors of modern poultry breeds are
uniformly found to be considerably more robust than modern broiler chickens and hybrid
turkeys (Wideman, 2000, 2001; Nganpiep and Maina, 2002; Wideman et al. 2004, 2007).

• Particulate deposition due to gravitational sedimentation and Brownian diffusion most
likely will occur where air velocities are low, particle residence time is prolonged, and at
sites of air flow reversal. Accordingly, particles are highly likely to be deposited
throughout the alveoli of mammalian lungs, precisely at the level where gas exchange
must occur, and where membrane fibrosis is highly detrimental due to the loss of
elasticity (alveoli must inflate and deflate during the respiratory cycle; fibrosis
significantly increases respiratory effort in birds). In contrast, convective air flow does
not penetrate the gas exchange capillaries of avian lungs, thus particle deposition within
the air capillaries should be minimal or non-existent. Within the avian lung parenchyma,
air flow is bidirectional in neopulmonic parabronchi which comprise 25%, at most, of the
lung volume.

• Interstitial inflammation, granuloma development and giant cell formation are normal
patterns of avian responses to pulmonary entrapment of particulates delivered either via
the inspired air or via the bloodstream. Absent respiratory disease attributable to

-------
457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

479

480

481

482

483

484

485

486

Avian Respiration Synopsis as related to the Libby Amphibole; EPA 2011 RFW

pathogens, all available evidence indicates these intrapulmonary inflammatory responses
have minimal impact on the function or viability of affected birds.

• Assuming equal levels of "exposure", the above considerations indicate that otherwise
healthy mammals are likely to be more sensitive to particle inhalation than clinically
healthy birds.

4. Conclusion: The experiments conducted by Nakaue, Pierson and Heifer (1982) and Bland,
Nakue, Goeger and Heifer (1985) are highly instructive: 20 consecutive days of intensive aerosol
exposure to volcanic ash particles of a respirable size did elicit intrapulmonary histological
changes but failed to alter any routine indices of broiler performance, nor was mortality affected.
Broiler chickens are considerably less robust than wild birds (vide supra). Peacock and Peacock
(1965) demonstrated that most adult Leghorn chickens survived several years after milligram
quantities of asbestos fibers were instilled directly into their air sacs and (presumably) into the
lung parenchyma. It is my opinion that some birds in the affected area are likely to exhibit
histological evidence of intrapulmonary LA particulate exposure, but that little or no impact on
the physiological function or viability of resident avian populations will be discernable.

Robert F. Wideman, Jr., Ph.D.

Professor and Associate Director
Center of Excellence for Poultry Science
Division of Agriculture
University of Arkansas

-------
Avian Respiration Synopsis as related to the Libby Amphibole; EPA 2011 RFW

REFERENCES

Abe, T., K. Nakamura, H. Tojo, M. Mase, T. Shibahara.1998. Histology, Immunohistochemistry,
and ultrastructure of hydropericardium syndrome in adult broiler breeders and broiler
chickens. Avian disease 42(3):606-612.

Anderson, D. P., C. W. Beard, and R. P. Hanson., 1966. Influence of poultry house dust,

ammonia, and carbon dioxide on the resistance of chickens to Newcastle disease virus.
Avian Dis. 10:177-188.

Bakutis, B., E. Monstviliene, and G. Januskeviciene. 2004. Analyses of airborne contamination
with bacteria, endotoxins and dust in livestock barns and poultry houses. Acta Vet. Brno
73:283-289.

Bland, M. C., H. S. Nakaue, and M. P. Goeger. 1985. Duration of exposure-histological effects
on broiler lungs, performance, and house environment with Mt. St. Helen's volcanic ash
dust. Poult. Sci. 64:51-58.

Bottje, W.B., S. Wang, F.J. Kelly, C. Dunster, A. Williams, and I. Mudway. 1998. Antioxidant
defenses in lung lining fluid of broilers: Impact of poor ventilation conditions. Poult. Sci.
77:516-522.

Brackenbury, J. H., 1987. Ventilation of the lung-air sac system. Chapter 2, pp39-69 In:Bird
Respiration, Volume I. T. J. Seller, ed. CRC Press, Boca Raton, FL.

Brain, J. D., R. M. Molina, M. M. DeCamp, and A. E. Warner. 1999. Pulmonary intravascular

macrophages: their contribution to the mononuclear phagocyte system in 13 species. Am.
J. Physiol. 276:L146-L154.

Brambilla, C., J. Abraham, E. Brambilla, K. Benirschke, and C. Bloor. 1979. Comparative
pathology of silicate pneumoconiosis. Am. J. Pathol. 96:149-170.

Brown, R. E., J. D. Brain, and N. Wang. 1997. The avian respiratory system: a unique model for
studies of respiratory toxicosis and for monitoring air quality. Environ. Health Perspect.
105:188-200.

Duncker, H.-R, 1974. Structure of the avian respiratory tract. Respir. Physiol. 22:1-19.

Fedde, M. R, 1986. Respiration. Chapter 6, ppl91-220 In: Avian Physiology, 4th edition. P. D.
Sturkie, ed. Springer-Verlag, New York, NY.

Fedde, M. R. 1998. Relationship of structure and function of the avian respiratory system to
disease susceptibility. Poult. Sci. 77:1130-1138.

Ficken, M. D., J. F. Edwards, and J. C. Lay. 1986. Induction, collection and partial

characterization of induced respiratory macrophages of the turkey. Avian Dis. 30:766-
771.

Fulton, R.M., W.M. Reed, and D.B. DeNicola. 1990. Light microscopic and ultrastructural
characterization of cells recovered by respiratory-tract lavage of 2- and 6-week old
chickens. Avian Dis. 34:87-98.

Gross, W. B. 1961. The development of "air sac" disease. Avian Dis. 5:431-439.

Gross, W. B. 1990. Factors affecting the development of respiratory disease complex in
chickens. Avian Dis. 34:607-610.

-------
Avian Respiration Synopsis as related to the Libby Amphibole; EPA 2011 RFW

Hamal, K., R. F. Wideman, N. B. Anthony, and G. F. Erf. 2008. Expression of inducible nitric
oxide synthase in lungs of broiler chickens following intravenous cellulose microparticle
injection. Poult. Sci. 87:636-644.

Hamal, K., R. F. Wideman, N. B. Anthony, and G. F. Erf. 2010. Differential expression of
vasoactive mediators in the microparticle challenged lungs of chickens that differ in
susceptibility to pulmonary arterial hypertension. Am. J. Physiol. Regul. Comp. Physiol.
298:R235-R242.

Hayter, R. B., and E. L. Besch. 1974. Airborne-particle deposition in the respiratory tract of
chickens. Poult. Sci. 53:1507-1511.

Hower, E.W. 1985. Fatal systemic toxoplasmosis in a wild turkey. J. of wild diseases 21(4):446-
449.

Johnston, S.D., S. Orgeig, O. Lopatko, and C.B. Daniels. 2000. Development of the pulmonary
surfactant system in two oviparous vertebrates. Am. J. Physiol. Regulatory Integrative
Comp. Physiol. 278:R486-R493.

Jukes, M. G. M., 1971. Control of respiration. Chapter 6, ppl71-196 In: Physiology and
Biochemistry of the Domestic Fowl. Volume 1. D. J. Bell and B. M. Freeman, eds.
Academic Press, New York, NY.

King, A. S., 1993. Apparatus respiratorius (systema respiratorium). Chapter 8, pp257-299 In:

Handbook of Avian Anatomy: Nomina Anatomica Avium, 2nd edition. J. J. Baumel, A. S.
King, J. E. Breazile, H. E. Evans, and J. C. Vanden Berge, eds. Nuttal Ornithological
Club, c/o Museum of Compo Zool., Harvard Univ., Cambridge, MA.

King, A. S., and V. Molony, 1971. The anatomy of respiration, Chapter 5, pp93-169 In:

Physiology and Biochemistry of the Domestic Fowl. Volume 1. D. J. Bell and B. M.
Freeman, eds. Academic Press, New York, NY.

King, A. S., and J. McLelland, 1984. Respiratory system. Chapter 7, ppl 10-144 In: Birds, Their
Structure and Function, 2nd edition. Bailliere Tindall, London.

Klika, D.W., D.W. Scheuermann, M.H.A. De Groodt-Lasseel, I. Bazantova, and A. Switka.
1996. Pulmonary macrophages in birds (barn owl, Tyto tyto alba), domestic fowl (Gallus
gallus f. domestica), quail (coturnix coturnix), and pigeons (Columbia livia). The
Anatomical Record. 246:87-97.

Lai, H. T. L., M. G. B. Nieuwland, B. Kemp, A. J. A. Aarnink, and H. K. Parmentier. 2009.

Effects of dust and airborne dust components on antibody responses, body weight gain,
and heart morphology of broilers. Poult. Sci. 88:1838-1849.

Lorenzoni, A. G., and R. F. Wideman. 2008. Intratracheal administration of bacterial

lipopolysaccharide elicits pulmonary hypertension in broilers with primed airways. Poult.
Sci. 87:645-654.

Lorenzoni, A. G., G. F. Erf, N. C. Rath, and R. F. Wideman. 2009. Cellular component of lavage
fluid from broilers with normal versus aerosol-primed airways. Poult. Sci. 88:303-308.

Lund, C. C., L. A. Shaw, and C. K. Drinker. 1921. Quantitative distribution of particulate

material (manganese dioxide) administered intravenously to the dog, rabbit, guinea pig,
rat, chicken, and turtle. J. Exp. Med. 33: 231-238.

Lutz, P. L., I. S. Longmuir, and K. Schmidt-Nielsen, 1974. Oxygen affinity of bird blood. Respir.
Physiol. 20:325-330.

Maina, J.N., and H.M. Cowley. 1998. Ultrastructural characterization of the pulmonary cellular
defences in the lung of a bird, the rock dove, Columba livia. Proc. R. Soc. Lond. B.
265:1567-1572.

-------
Avian Respiration Synopsis as related to the Libby Amphibole; EPA 2011 RFW

McLelland, J., and V. Molony, 1983. Respiration. Chapter 4, pp63-89 In: Biochemistry and
Physiology of the Domestic Fowl, Vol 4. B. M. Freeman, ed. Academic Press, New
York, NY.

Mensah, G. A., and J. D. Brain. 1982. Deposition and clearance of inhaled aerosol in the
respiratory tract of chickens. J. Appl. Physiol. 53:1423-1428.

Nakamura, K., M. Yamada, S. Yamaguchi, M. Mase, Yamada. 2001. Proliferation of lung

macrophages in acute fatal viral infections in chickens. Avian diseases 45(4):813-818.

Nakaue, H. S., M. L. Pierson, and D. H. Heifer. 1982. Effect of Mount St. Helen's volcanic ash
on broiler performance and health and on house environment. Poult. Sci. 61:693-698.

Nickel, R, A. Schummer, E. Seiferle, W. G. Siller, and P. A. L. Wight, 1977. Respiratory system
(Systema respiratofium). pp62-69 In: Anatomy of the Domestic Birds. Springer-Verlag,
New York, NY.

Nganpiep, L.N., and J.N. Maina. 2002. Composite cellular defence stratagem in the avian
respiratory system: functional morphology of the free (surface) macrophages and
specialized pulmonary epithelia. J. Anat. 200:499-516.

Peacock, P. R., and A. Peacock. 1965. Asbestos-induced tumors in White Leghorn fowls. Ann.
N. Y. Acad. Sci. 132:501-503.

Powell, F. L., 1982. Diffusion in avian lungs. Federation Proc. 41:2131-2133.

Qureshi, M. A., J. N. Petitte, S. M. Laster, and R. R. Dietert. 1993. Avian macrophages:

contribution to cellular microenvironment and changes in effector function following
activation. Poult. Sci. 72:1280-1284.

Reese, S., G. Dalamani, and B. Kaspers. 2006. The avian lung-associated immune system: a
review. Vet. Res. 37:311-324.

Scheid, P., and J. Piiper, 1987. Gas exchange and transport. Chapter 4, pp97-129 In: Bird

Respiration, Volume I. T. J. Seller, ed. CRC Press, Boca Raton, FL. Schmidt-Nielsen, K.,
1975. Recent advances in avian respiration. Symp. Zool. Soc. Lond.35:33-47.

Staub, N. C. 1994. Pulmonary intravascular macrophages. Annu. Rev. Physiol. 56:47-67.

Stearns, R C., G. M. Barnas, M. Walski, and J. D. Brain, 1987. Deposition and phagocytosis of
inhaled particles in the gas exchange region of the duck, Anas platyrhynchos. Respir.
Physiol. 67:23-36.

Toth, T.E. and P.B. Siegel. 1986. Cellular defense of the avian respiratory tract: paucity of free-
residing macrophages in the normal chicken. Avian Dis. 30:67-75.

Toth, T.E., P. Siegel, and H. Veit. 1987. Celullar defense of the avian respiratory system. Influx
of phagocytes: Elicitation versus activation. Avian Diseases 31:861-867.

Toth, T.E., H. Robert, T. Caceci, P. Siegel, and O. Duane. 1988. Cellular defense of the avian
respiratory phagocytes activated by Pasteurella multocida. Inflam. Immunol. 56:1171-
1179.

Tottori, J., R. Yamaguchi, Y. Murakawa, M. Sato, K. Uchida, and S. Tateyama. 1997.

Experimental production of ascites in broiler chickens using infectious bronchitis virus
and Escherichia coli. Avian Dis. 41:214-220.

Wang, W., G. F. Erf, and R. F. Wideman. 2002. Effect of cage vs. floor litter environments on
the pulmonary hypertensive response to intravenous endotoxin and on blood-gas values
in broilers. Poult. Sci. 81:1728-1737.

-------
Avian Respiration Synopsis as related to the Libby Amphibole; EPA 2011 RFW

Wang, W., R. F. Wideman, T. K. Bersi, and G. F. Erf. 2003. Pulmonary and hematological

inflammatory responses to intravenous cellulose micro-particles in broilers. Poult. Sci.
82:771-780.

Warner, A. E., R. M. Mulina, C. F. Bellows, J. D. Brain, and S. D. Hyg. 1994. Endotoxemia

enhances pulmonary mononuclear cell uptake of circulating particles and pathogens in a
species without pulmonary intravascular macrophages. Chest 105 (3):50S-51S.

Weidner, W. J., and C. T. Lancaster. 1999. Effects of monastral blue on pulmonary arterial blood
pressure and lung and liver particle retention in chickens. Poult. Sci. 78:878-882.

Wells, R M. G., 1976. The oxygen affinity of chicken haemoglobin in whole blood and
erythrocyte suspensions. Resp. Physiol. 27:21-31.

Whyte, R. T. 1993. Aerial pollutants and the health of poultry farmers. World's Poult. Sci. J.
49:139-156.

Wideman, R. F. 2000. Cardio-pulmonary hemodynamics and ascites in broiler chickens. In:

Poult, and Av. Biol. Rev. Ed. R. R. Dietert and M. A. Ottinger. 11 (1):21 -43.

Wideman, R. F. 2001. Pathophysiology of heart/lung disorders: pulmonary hypertension
syndrome in broiler chickens. World's Poult. Sci. J. 57:289-307.

Wideman, R. F., G. F. Erf, M. E. Chapman, W. Wang, N. B. Anthony, and L. Xiaofong. 2002.
Intravenous micro-particle injections and pulmonary hypertension in broiler chickens:
Acute post-injection mortality and ascites susceptibility. Poult. Sci. 81:1203-1217.

Wideman, R. F., M. E. Chapman, W. Wang, and G. F. Erf. 2004. Immune modulation of the

pulmonary hypertensive response to bacterial lipopolysaccharide (endotoxin) in broilers.
Poult. Sci. 83:624-637.

Wideman, R. F., M. E. Chapman, K. R. Hamal, O. T. Bowen, A. G. Lorenzoni, G. F. Erf, and N.
B. Anthony. 2007. An inadequate pulmonary vascular capacity and susceptibility to
pulmonary arterial hypertension in broilers. Poult. Sci. 86:984-998.

Wideman, R. F., K. R. Hamal, M. T. Bayona, A. G. Lorenzoni, D. Cross, F. Khajali, D. D.

Rhoads, G. F. Erf., and N. B. Anthony. 201 la. Plexiform lesions in the lungs of domestic
fowl selected for susceptibility to pulmonary arterial hypertension: Incidence and
histology. Anatomical Record 294:739-755, 2011b.

Wideman, R. F., and K. R. Hamal. 2011. Idiopathic pulmonary arterial hypertension: an avian
model for plexogenic arteriopathy and serotonergic vasoconstriction. J. Pharmacol.
Toxicol. Methods 63:283-295.

Winkler, G. 1988. Pulmonary intravascular macrophages in domestic animal species: review of
structural and functional properties. Am. J. Anat. 181:217-234.

Yamaguchi, R., J. Tottori, K. Uchida, S. Tateyama, and S. Sugano. 2000. Importance of
Escherichia coli infection in ascites in broiler chickens shown by experimental
production. Avian Dis. 44:545-548.

Zeng, X., K.E. Yutzey and J. A. Whitsett. 1998. Thyroid transcription factor-1, hepatocyte

nuclear factor-3p and surfactant protein A and B in the developing chicken lung. J. Anat.
193:399-408.

Zucker, B. A., S. Trojan, and W. Muller. 2000. Airborne gram-negative bacterial flora in animal
houses. J. Vet. Med. B 47:37-46.

-------
Avian Respiration Synopsis as related to the Libby Amphiboie; EPA 2011 RFW

Paleopulmonlc Mediodorsal

parabronchi secondary

Figure 1. Schematic arrangement of avian lungs and air sacs. Deep within the thoracic cavity the
trachea bifurcates at the syrinx (the avian organ of phonation) to form the right and left
extrapulmonary primary bronchi. These bronchi penetrate the respective lungs to become the
intrapulmonary primary bronchi. Within the lungs of domestic fowl, the medioventral,
mediodorsal, lateroventral, and laterodorsal secondary bronchi branch from the
intrapulmonary primary bronchus. The bronchi and air sacs connect via ostea.

-------
Avian Respiration Synopsis as related to the Libby Amphiboie; EPA 2011 RFW

Torus marginalis

pars intrapulmonalis

Medial view of the right lung illustrating: the intrapulmonary
primary branch us; the medioventral (MV), mediodorsal (MD)
and lateroventral (LV) secondary bronchi, paleopulmonic
parabronchi (tertiary bronchi) connecting the MV and MD
secondary bronchi; and, ostea (openings)to air sacs. The
Costal sulcus represents a rib indentation.

Figure 2. Details of the primary and secondary bronchi within avian lungs. The intrapulmonary
primary bronchus penetrates from the cranial to the caudal margins of the lung, opening
caudally into the osteum of the abdominal air sac. Within the lungs the secondary bronchi
branch from the intrapulmonary primary bronchus.

-------
Avian Respiration Synopsis as related to the Libby Amphiboie; EPA 2011 RFW

Air sac system of the fowl (Weik, 1963).

A: trachea, B: lung; C: cervical vertebrae; D: thoracic vertebrae;

E: ribs; F: ilium; G: ischium, H: pubis; J: humerus; K: scapula;

L: coracoid; M: sternum; a: clavicular air sac; b: cervical air sacs;
c: cranial thoracic air sac; d: caudal thoracic air sac; e: abdominal
air sac

Figure 3. The non-inflating avian lungs (B) are partially encased by 5 ribs (E) as indicated by the
costal sulci (indentations) in the dorsal-lateral aspect of the lungs. The air sacs are shown in their
anatomically correct positions.

-------
Avian Respiration Synopsis as related to the Libby Amphiboie; EPA 2011 RFW

Abdominal
air sac

Paleopulmonic Mediodorsal

parabronchi secondary bronchi

Medioventral
secondary
bronchi

Cranial thoracic
air sac

Lateroventral
secondary bronchus A

Intrapulmonary primary
bronchus

Trachea

Caudal thoracic
air sac

Paleopulmonic Mediodorsal

parabronchi secondary bronchi

Cranial thoracic Neopulmonic

air sac parabronchi B

Figure 4. Scheme of the organization of the parabronchi in birds.

(A) Only paleopulmonic parabronchi are present in some birds (e.g., penguin and emu). (B) In
addition to paleopulmonic parabronchi, a variably developed net of neopulmonic parabronchi is
present in most birds (Duncker, 1972).

-------
Avian Respiration Synopsis as related to the Libby Amphiboie; EPA 2011 RFW

Infundibulum

Atrium

Musculus atrial js

Septum interatrialium

Fig. 10 Diagram of part of a Para-
bronchus (Gal/us), to show the archi-
tecture of the interior and the blood
vessels. For orientation see inset of
transverse section of a parabronchus.
See Annot. 60—63, 66.

capillar®

V.atrialis
Venule septalis
Venula intraparabronchialis
V. intra pa rabronchia lis

V.interparabronchiolis
Venula intraparabronchialis

A.interparabronchialis
Arteriola intraparabronchialis

TS. para bronchus

Area enlarged

Figure 5. Section through part of the wall of a parabronchus. Atria 100-200urn in diameter form
pockets projecting 50|im into the luminal wall. Spiral bands of smooth muscle (Musculus
atrialis) underlie the parabronchial luminal epithelium and encircle the opening to each atrium.
One or more funnel-shaped infundibula penetrate from the atrial floor into the parabronchial
wall, with multiple freely anastomosing air capillaries originating from each infundibulum and
radiating outward toward the periphery (outer boundary) of the parabronchus.

-------
Avian Respiration Synopsis as related to the Libby Amphiboie; EPA 2011 RFW

Figure 6. Section through two adjacent parabronchi. a: interatrial septa; b: atria; c: air capillaries;
d: outer connective tissue septa; e: blood vessels; f: anastomotic connections between air
capillaries. The air capillaries radiate outward toward the periphery (outer boundary) of the
parabronchi.

-------
Avian Respiration Synopsis as related to the Libby Amphiboie; EPA 2011 RFW

(from pulmonary artery)

Figure 7. Interparabronchial arteries supply deoxygenated blood to Intraparabronchial arterioles
branching inward into the parabronchial wall to form a net of blood capillaries surrounding each
air capillary. Gas exchange occurs at the blood-gas barrier at the interface between blood
capillaries and air capillaries. Venules collect the oxygenated blood at the base of the atria and
infundibula adjacent to the parabronchial lumen.

-------
Avian Respiration Synopsis as related to the Libby Amphiboie; EPA 2011 RFW

Inspiration

Expiration

Paleopulmonic

Figure 8. Schematic representation of the pathway of gas flow through the paleopulmonic and
neopulmonic tertiary parabronchi during inspiration (A, upper panel) and expiration (B, lower
panel). IPB: intrapulmonary primary bronchus; MD: mediodorsal secondary bronchi; MV:
medioventral secondary bronchi. Outward arrows on air sacs (upper panel) = inflation caused by
negative thoraco-abdominal pressures (suction); Inward arrows on air sacs (lower panel) =
deflation caused by positive thoraco-abdominal pressures. Arrows in primary, secondary and
tertiary parabronchi show directions of convective air flow.

-------
Avian Respiration Synopsis as related to the Libby Amphiboie; EPA 2011 RFW

Particle Size Distributions of LA Particles In Duff from Libby OU3 {N = 1,547)

Length

Length (um)

Width (um)

1°

Aspect Ratio

Error bounds are based on a 90% confidence interval.

Figure 9. Particle size distributions for Libby Amphiboie (LA) in duff.

-------
Avian Respiration Synopsis as related to the Libby Amphiboie; EPA 2011 RFW

FIGURE 2. Pathway of gas flow in the avian respiratory system
during inspiration. Enlargement of the body cavity by inspiratory
muscle action lowers pressure in the air sacs relative to that in the
atmosphere and gas flows into the system. Gas docs not enter the
medioventral secondary bronchi, but passes into the mediodorsal
secondary bronchi. Some of the gas passes through the paleopulmonic
parabronchi. and the remainder passes into the neopulmonic
parabronchi and caudal air sacs.

FIGURE 3. Pathway of gas flow in the avian respiratory system
during expiration. Reduction in volume of the body cavity by
expiratory muscle action increases pressure in the air sacs relative to
that in the atmosphere and gas flows out of the system. Compression
of intrapulmonary primary bronchus causes gas coming from the
caudal air sacs to pass through neopulmonic parabronchi, into
mediodorsal secondary bronchi and through the paleopulmonit:
parabronchi. Gas from the cranial air sacs docs not pass through
parabronchi on the way to the primary bronchus and trachea.

Figures from Fedde, 1998.

-------
FINAL

ATTACHMENT D
STUDY REPORTS

(See attached compact disk)

-------
Part 2 Baseline Ecological Risk Assessment for
Asbestos - Non-Operable Unit 3

-------
This page left blank intentionally

-------
Site-Wide Baseline Ecological Risk Assessment
For Exposure to Asbestos
Part 2 (Non-OU3)

Libby Asbestos Superfund Site
Libby, Montana

December 2014

Prepared for:

U.S. Environmental Protection Agency
Region 8
1595 Wynkoop Street
Denver, Colorado 80202

Prepared by:

VSh.

CDM Federal Programs Corporation
555 17th Street, Suite 1200
Denver, Colorado 80202

Under a contract with:

U.S. Army Corps of Engineers

Omaha District
Offutt AFB, Nebraska 68113

-------
This page left blank intentionally

-------
Libby Asbestos Superfund Site,

Libby, Montana
Site-Wide Baseline Ecological Risk Assessment
Part 2 (Non-OUS)

Reviewed by: _ Date:

Dan W.
-------
This page left blank intentionally

-------
Table of Contents

Section 1 Introduction........................................................................................................................................... i-i

1.1 Purpose of this Document 1-1

1.2 Document Organization 1-1

Sectli e Characterization.......................................................................................................................... 2-1

2.1 Overview 2-1

2.2 Operable Units 2-1

2.3 Physical Setting 2-2

2.3.1 Climate 2-3

2.3.2 Surface Water Features 2-3

2.4 Ecological Setting 2-4

2.4.1 Aquatic Setting 2-4

2.4.2 Terrestrial Setting 2-4

2.4.3 Federal and State Species of Special Concern 2-5

2.5 Nature and Extent of LA Contamination at the Site 2-5

2.5.1 Mineral Characteristics of LA 2-5

2.5.2 Concentrations of LA in Environmental Media 2-6

2.5.2.1 Water Source Study 2-6

2.5.2.2 Nature and Extent of LA in Surface Water and Sediment 2-6

2.5.2.3 Porewater in the Tributaries 2-7

Section 3 Problem Formulation 3-1

3.1 Conceptual Site Model 3-1

3.1.1 Potential Sources of Contamination 3-1

3.1.2 Migration Pathways in the Environment 3-2

3.1.3 Potentially Exposed Ecological Receptors 3-2

3.1.4 Exposure Pathways of Chief Concern 3-2

3.1.4.1 Fish 3-2

3.1.4.2 Benthic Invertebrates 3-3

3.1.4.3 Amphibians 3-3

3.2 Management Goal and Assessment Techniques 3-3

3.2.1 Management Goal 3-3

3.2.2 Definition of Population 3-3

3.2.3 Assessment Endpoints 3-3

3.2.4 Measures of Effect 3-3

Section 4 Risk Characterization for Fish 4-1

4.1 Reported Effects 4-1

4.2 0U3 Results 4-1

4.3 Risk Characterization 4-2

4.3.1 In Situ Eyed Egg and Alevin Exposure Study 4-2

4.3.2 In Situ Juvenile Fish Study 4-2

4.3.3 In Situ Lesion Study 4-2

4.4 Fish Summary 4-3

Section 5 Risk Characterization for Benthic Macroinvertebrates 5-1

SmiUl '

Non-OU3 Asbestos BERA_FINAL.docx

-------
Table of Contents • Baseline Ecological Risk Assessment Part 2 (Non-OU3), Libby, Montana

5.1 Reported Effects 5-1

5.2 OU3 Results 5-1

5.3 Risk Characterization 5-2

5.4 Benthic Macroinvertebrate Summary 5-2

Section 6 Risk Characterization for Amphibians 6-1

6.1 Reported Effects 6-1

6.2 0U3 Results 6-1

6.3 Risk Characterization 6-2

6.3.1 Laboratory Toxicity Tests 6-2

6.3.2 In Situ Lesion Studies 6-2

6.4 Amphibian Summary 6-2

Seetli certainty Assessment 7-1

7.1 Uncertainties in Nature and Extent of Contamination 7-1

7.1.1 Representativeness of Samples Collected 7-1

7.1.2 Accuracy of Analytical Measurements 7-1

7.1.2.1 TEM 7-2

7.1.2.2 PLM 7-2

7.2 Uncertainties in Exposure Assessment 7-3

7.3 Uncertainties in Toxicity Assessment 7-3

7.3.1 Absence of Toxicity Data for LA 7-3

7.3.2 Extrapolation of 0U3 Study Results to the Site 7-3

7.3.3 Uncertainties in 0U3 Studies 7-4

7.4 Uncertainties in Risk Characterization 7-4

7.5 Summary of Uncertainties 7-4

Section 8 Summary anil Conclusions 8-1

Section 9

"smith

Non-OU3 Asbestos BERA_FINAL.docx

-------
Table of Contents • Baseline Ecological Risk Assessment Part 2 (Non-OU3), Libby, Montana

List of Tables

Table 2-1 State Species of Concern

Table 2-2 Description of Surface Water, Sediment, and Porewater Sampling Locations

Table 2-3 Summary Statistics for LA in Surface Water

Table 2-4 Summary Statistics for LA in Sediment

Table 2-5 Summary Statistics for LA in Porewater

List of Figures

Figure 2-1 Site Location Map

Figure 2-2 Mining-Related Site Features

Figure 2-3 Operable Unit Boundaries

Figure 2-4 Wind Rose for LBBM8 (2007-2013]

Figure 2-5 Photographs of Vermiculite and Asbestos

Figure 2-6 Surface Water Sampling Locations in 0U4

Figure 2-7 Sediment Sampling Locations in 0U4

Figure 2-8 Porewater Sampling Locations in 0U4

Figure 2-9 Sampling Locations in OU7

Figure 3-1 Conceptual Site Model for Ecological Receptors to Asbestos

List of Appendices

Appendix A Responses to Comments

Appendix B Detailed Sample and Analytical Information

Appendix C OU5 Confirmation Soil Samples from the Fishing Pond

Smiifl

Non-OU3 Asbestos BERA FINAL.docx

-------
Table of Contents • Baseline Ecological Risk Assessment Part 2 (Non-OU3), Libby, Montana

Acronyms

% percent

< less than

> greater than or equal to

°F degrees Fahrenheit

BERA baseline ecological risk assessment

BNSF Burlington Northern and Santa Fe

CB&I CB&I Federal Services, LLC

CSM conceptual site model

EDS energy dispersive spectra

EPA U.S. Environmental Protection Agency

j Ext extension

FWS Fish and Wildlife Service

Grace W.R. Grace and Company

Hwy highway

LA Libby amphibole asbestos

LRC lower Rainy Creek

MDEQ Montana Department of Environmental Quality

MFL million fibers per liter

ND non-detect

NIST National Institute of Standards and Technology

NVLAP National Voluntary Laboratory Accreditation Program

OU operable unit

PLM polarized light microscopy

PLM-Grav polarized light microscopy gravimetric

PLM-VE PLM visual area estimation

QA quality assurance

QAPP quality assurance project plan

QC quality control

RAWP response action work plan

ROD record of decision

SAP sampling and analysis plan

SOP standard operating procedure

TEM transmission electron microscopy

Tr trace

USACE U.S. Army Corps of Engineers

-------
Section 1

Introduction

1.1 Purpose of this Document

The purpose of Part 2 is to present a baseline ecological risk assessment (BERA] for all operable units
(OUs], with the exception of 0U3 (see Part 1], of the Libby Asbestos Superfund Site. This non-0U3
BERA will describe the likelihood, nature, and extent of adverse effects in ecological receptors exposed
to Libby amphibole asbestos (LA] at the Site outside of 0U3 as a result of releases of asbestos to the
environment from past mining, milling, and processing activities. This information, along with other
relevant information, is used by risk managers to decide whether remedial actions are needed to
protect ecological receptors at the Site from the effects of exposure to mining-related environmental
asbestos contamination. This document has undergone review by U.S. Fish and Wildlife Service (FWS]
and Montana Department of Environmental Quality (MDEQ], Appendix A contains a summary of the
comments received and the responses prepared by U.S. Environmental Protection Agency (EPA],

1.2 Document Organization

¦ Section 1 - Introduction. This section provides the purpose and organization of this document.

¦ Section 2 - Site Characterization. This section describes the location, history, and
environmental setting of all OUs, except 0U3, including information on the nature and extent of
asbestos contamination in the environment.

¦ Section 3 - Problem Formulation. This section describes the ecological problem formulation,
including the site conceptual model for exposure to asbestos (potential receptors will be
identified for each OU], the selection of assessment endpoints, and a description of the
measures of effect used to characterize the effects of asbestos exposure.

¦ Section 4 - Risk Characterization for Fish. This section presents the risk characterization for
fish.

¦ Section 5 - Risk Characterization for Benthic Macroinvertebrates. This section presents the
risk characterization for benthic macroinvertebrates.

¦ Section 6 - Risk Characterization for Amphibians. This section presents the risk
characterization for amphibians.

¦ Section 7 - Uncertainty Assessment. This section presents the uncertainty assessment, and
discusses the sources of uncertainty in the risk evaluation for ecological receptors.

¦ Section 8 - Summary and Conclusions. This section presents overall conclusions for the non-
0U3 BERA.

¦ Section 9 - References. Lists all the references used in the preparation of this report.
All referenced tables, figures, and appendices are provided at the end of this document.

ISk

smitn

Non-OU3 Asbestos BERA FINAL.docx

1-1

-------
Section 1 • Introduction

This page left blank intentionally

1-2

Cs»mnh

Non-OU3 Asbestos BERA_FINAL.docx

-------
Section 2

Site Characterization

2.1 Overview

Libby is a community in northwestern Montana that is located near a former vermiculite mine
(Figure 2-1). The vermiculite mine near Libby began limited operations in the 1920s and was
operated on a larger scale by the W.R. Grace and Company (Grace) from approximately 1963 to 1990.
Operations at the mine included mining and milling of the vermiculite ore. After milling, concentrated
ore was transported down Rainy Creek Road by truck to a screening facility (known today as the
former Screening Plant) adjacent to Montana Highway 37, near the confluence of Rainy Creek and the
Kootenai River (Figure 2-2). Here, the ore was size-sorted, and transported by rail or truck to
processing facilities in Libby and nationwide. At the processing plants, the ore was exfoliated by rapid
heating and exported to market by rail or truck.

Historic maps show the location of a processing plant at the edge of the former Stimson Lumber Mill,
near present day Libby City Hall. This older processing plant was taken off-line and demolished
sometime in the early 1950s. Another processing plant (known today as the former Export Plant) was
located near downtown Libby, near the intersection of the Kootenai River and Montana Highway 37
(Figure 2-2). Expansion operations at the Export Plant ceased sometime prior to 1981, although site
buildings were still used to bag and export milled ore until 1990.

During mine operations, invoices indicate shipment of nearly 10 billion pounds of vermiculite from
Libby to processing centers and other locations. Most of this was shipped and used within the United
States and was often sold under the brand name Zonolite. Vermiculite material was used in a variety of
commercial products that were marketed and sold to the general public. Before the mine closed in
1990, Libby produced approximately 80 percent (%) of the world's supply of vermiculite.

2.2 Operable Units

To facilitate a multi-phase approach to remediation of the Libby Site, eight separate OUs have been
established. Official OU boundaries will not be determined until the record of decision (ROD) is
published for each OU. OU1 and OU2 boundaries have been established. All remaining OUs have "study
boundaries" which will be finalized once their ROD is published. All OUs are shown on Figure 2-3 and
include:

¦ OU1. OU1 is defined geographically by the parcel of land that included the former Export Plant
and the Highway 37 embankments, and is situated on the south side of the Kootenai River, just
north of the downtown area of the City of Libby. The property is bound by the Kootenai River on
the north, the Burlington Northern and Santa Fe (BNSF) railroad thoroughfare on the south, and
residential properties on the east and west.

¦ OU2. OU2 includes areas impacted by contamination released from the former Screening Plant.
These areas include the former Screening Plant, the Flyway property, the Highway 37 right-of-
way adjacent to the former Screening Plant and/or Rainy Creek Road, and privately owned
properties.

ISk

smitn

Non-OU3 Asbestos BERA FINAL.docx

2-1

-------
Section 2 • Site Characterization

¦ OU3. The mine OU includes the property in and around the Zonolite Mine owned by Grace or
Grace-owned subsidiaries (excluding 0U2] and any area (including any structure, soil, air,
water, sediment or receptor] impacted by the release and subsequent migration of hazardous
substances and/or pollutants or contaminants from such property, including, but not limited
to, the mine property, the Kootenai River and sediments therein, Rainy Creek, Rainy Creek
Road and areas in which tree bark is contaminated with such hazardous substances and/or
pollutants and contaminants.

¦ OU4. 0U4 is defined as residential, commercial, industrial (not associated with former Grace
operations], and public properties, including schools and parks in and around the City of Libby,
or those that have received material from the mine not associated with Grace operations (e.g.,
properties that have utilized vermiculite from the mine in a garden or flowerbed],

¦ OU5. 0U5 is defined geographically by the parcel of land that included the former Stimson
Lumber Company. 0U5 is bound by the high bank of Libby Creek to the east, the BNSF railroad
to the north, and residential/commercial/industrial property within 0U4 to the south and west.
This OU is approximately 400 acres in size and is currently occupied by various vacant
buildings as well as multiple operating businesses (lumber processing, log storage, excavation
contractor, etc.]. Within the boundary of OU5 exists the Libby Groundwater Superfund Site,
which is not associated with the Libby Site.

¦ OU6. Owned and operated by the BNSF railroad, OU6 is defined geographically by the BNSF
property boundaries from the eastern boundary of 0U4 to the western boundaiy of OU7 and
extent of contamination associated with the Libby and Troy rail yards.

¦ OU7. The Troy OU includes all residential, commercial, and public properties in and around the
town of Troy, Montana, approximately 2 0 miles west of downtown Libby.

¦ OU8. United States and Montana State Highway rights-of-way and secondary state route rights-
of-way within the boundaries of 0U4 and OU7.

This risk assessment will focus on all OUs (with the exception of OU3], hereafter referred to as the
Site. The sections below describe in detail the physical setting of the Site.

2.3 Physical Setting

Libby is situated along the Kootenai River, at the confluence of several smaller creeks, in a relatively
narrow river valley. Mountains and national forest land surround the Kootenai Valley on all sides: the
Cabinet Mountains to the south, the Purcell Mountains to the north, and the Salish Mountains to the
east. The elevation of Libby is approximately 2,000 feet above sea level. The area is primarily
coniferous forest and heavily vegetated. The biome classification for the Kootenai Valley is the taiga,
which is also known as the northern coniferous forest or boreal forest biome.

Troy (OU7] is located within the Kootenai River valley northwest of Libby at an elevation ranging
from 1,850 feet above mean sea level along the Kootenai River to 2,500 feet above mean sea level on
the mountain slopes surrounding the valley. OU7 is approximately 8 miles long and 1.8 miles wide at
its broadest point. The topography of OU7 varies from gently graded, open land along the Kootenai
River to terraced hillsides and steep forested mountains adjacent to the river valley.

2-2

Non-OU3 Asbestos BERA_FINAL.docx

smith

-------
Section 2 • Site Characterization

imate

Climate at the Site is relatively moist, with annual precipitation in the Kootenai Valley averaging slightly
over 20 inches (this includes approximately 60 inches of snowfall]. Surrounding higher elevations
receive significantly more precipitation. During the winter months, moist Pacific air masses generally
dominate, serving to moderate temperatures and bring abundant humidity, rain, and snow. Colder,
continental air masses occasionally drop temperatures significantly, but generally only for shorter
periods. The average temperature in December and January are 25 to 30 degrees Fahrenheit (°F],

During summer, the climate is warmer and dryer, with only occasional rain showers and significantly
lower humidity and soil moistures. High temperatures of greater than 90°F are common. The average
temperature in July is approximately 65 to 70°F. Spring and fall are transition periods.

Due to its valley location along the Kootenai River and downstream of the Libby dam, fog is common in
the Kootenai Valley. This effect is most pronounced during winter and in the mornings. Inversions,
which trap stagnant air in the valley, are also common. Winds in the Kootenai Valley are generally light,
averaging approximately six to seven miles per hour. Prevailing winds are from the southwest (Figure
2-4], but daily wind direction is significantly affected by temperature differences brought about by the
large amount of vertical relief surrounding the area.

2.3.2 Surface Water Features

The Site is contained within the Kootenai drainage basin and the Kootenai River and Fisher River sub-
basins. The Kootenai drainage basin is contained in both Canada and the United States encompassing
about 18,000 square miles or 11,520,000 acres.

The Kootenai River, which transects 0U4 (Figure 2-1], has its origins in British Columbia's Kootenay
National Park in Canada. From there, it flows 485 miles into northwest Montana and through the
towns of Libby and Troy. The river continues into northern Idaho, then back into Canada and
Kootenay Lake. Ultimately, it joins with the Columbia River. Seventeen miles north of Libby, the river
is held back by the Libby Dam, creating a 90-mile long reservoir called Lake Koocanusa that reaches
into Canada (LibbyMT.com 2013], Atthis time, the Kootenai River is part of OU3 and was included in
the risk evaluation for OU3 (Part 1],

Kootenai River tributaries in 0U4 and OU7 are characteristically high-gradient mountain streams with
bed material consisting of various mixtures of sand, gravel, rubble, boulders, and drifting amounts of
clay and silt, predominantly of glacio-lacustrine origin. Fine materials, due to their instability during
periods of high stream discharge, are continually abraded and redeposited as gravel bars, forming
braided channels with alternating riffles and pools. Stream flow in unregulated tributaries generally
peaks in May and June, after the onset of snow melt, then declines to low flows from November
through March. Flows also peak with rain-on-snow events. As previously stated, the Site has a
relatively moist climate with annual valley precipitation slightly over 20 inches. Higher elevations
receive significantly more precipitation and account for much of the creek flow. Seasonal fluctuations
cause varying levels of runoff and creek flow. Typically, runoff is most significant in spring when snow
at higher elevations begins to melt. Summer precipitation does occur; however, typical summer
weather is hot and dry and creek flow is moderated by high elevation lakes.

In OU5, a fishing pond is currently under development. The hole for the pond has been excavated, but
it has not yet been filled with water. It is planned that Libby Creek will be used to provide a water
source for the pond upon its completion.

-------
Section 2 • Site Characterization

2.4 Ecological Setting

quatic Setting

Within the Site, there are multiple streams and a fishing pond in 0U5 that provide habitat for a range of
aquatic species, including fish, benthic macroinvertebrates, and amphibians. Site-specific population
surveys have not been performed outside of 0U3. However, information gathered for the Rainy Creek
watershed as part of 0U3 studies indicate that the most common species of fish are western cutthroat
trout, rainbow trout, and "cutbow" trout (a rainbow/cutthroat hybrid]. Aquatic invertebrate
community surveys in 0U3 indicate that the most common types of aquatic invertebrates observed
include mayflies, stoneflies, caddisflies, true flies, and beetle larvae. The most common amphibian
species observed are the tree frog, spotted frog, and western toad. Additional details regarding the
population surveys for 0U3 can be found in Section 4.3 (fish], Section 5.3 (benthic macroinvertebrates],
and Section 6.3 (amphibians] of Part 1. Due to the proximity of 0U3 to the Site and similarities in
terrain and habitat, it can reasonably be assumed that similar groups of organisms are present at the
Site. It is recognized that all creeks for which environmental data are available may not have all groups
of organisms present due to variations in environmental conditions (e.g., a creek may have seasonal
fluctuation in flow such that the habitat is not supportive of fish populations]. However, for the
purposes of this risk assessment, it has been assumed that all receptor types may be present, and all
creeks are evaluated as such.

2.4.2 Terrestrial Setting

Although there is forested land that surrounds the Site, it is currently being evaluated as part of 0U3
in Part 1 (as the extent of LA contamination in the forest has not yet been defined]. The remaining
land at the Site has largely been developed for human use, both residential and commercial use, and
habitat is not optimal to support terrestrial receptors. A brief discussion of terrestrial habitat
availability is presented below for each OU:

¦ OU1. Numerous investigations and removal events have occurred at 0U1 to address
contamination at the former Export Plant. 0U1 is now a landscaped park with paved access
and parking. The main features of the park include two boat ramps, a pavilion with surrounding
lawn areas, and picnic tables. Because the majority of 0U1 is landscaped park and frequented by
recreational visitors, it is not expected to provide significant habitat for terrestrial ecological
receptors.

¦ OU2. Similar to 0U1, numerous investigations and removal events have occurred at 0U2 to
address contamination at the former Screening Plant. The former Screening Plant (Subarea 1] is
currently privately owned and is being used for residential purposes. The Flyway property and
Wise property (Subareas 2 and 3] are currently vacant, undeveloped areas of land and the road
right-of-way adjacent to the former Screening Plant (Subarea 4] runs along Highway 37. None
of these subareas are expected to provide significant habitat for terrestrial ecological receptors.

¦ OU4, OU7. 0U4 and 0U7 include residential, commercial, industrial, and public properties in
and around the City of Libby and Troy. Because this land has been developed for human use, it
is not considered to provide significant habitat for terrestrial ecological receptors.

¦ OU5. 0U5 is predominantly an industrial area, occupied by various vacant buildings, as well as
multiple operating businesses (lumber processing, log storage, excavation contractor, etc.].

There is a small, isolated forested area within 0U5; however, due to the fragmented nature of

2-4

Non-OU3 Asbestos BERA_FINAL.docx

-------
Section 2 • Site Characterization

the habitat and proximity to human activity and disturbance, terrestrial receptor use is not
anticipated to be significant.

¦ OU6, OU8. The rail line and road rights-of-way do not serve as viable habitat for terrestrial
receptors due to their limited area, frequent disturbance and proximity to transportation
corridors.

As noted above, extensive soil removal actions have been performed in 0U1 and 0U2 to address LA
contamination at these former mine facilities. In addition, soil removal actions have also been taken at
properties in 0U4, 0U5, 0U6, and 0U7 to protect human health. The action levels used as the basis of
these soil cleanup efforts would also be protective of ecological receptors based on the results of the
0U3 BERA.

2.4.3 Federal and State Species of Specie cern

There is only one federally-listed protected species that has been reported to occur in or about the
vicinity of the Site, the bull trout (Salvelinus confluentus]. Critical habitat for bull trout has also been
designated, and the following streams in the area as follows, Fisher River, Libby Creek, O'Brien Creek,
Pipe Creek, Quartz Creek, and Callahan Creek. Species of concern to the State of Montana that have
been observed to occur in the general vicinity of the Site are listed in Table 2-1. This includes two
amphibians, three fish, and seven invertebrates. However, not all of these species are equally likely to
occur within the Site. Based on an evaluation of where the species was reported, the following listed
species are considered to be the most likely to occur at the Site:

Federal

¦ Bull Trout [Salvelinus confluentus]

State of Montana

¦ Coeur d'Alene Salamander [Plethodon idahoensis)

¦ Boreal Toad, Green (also known as Western Toad] (Bufo boreas]

¦ Bull Trout [Salvelinus confluentus]

¦ Torrent Sculpin [Cottus rhotheus)

¦ Westernslope Cutthroat Trout [Oncorhynchus clarkii lewisi)

2.5 Nature and Extent of LA Contamination at the Site

laracteristics

The vermiculite deposit near Libby contains a distinct form of naturally-occurring amphibole asbestos
that is comprised of a range of mineral types and morphologies (see Figure 2-5], In the spring of 2000,
the U.S. Geological Survey performed electron probe micro-analysis and x-ray diffraction analysis of
30 samples collected from asbestos veins at the mine (Meeker etal. 2003], The results indicated that
there were several mineral varieties of amphibole asbestos present, including (in order of decreasing
abundance] winchite, richerite, and tremolite, with lower levels of magnesio-riebeckite, edenite, and
magnesio-arfvedsonite. Although Meeker etal. (2003] did not report the presence of actinolite, the
authors note that, depending on the valence state of iron and data reduction methods utilized by other
analytical laboratories, some minerals may also be classified as actinolite. The mixture of asbestos
present at the Site is referred to as Libby amphibole asbestos (LA],

-------
Section 2 • Site Characterization

2.5.2 Concentrator i Environmental Media

Multiple studies have been carried out at the Site resulting in the collection of environmental samples
for a variety of media. All of the sampling and analytical methods have been planned in sampling and
analysis plans (SAPs] and associated quality assurance project plans (QAPPs] and conducted in
accordance with standard operating procedures (SOPs] for sampling and analysis. Consequently, all
data collected under these EPA-approved SAP/QAPPs/SOPs are considered to be appropriate for use
in the risk assessment. The studies for which data have been included in this risk assessment are
detailed in the sections below.

The major tributaries to the Kootenai River in 0U4 and 0U7 for which data are available include the
following: Cedar Creek, Cherry Creek, Fisher River, Flower Creek, Granite Creek, Libby Creek,
Parmenter Creek, O'Brien Creek, Pipe Creek, Quartz Creek, and Callahan Creek. Figure 2-6 through
Figure 2-8 present the locations where samples were collected for surface water1, sediment, and
pore water, respectively, for all studies near Libby. Figure 2-9 presents the locations where surface
water, sediment, and porewater samples were collected near Troy. Table 2-2 provides descriptions of
the sampling locations for surface water, sediment, and porewater. Table 2-3 through Table 2-5
provide summary statistics of surface water, sediment, and porewater, respectively, for each creek.
Appendix B contains detailed sample and analytical information for the samples included in this risk
assessment. This information was queried from the Libby and Troy Scribe databases on October 30,
2014.

2.5.2.1 Water Source Study

The water source study was completed in two phases in accordance with the Water Source
Identification Study - Phase I SAP (EPA 2011] and the Water Source Identification Study - Phase II
SAP/QAPP (EPA 2012a], The goal of this study was to identify a new source of water for use during
construction activities; however, these data have utility for use in this risk assessment and have been
included. Because other surface water studies have shown that LA concentrations are dependent upon
flow conditions, this study was separated into two sampling efforts to ensure collected surface water
data are representative of both high flow (spring] and low flow (fall] conditions. The water source
study measured asbestos concentrations at each location under both low flow and high flow
conditions. Surface water samples were collected at locations that are relevant to this risk assessment
from Libby Creek, Pipe Creek, Cedar Creek, Cherry Creek, Granite Creek, Flower Creek, Parmenter
Creek, and Quartz Creek. The results for this study are summarized in Data Summary Report: Water
Source Identification Study (EPA 2013a],

2.5.2.2 Nature and Extent of LA in Surface Water and Sediment

The nature and extent of LA in surface water and sediments study was completed in accordance with
the Nature and Extent of LA Contamination in Surface Water and Sediment SAP/QAPP (EPA 2012b],
During this study, surface water and sediment samples were collected from the major tributaries to
the Kootenai River in order to investigate the nature and extent of LA in surface water and sediment.
Sampling locations were selected for major tributaries to the Kootenai River by preferentially
choosing tributaries that have had a past removal action. The tributaries that were selected for
sampling include Granite Creek, Libby Creek, Callahan Creek, Flower Creek, Pipe Creek, and the Fisher
River. Up to three sampling locations along these tributaries were selected for surface water and
sediment sampling so that influences of removal actions and human interaction could be

1 Surface water samples from Libby Creek will be used as a surrogate for surface water in the fishing pond in OU5, because
this is the water source that will be used to fill the pond upon its completion.

2-6 smith

Non-OU3 Asbestos BERA_FINAL.docx

-------
Section 2 • Site Characterization

characterized. This study was separated into two sampling efforts to ensure collected surface water
data are representative of both high flow (spring] and low flow (fall] conditions. Sediment samples
were collected during low flow conditions. The results for this study are summarized in Data Summary
Report Nature and Extent of LA Contamination in Surface Water and Sediment (EPA 2013b],

2.5.2.3 Porewater in the Tributaries

The purpose of this study was to collect sediment porewater samples from locations in tributaries to
the Kootenai River in support of this ecological risk assessment. Sampling locations for sediment
porewater collection were the same tributary locations as sampled during the 2012 Nature and Extent
Study in Surface Water and Sediment (EPA 2012b] as presented in Section 2.3.2.2, with the addition of
three locations (two in O'Brien Creek and one in the Fisher River] as outlined in Sediment Porewater
Study of Kootenai River Tributaries (EPA 2013c], Sampling locations were added in O'Brien Creek
because this tributary is critical habitat for bull trout. Although the Fisher River (also critical habitat
for bull trout] was sampled in the 2012 Nature and Extent Study, an additional sampling location was
added further upstream for this study because the headwaters were thought to be more
representative of potential trout spawning habitat. Because these three new sampling locations were
not sampled for surface water and sediment as part of the 2012 Nature and Extent Study, samples for
these media were collected during this study.

-------
Section 2 • Site Characterization

This page left blank intentionally

2-8

Cs»mnh

Non-OU3 Asbestos BERA_FINAL.docx

-------
Section 3

Problem Formulation

Problem formulation is a systematic planning step that identifies the major concerns and issues to be
considered in an ecological risk assessment, and describes the basic approaches that will be used to
characterize ecological risks that may exist (EPA 1997], As discussed in EPA (1997], problem
formulation is generally an iterative process, undergoing refinement as new information and findings
become available.

A conceptual site model (CSM] is a schematic summary of what is known about the nature of source
| materials at a site, the pathways by which contaminants may migrate through the environment, and
| the scenarios by which ecological receptors may be exposed to site-related contaminants. When

information is sufficient, the CSM may also indicate which of the exposure scenarios for each receptor
are likely to be the most significant, and which (if any] are likely to be sufficiently minor that detailed
evaluation is not needed.

Figure 3-1 presents the CSM for exposure scenarios of potential concern to each main ecological
receptor group, including fish, benthic macroinvertebrates, amphibians, birds, mammals, terrestrial
plants, and soil invertebrates. The following sections provide a more detailed discussion of the main
elements of the CSM.

ntial Sources of Contamination

Historic mining and milling activities at the mine site resulted in releases of asbestos fibers to air.
Fibers released to air were carried downwind in air (mainly to the northeast as demonstrated in
Figure 2-4], followed by deposition of the fibers to soil and water.

Because creeks in the Libby and Troy areas are perennial streams and experience significant flow
fluctuations during the spring and following heavy precipitation events, many creeks have had riprap
placed at various sections by the U.S. Army Corps of Engineers (USACE], Lincoln County, the City of
Libby, and/or private land owners to control erosion (CDM Smith 2008a], Material used for the
construction of riprap sections in the creeks included: 1] quarried argillite and siltstone
(metasediments] from the Wallace Formation of the Precambrian Belt Group, 2] quarried syenite from
the Rainy Creek ultramafic complex, 3] basalt, and 4] concrete debris, tree stumps, and wood lagging.
The syenite is exposed at the Vermiculite Mountain Mine, and riprap constructed with this material is
thought to have originated at the mine. LA material in the form of biotite pyroxenite, magnetite
pyroxenite, and LA are often found in the presence of the syenite (CDM Smith 2008a],

In 2007 and 2008, several creeks in the Libby and Troy areas were investigated to evaluate the
presence and extent of LA in materials used for the construction of riprap. The results of these studies
can be found in Flower Creek Investigation Summary (CDM Smith 2007], Granite-Callahan Investigation
Summary Memo (CDM Smith 2008b] and Summary of Creek Investigations Completed for Libby Asbestos
SuperfundSite Operable Units 4 and 7 (CDM Smith 2008a], These creek studies determined that
several of the creeks were lined with riprap materials that contained LA. As a result, removal of LA-
contaminated riprap material from along the creek embankments was performed as directed in the

3.1 Conceptual Site Model

3-1

Non-OU3 Asbestos BERA_FINAL.docx

-------
Section 3 • Problem Formulation

addendum to the Response Action Work Plan (RAWP) Addendum for Flower Creek (CDM Smith 2008c],
Granite Creek (CDM Smith 2008d], Callahan Creek (CDM Smith 2008e], Libby Creek (CDM Smith
2009a], and Pipe Creek (CDM Smith 2009b],

Over 7,000 cubic yards of material was removed from the creeks. Soil clearance samples were collected
in accordance with the Response Action Sampling and Analysis Plan (CDM Smith 2008f] and analyzed by
polarized light microscopy (PLM] in accordance with NIOSH 9002. Tremolite-actinolite results ranged
from non-detect to less than (<] 1%. Following the clearance of the excavated areas, the creek
embankments were restored in accordance with the respective RAWP addendum. In Granite Creek,
shotcrete was applied to areas where LA-contaminated riprap was left in place to minimize release of
LA fibers.

1 Migratit 'ays in the Environment

Asbestos that is present in soil may be carried in surface water runoff (e.g., from rain or snowmelt]
into local creeks resulting in contamination of waters and sediments in the creeks. Because the riprap
material previously placed into the creeks has either been removed or coated with shotcrete, erosion
of overbank material along the creeks is the main source of on-going release of asbestos to the
environment.

I Potentially Exposed Ecological Receptors

As discussed in Section 2.4.1, there are several ecological receptors that are likely to occur at the Site
that could be exposed to LA. However, it is generally not feasible or necessary to evaluate risks to each
species individually. Rather, it is usually appropriate to group receptors with similar behaviors and
exposure patterns, and to evaluate the risks to each group. For aquatic and semi-aquatic receptors,
organisms are usually evaluated in three groups:

¦ Fish

¦ Benthic macro i nve rte brates

¦ Amphibians (aquatic life stages]

Evaluation of risks to terrestrial receptors will not be performed as part of this risk assessment
because the OUs included in this risk assessment were developed for human use and do not have
habitat that would support significant terrestrial receptor populations. In addition, potential risks to
terrestrial receptors in the forested areas that surround Libby and Troy, where there is habitat, are
being evaluated under the OU3 BERA.

1 Expo iways of Chief Concern

3.1.4.1 Fish

The primary exposure pathway of concern for fish is direct contact with asbestos fibers suspended in
surface water. Fish may also be exposed to asbestos by incidental ingestion of sediment while feeding,
ingestion of contaminated prey items, and direct contact with sediment Incidental ingestion of
sediment is likely to be a minor source of exposure, especially for fish (e.g., trout] that feed mainly in
the water column. Likewise, ingestion of prey items is likely to be minor because asbestos is not
expected to bioaccumulate in food web items. Direct dermal contact with sediment is also likely to be
minor, at least for fish that reside mainly in the water column.

3-2

Non-OU3 Asbestos BERA_FINAL.docx

smith

-------
Section 3 • Problem Formulation

3.1.4.2 Benthic Invertebrates

Benthic invertebrates that reside in the upper layer of stream sediment may be exposed to asbestos by
direct contact with surface water. In addition, benthic organisms may be exposed by direct contact
with fibers in sediment and/or sediment pore water, and also by ingestion of fibers while feeding in
the sediment. For this type of organism, distinguishing between direct contact and ingestion exposure
is often not possible, so the pathways are often evaluated together.

3.1.4.3 Amphibians

Amphibians (e.g., frogs, toads] inhabit both aquatic and terrestrial (mainly riparian] environments,
with early life stages being primarily aquatic and later life stages primarily terrestrial. In their early
aquatic life stages, amphibians may be exposed to contaminants in surface water mainly by direct
contact. They may also be exposed to contaminants in sediment by direct contact and incidental
ingestion and to contaminants in aquatic prey items by ingestion.

3.2 Management Goal and Assessment Techniques

3.2.1 Management Goal

A management goal is a statement of the basic objectives that the risk manager wishes to achieve at a
site. The overall management goal identified for ecological health at the Site for asbestos
contamination is:

Ensure adequate protection of ecological receptors within the Site from the adverse effects of
exposures to mining-related releases of asbestos to the environment. "Adequate protection" is
generally defined as the reduction of risks to levels that will result in the recoveiy and
maintenance of healthy local populations and communities of biota (EPA 1999],

litii lulation

A "population" can be defined in multiple ways. For the Site, the assessment populations are defined
as the groups of organisms that reside in locations that have been impacted by mining-related
releases. For aquatic receptor exposures to asbestos, these locations include the tributaries within the
Site (Cedar Creek, Cherry Creek, Fisher River, Flower Creek, Granite Creek, Libby Creek, Parmenter
Creek, O'Brien Creek, Pipe Creek, Quartz Creek, and Callahan Creek],

3.2.3 Assessment End points

Assessment endpoints are explicit statements of the characteristics of the ecological systems that are
to be protected. Because the risk management goals are formulated in terms of the protection of
populations and communities of ecological receptors, the assessment endpoints selected for use in
this problem formulation focus on endpoints that are directly related to the management goals, such
as mortality, growth, and reproduction.

If effects on these three assessment endpoints are absent or minimal, it is likely that ecologically
significant population-level effects will not occur.

3.2.4 Measur

Measures of effect are quantifiable ecological characteristics that can be measured, interpreted, and
related to the valued ecological components chosen as the assessment endpoints (EPA 1997,1998],
There are a number of alternative measures of effect that may be investigated as part of an ecological
risk assessment. Because there are no established toxicity benchmarks for LA in the literature to

-------
Section 3 • Problem Formulation

support a hazard quotient derivation approach, this risk assessment relies upon the results from 0U3
(Part 1] for characterizing measures of effects at the Site.

3-4

Non-OU3 Asbestos BERA_FINAL.docx

-------
Section 4

Risk Characterization for Fish

This section presents the risk characterization for fish, including information on known reported
effects of asbestos on fish, a summary of results from Part 1, risk characterization, and a summary for
fish.

4, L Reported Effects

As noted in in Section 4.1 of Part 1, adverse effects in fish resulting from exposure to asbestos have not
been extensively studied, but several relevant reports were located related to the toxicity of chrysotile.
A range of effects have been reported from asbestos exposures including, but not limited to, epidermal
| hypertrophy and necrosis in kidneys and gills, decreased weight, development of epidermal tumors,
| thickening of epidermal tissue, increased mucous cell density in the intestinal tract, constricted kidney
j tubules, abnormal levels of lipids, endoplasmic reticulum in the liver, reduced reproduction, and
I lateral line degradation. Outside of the studies performed in support of 0U3, no studies were located
on the toxicity of LA to fish.

4.2 OU3 Results

Four lines of evidence are available to help evaluate the effects of exposure of fish to LA in site waters,
including:

¦ In situ toxicity studies of eyed eggs and alevins

¦ In situ toxicity tests of juvenile trout

¦ Fish population studies

¦ Resident fish lesion studies

The population studies indicates that trout population structure in lower Rainy Creek (LRC] is
different from reference streams, with decreased fish density, increased fish size, and slightly
decreased biomass. This observation could be consistent with a hypothesis that LA in site waters is
toxic to trout and results in a decreased number of fish, but several observations suggest that LA is not
the likely cause of the difference:

¦ There are several habitat quality factors that are lower in LRC than reference streams
(especially spawning gravel, woody debris, water temperature, and pool availability]. These
habitat factors show a relatively strong correlation with trout density, suggesting that habitat
likely accounts for much of the apparent difference.

¦ In situ toxicity studies of early life stage trout indicate there might be a small decrease in
hatching success of eyed eggs in lower Rainy Creek than in reference streams, but this cannot be
attributed to LA. Moreover, the difference is sufficiently small (<10%] that a substantial effect
on population density would not be expected (Toll et al. 2013],

ISk

smitn

Non-OU3 Asbestos BERA FINAL.docx

4-1

-------
Section 4 • Risk Characterization for Fish

¦ No effects that might contribute to decrease survival of larger fish have been detected, either in
caged juvenile fish studies or studies of resident fish. This is consistent with numerous other
studies which indicate that early life stages of fish are usually more sensitive to toxicants that
larger fish.

Taken together, the weight of evidence suggests that LA in waters of LRC is not causing adverse effects
on resident trout. By extension, effects of LA on fish in the Kootenai River (including federally
protected species such as the white sturgeon and bull trout] are therefore not of concern, since
concentrations of LA in the Kootenai River are substantially lower than in LRC.

Confidence in this conclusion is medium to high. However, observations from the in situ exposure
studies are limited to the conditions and concentration values that occurred during the studies, and if
substantially higher concentrations were to occur in other years, the consequences, if any, cannot be
predicted. While observations from fish population surveys are often variable between years, results
at this site were relatively consistent across two years, so confidence in these studies is good.

4.3 Risk Characterization

' Situ evin Exposure Study

Mean concentrations of total LA in sediment porewater in lower Rainy Creek measured during the 0U3
study ranged from 41 to 42 million fibers per liter (MFL] and 9 to 31 MFL in the overlying surface water.
In comparison, total LA concentrations in sediment porewater in tributaries at the Site (measured using
the same porewater sampling and analysis methods as employed during the in situ study] are
substantially lower, with only one sample having total LA detected at 0.3 MFL (see Table 2-5], Total LA
concentrations in surface water in tributaries at the Site are also substantially lower, ranging from non-
detectto 0.084 MFL on average (see Table 2-3],

Because concentrations of LA in sediment porewater and surface water in Site tributaries are
considerably lower than concentrations in lower Rainy Creek, it can reasonably be expected that
effects in fish exposed to LA in tributaries at the Site would be less than those observed in the OU3
study. Because the in situ eyed egg study concluded that effects to fish from LA were minimal and not
considered to be large enough to be of significant ecological concern in OU3, the same is concluded for
the Site.

4.3.2 In Situ Juvenile Fish Study

For the OU3 study, average total LA concentrations in lower Rainy Creek surface water ranged from
about 10 to 30 MFL, with an apparent tendency to increase in the downstream direction. LA was
occasionally detected in one reference location (mean = 2.9 MFL], but was only rarely detected at
another reference location (mean <0.02 MFL], Total LA surface water concentrations in tributaries at
the Site are lower than those observed in lower Rainy Creek during the study, and ranged from non-
detectto 0.084 MFL on average (see Table 2-3], Therefore, because LA concentrations at the Site are
lower than OU3, it can reasonably be assumed that there would also be no adverse impacts on survival
or growth in juvenile trout and no external and histological lesions in juvenile trout as a consequence
of exposures to LA in surface water at the Site.

4.3.3 In Situ Lesion Study

Based on the findings of the OU3 in situ lesion study, and because concentrations of total LA in surface
water in tributaries at the Site are lower than those measured in OU3 (see Table 2-3], it can
reasonably be expected that LA is not a contributing factor to lesion development in trout at the Site.

4-2 smith

-------
Section 4 • Risk Characterization for Fish

4.4 Fish Summary

Risks to fish populations based on in situ early life stage toxicity testing, in situ lesions and population
surveys suggests that LA in surface water and porewater at 0U3 is not causing adverse effects on
resident trout. By extension, effects of LA on fish in tributaries at the Site are not of concern, since
concentrations of LA in Site waters are substantially lower than concentrations in 0U3.

-------
Section 4 • Risk Characterization for Fish

This page left blank intentionally

4-4

Non-OU3 Asbestos BERA_FINAL.docx

-------
Section 5

Risk Characterization for Benthic
Macroinvertebrates

This section presents the risk characterization for benthic macroinvertebrates, including information
on known reported effects of asbestos on benthic macroinvertebrates, a summaiy of results from Part
1, risk characterization, and a summary for benthic macroinvertebrates.

5.1 Reported Effects

As noted in Section 5.1 of Part 1, adverse effects in benthic macroinvertebrates resulting from
exposure to asbestos have not been extensively studied, but several relevant reports were located
related to the toxicity of chrysotile. A range of effects have been reported including, but not limited to,
decrease in larval survival, reduced weight gain, and a reduction in release of larva by brooding adults.
Outside of those studies performed for 0U3, no studies were located on the toxicity of LA to benthic
macroinvertebrates.

5.2 OU3 Results

Two lines of evidence are available to evaluate effects of site contaminants on benthic
macroinvertebrates, including:

¦ Laboratory-based site-specific sediment toxicity tests in two species of organism (H. azteca, and
C. tentans)

¦ Site-specific benthic community population studies, augmented with habitat quality studies

The site-specific sediment toxicity tests indicate that effects on growth and reproduction were not
apparent in H. azteca, and were minor in C. tentans. However, an effect of site sediment on survival
was noted in both species, with C. tentans being more impacted (9-25% decrease] than H. azteca (4-
6% decrease]. It is difficult to judge if LA is the likely cause, because quantitative estimates of LA
concentration in the two site sediments are sufficiently uncertain that the presence of a dose-response
relationship cannot be ascertained. Even if LA is the cause, the applicability of these results to other
species, and hence the potential magnitude of effects on the benthic macroinvertebrate community as
a whole, are difficult to judge from this line of evidence alone, and are best determined by evaluating
the site-specific population studies presented below.

The site-specific population studies suggest that benthic macroinvertebrate communities along lower
Rainy Creek may occasionally rank as slightly impaired compared to off-site reference locations, but
are not impaired compared to upper Rainy Creek. The differences are not extensive and might be due,
at least in part, to differences in habitat quality.

Taken together, these findings support the conclusion that LA contamination in lower Rainy Creek
may be causing small to moderate effects on survival of some species, but the overall benthic
macroinvertebrate community is not substantially impacted.

ISk

smitn

Non-OU3 Asbestos BERA FINAL.docx

5-1

-------
Section 5 • Risk Characterization for Benthic Macroinvertebrates

Confidence in this conclusion is medium to high. One potential limitation to the site-specific studies is
that the test species are not expected to occur in mountain streams, and native species (mainly
mayflies, stoneflies, caddisflies, true flies, and beetle larvae] might have differing sensitivities. While
benthic community and habitat surveys often display considerable variability between years, in this
case the results are relatively consistent between two years, providing good confidence in the survey
results.

5.3 Risk Characterization

The concentration of LA in 0U3 sediments was estimated to be 3% and 5% during the 0U3 benthic
macroinvertebrate toxicity tests for H. Azteca and C. tentans based on analysis by PLM. Concentrations
in Site creek sediments are less than those at 0U3 during these studies, ranging from non-detect to
trace (see Table 2-4], Because Site sediment LA concentrations are lower than in 0U3 sediments, it
can reasonably be expected that impacts on growth, survival, and reproduction of benthic
macroinvertebrates due to LA in sediment at the Site would be less than those observed for 0U3.

5.4 Benthic Macroinvertebrate Summary

Risks to benthic macroinvertebrate populations based on laboratory toxicity testing and population
surveys suggests that LA in surface waters at 0U3 is not causing adverse effects on benthic
macroinvertebrates. By extension, effects of LA on benthic macroinvertebrates in tributaries at the
Site are not of concern, since concentrations of LA in Site sediments are substantially lower than
concentrations in 0U3.

5-2

Non-OU3 Asbestos BERA_FINAL.docx

-------
Section 6

Risk Characterization for Amphibians

This section presents the risk characterization for amphibians, including information on known
reported effects of asbestos on amphibians, a summary of results from Part 1, risk characterization,
and a summary for amphibians.

6.1 Reported Effects

Outside of those performed for 0U3, no studies were located on effects of asbestos exposure on
amphibian species.

6.2 OU3 Results

| Two lines of evidence are available to evaluate potential effects of LA on amphibians in 0U3:

I * A site-specific laboratory-based sediment toxicity test

¦ A field survey of gross and histologic lesion frequency and severity in amphibians collected
from 0U3 and from reference areas

The site-specific sediment toxicity test did not produce any signs of overt toxicity in any organisms
exposed to 0U3 sediment. Both survival and growth were higher in organisms exposed to 0U3
sediment than for a reference sediment. The only observation of potential concern was an apparent
increase in the time to metamorphosis for some organisms that were exposed to 0U3 sediment. The
ecological significance of this apparent lag in the final stages of development is not certain, but
assuming the effect is only a lag (as opposed to an actual cessation of development], it is suspected the
effects would likely not be ecologically meaningful. However, it is plausible that the delay might
become important if ponds in high exposure areas were to dry up during this critical stage of
development.

The survey of external and histological lesions in field-collected organisms indicates that lesions in
organisms from 0U3 are not more frequent or more severe than in organisms from reference sites,
and that all lesions observed are likely the result of parasitism rather than asbestos exposure. This
supports the conclusion that LA is not causing any external or internal malformations of concern.

Taken together, these findings support the conclusion that sediments and waters in 0U3 are not likely
to be causing any ecologically significant adverse effects on amphibian populations.

Confidence in this conclusion is medium to high. The most significant uncertainty is whether the
apparent delay in the final stages of metamorphosis might be of concern. Further studies would be
needed to determine if the apparent lag in final stage development is reproducible, and whether
complete metamorphosis is ultimately achieved in exposed organisms.

ISk

smitn

Non-OU3 Asbestos BERA FINAL.docx

6-1

-------
Section 6 • Risk Characterization for Amphibians

6.3 Risk Characterization

6.3.1 Laboratory Toxicity Tests

The concentration of LA in the 0U3 sediment treatment used in the 0U3 toxicity test was estimated to
be between 4% and 7% based on analysis by PLM. Concentrations in Site creek sediments are less
than those at 0U3, ranging from non-detectto trace (see Table 2-4], Because Site sediment LA
concentrations are lower than 0U3 sediments, it can reasonably be expected that impacts on growth,
survival, and time to metamorphosis of amphibians at the Site would be less than observed for 0U3.

6.3.2 In Situ Lesion Studies

The concentration of LA in sediments from the 0U3 ponds during the 0U3 in situ lesion studies was
estimated to be between <0.2% and 5% based on analysis by PLM. Concentrations in Site creek
sediments are generally less than those evaluated in this study, ranging from non-detect to trace (see
Table 2-4], Concentrations of total LA measured in surface water samples from the OU3 ponds ranged
from 6.7 to 26 MFL on average. Total LA concentrations in Site creek surface water are less than those
evaluated in this study, ranging from non-detect to 0.084 MFL on average (see Table 2-2],

Because Site sediment and surface water LA concentrations are lower than in OU3, it can reasonably
be assumed that impacts on growth and lesion development in amphibians due to LA in surface water
and sediment at the Site would be less than observed for OU3.

6.4 Amphibian Summary

Risks to amphibian populations based on laboratory toxicity testing and in situ lesion studies suggests
that LA in surface water and sediment at OU3 is not causing adverse effects on amphibians. By
extension, effects of LA on amphibians in tributaries at the Site are not of concern, since
concentrations of LA in Site surface water and sediment are substantially lower than concentrations in
OU3.

6-2

Non-OU3 Asbestos BERA_FINAL.docx

-------
Section 7

Uncertainty Assessment

Quantitative evaluation of ecological risks is generally limited by uncertainty regarding a number of
important data. This lack of knowledge is usually circumvented by making estimates based on
whatever limited data are available, or by making assumptions based on professional judgment when
no reliable data are available. Because of these assumptions and estimates, the results of the risk
characterizations are themselves uncertain, and it is important for risk managers and the public to
keep this in mind when interpreting the results of a risk assessment The following text summarizes
the key sources of uncertainty influencing the results of this BERA.

7» L Uncertainties in Nature and Extent of Contamination

esentativeness of Samples Collected

j Concentration levels of LA in environmental media can vary as a function of location, and may also
j vary as a function of time. Thus, samples collected during a field sampling program may or may not
fully characterize the spatial and temporal variability in actual concentration levels. At the Site, field
samples were collected in accordance with SAPs/QAPPs that specifically sought to ensure that
samples were representative of the range of conditions across each exposure area (e.g., surface water
samples were collected during both high and low flow conditions]. However, in some locations, the
number of samples collected was relatively small. Thus, without the collection of very large numbers
of samples over both space and time, some uncertainty remains as to whether the samples collected
provide an accurate representation of the distribution of concentration values actually present.

In addition, it was not possible to sample all bodies of water at the Site. Sampling did occur in creeks
thought to have the greatest potential to have LA present due to previous efforts to stabilize the banks
with riprap material and are representative of what the "worst case" scenario could be for the Site.

Lastly, the fishing pond in 0U5 is yet to be filled in with water, making evaluation of future risks to
aquatic receptors difficult. Because Libby Creek will serve as the water source for the pond, samples
collected from Libby Creek were used as surrogates for the future pond. This introduces another level
of uncertainty in that there is the potential for LA to settle in the pond and not have equal
concentrations of LA as Libby Creek. Additionally, sediment data are not available for the fishing pond.
However, sampling of the subsurface soil where the pond was dug indicates that LA was not present at
the time of excavation. The results for these samples are presented in Appendix C. It could therefore
be assumed that LA is not present in the sediment of the pond at levels observed in 0U3 sediments.

iracy of Analyti jasurements

Unlike traditional chemistiy methods, where analytical results are based solely on the output of a
laboratory instrument, analytical results for asbestos are dependent upon subjective analyst
interpretations. Thus, high data quality is ensured through the use of laboratories and analysts that
are well-trained in asbestos analysis, and specifically trained in the analysis of LA.

ISk

smitn

Non-OU3 Asbestos BERA FINAL.docx

7-1

-------
Section 7 • Uncertainty Assessment

All analytical laboratories participating in the analysis of samples for the Site are accredited by the
National Institute of Standards and Technology (NIST] National Voluntary Laboratory Accreditation
Program (NVLAP] for the analysis of asbestos by transmission electron microscopy (TEM] and/or
PLM. This accreditation process includes the analysis of NIST/NVLAP standard reference materials, or
other verified quantitative standards, and successful participation in two rounds of proficiency testing
per year each of bulk asbestos by PLM and airborne asbestos by TEM as supplied by NIST/NVLAP. In
addition, each laboratory working for the Site is also required to pass an onsite EPA laboratory audit,
participate in ongoing analytical discussions with other project laboratories, and meet Site-specific
data reporting requirements.

Even with these quality assurance (QA] measures in place, due to the subjective nature of both TEM
and PLM analyses, results can differ between analysts and laboratories. Because of this, the analytical
quality control (QC] program for the Site performs regular evaluations of both within- and between-
laboratory variability in asbestos results for both analytical methods. A detailed evaluation of Site
QA/QC is presented in EPA (2014], In addition, information pertaining to laboratory audits and data
validation has been summarized in Annual Quality Assurance/Quality Control (QA/QC) Summary
Report (2010-2012] (CB&I Federal Services, LLC [CB&I] 2014], The following sections summarize
some of the method-specific uncertainties of the data utilized in the BERA.

7.1.2.1 TEM

When analyzing a filter for asbestos, the TEM analyst visually scans prepared grids for potential
asbestos fibers. When a structure is observed, the distinction between asbestos/non-asbestos and
asbestos type (e.g., chrysotile, actinolite, amosite] is determined based on a visual assessment of the
structure-specific selective area electron diffraction pattern and energy dispersive spectra (EDS],
comparing them to a spectral library of known asbestos types. Interpretation of the EDS requires
significant training as LA is inclusive of a range of asbestos mineral types (EPA 2008], EDS
interpretation is further complicated by the fact that spectra can differ between TEM instruments,
chemical composition can differ within an asbestos structure (e.g., the EDS obtained at the end of a
fiber may differ from the EDS at the center point], and spectra can be influenced by surrounding
matrix particles.

Results of the TEM laboratory QC analyses show that there are differences in structure counting and
recording methods within and between the analytical laboratories, with within-laboratory precision
being better than between-laboratory (CDM Smith 2014], Grid opening re-examination (recount]
results show there were some differences noted in the number of LA structures counted and in the
differentiation of LA structures from non-asbestos material structures with EDS that are similar to LA
(e.g., pyroxene]. Yet, despite these differences, the number of LA structures counted usually only
differed by one structure. For surface water and porewater samples, the between-laboratory
differences in structure counting and recording methods are not likely to be a large source of
uncertainty in reported water concentrations.

7.1.2.2 PLM

Most of the PLM methods currently available for the analysis of asbestos in solid media were
developed for the analysis of building materials containing relatively high asbestos levels and are not
generally intended for assessing low-level (<1%] asbestos contamination in soil. Indeed, even the
Libby-specific PLM visual area estimation (PLM-VE] method is not able to reliably detect the levels of
LA in soil below about 0.2% by mass (EPA 2008], When performing a PLM-VE analysis, the analyst
utilizes visual estimation techniques (e.g., standard area projections, photographs, drawings, or

7-2

Non-OU3 Asbestos BERA_FINAL.docx

-------
Section 7 • Uncertainty Assessment

trained experience] to estimate the asbestos content of the soil and results are reported semi-
quantitatively based on visual comparisons to LA-specific reference materials. The "detection limit"2 is
dependent upon the ability of the analyst, but is typically about 0.2% to 0.3% LA (by mass] (EPA
2008], This means that soil LA concentrations below about 0.2%, may not be reliably identified by
PLM-VE, and some soils ranked as Bin A (non-detect] by PLM-VE likely contain low levels of LA that
cannot be reliably detected. Thus, the difference between Bin A (non-detect] and Bin B1 (trace LA
present at levels less than 0.2 %] is not always distinct. As such, result reproducibility is especially
difficult for Bin A and Bin B1. Because risk conclusions do not differ for sediments that are Bin A
versus Bin Bl, the distinction between these two bins is not critical.

7.2 Uncertainties in Exposure Assessment

Exposure pathways selected for quantitative evaluation in this risk assessment do not include all
potential exposure pathways for all ecological receptors. Exposure pathways that were not evaluated
include:

¦ Inhalation of dust particles for amphibians

¦ Ingestion of prey items for fish, benthic macroinvertebrates, and amphibians

Omission of these pathways may tend to lead to an underestimation of total risk to the exposed
receptors. As discussed previously in Section 3.1.4, many of these exposure pathways are likely to be
minor compared to other pathways that were evaluated, and the magnitude of the underestimation is
not likely to be significant in most cases. However, the exclusion of some exposure pathways may tend
to underestimate predicted risks in some cases.

7.3 Uncertainties in Toxicity Assessment

»nce of Toxic < <

As noted in the sections above, adverse effects in aquatic receptors resulting from exposure to
asbestos have not been extensively studied, but several relevant reports were located related to the
toxicity of chrysotile. No studies were located on the toxicity of LA to aquatic receptors. Because of
this, this risk assessment relied heavily on studies performed at OU3 (see below],

7.3.2 Extrapolation of OU3 Study Res e Site

Because toxicity data for LA effects on aquatic receptors are not available in the literature, this risk
assessment utilized information gathered during the remedial investigation for OU3. This information
is applicable to the Site due to the proximity of OU3 to the Site and the similar environmental
conditions. Concentrations of LA measured in environmental media at the Site are lower than
concentrations in OU3, making it reasonable to draw the conclusion that ecological impacts due to LA
exposures at the Site would be less than what has been observed at OU3.

2 For this report, the "detection limit" is defined as the concentration that must be present in a sample such that the method
will be able to detect LA 95% of the time.

-------
Section 7 • Uncertainty Assessment

7.3.3 Uncertainties 3 Studies

As noted in the Weight of Evidence evaluations presented in Part 1 (see Sections 4.5, 5.5, and 6.5 in
Part 1], confidence in the conclusions of the 0U3-specific studies is generally medium to high, but
there were specific uncertainties noted with each study. Despite these uncertainties, the conclusions
from these studies are directly applicable to the Site because of the similarities in the exposure media,
ecological receptors, and exposure pathways between 0U3 and Site.

7.4 Uncertainties in Risk Characterization

Assessment endpoints for the receptors at the Site are based on the sustainability of exposed
populations, and risks to some individuals in a population may be acceptable if the population is
expected to remain healthy and stable. However, even if it is possible to accurately characterize the
distribution of risks or effects across the members of the exposed population, estimating the impact of
those effects on the population is generally difficult and uncertain. The relationship between adverse
effects on individuals and effects on the population is complex, depending on the demographic and life
history characteristics of the receptor being considered as well as the nature, magnitude and
frequency of the stresses of LA and associated adverse effects. Thus, the actual risks that will lead to
population-level adverse effects will vary from receptor to receptor.

7.5 Summary of Uncertainties

Although there are some uncertainties and limitations associated with the conclusions for the Site as
noted above, these uncertainties do not erode confidence in the overall finding that ecological
receptors at the Site are unlikely to be adversely impacted by LA exposures.

smith

Non-OU3 Asbestos BERA_FINAL.docx

-------
Section 8

Summary and Conclusions

EPA planned and performed a number of studies to investigate whether ecological receptors in 0U3 of
the Libby Asbestos Superfund Site were adversely impacted by the presence of LA in the environment.
These studies and their findings have been applied to the Site because of similar ecological settings.

For aquatic receptors, studies of fish, benthic macroinvertebrates, and amphibians exposed to LA in
surface water, sediment, or porewater from 0U3 revealed no evidence of ecologically significant
effects that were attributable to LA. These studies indicate that aquatic receptors in 0U3 are unlikely
to be adversely impacted by LA. Because concentrations of LA at the Site in environmental media are
substantially lower than those in 0U3, it can reasonably be expected that aquatic receptors at the Site
are also unlikely to be adversely impacted by LA.

For terrestrial receptors, because the 0U3 boundary has not been formally delineated, the forested
areas surrounding Libby and Troy were evaluated as part of the 0U3 risk assessment (Part l].The
land at the Site has largely been developed for human use, both residential and commercial use, and
habitat is not optimal to support terrestrial receptors. Because of this, terrestrial receptors were not
evaluated in this risk assessment.

ISk

smith

Non-OU3 Asbestos BERA_

FINAL.docx

8-1

-------
Section 8 • Summary and Conclusions

This page left blank intentionally

8-2

Cs»mnh

Non-OU3 Asbestos BERA_FINAL.docx

-------
Section 9

References

CB&I. 2014. Annual QA/QC Summary Report (2010-2012] for Task Order 2019 Quality Assurance
(QA] Support for the Libby Asbestos Site. In preparation (2014],

CDM Smith 2007. Flower Creek Investigation Summaiy Memo. Prepared for Volpe. November.

. 2008a. Summary of Creek Investigations Completed for Libby Asbestos Superfund Site

Operable Units 4 and 7, October 2008.

. 2008b. Granite-Callahan Investigation Summary Memo. Prepared for Volpe. May.

| . 2008c. Addendum to the Response Action Work Plan Flower Creek Removal Plan Libby,

I Montana. June 20.

j . 2008d. Addendum to the Response Action Work Plan Granite Creek Removal Plan Libby,

Montana. July 18.

. 2008e. Addendum to the Response Action Work Plan Callahan Creek Removal Plan Libby,

Montana. July 25.

. 2008f. Response Action Sampling and Analysis Plan, Revision 1, Libby Asbestos Project, Libby,

Montana. April 9.

. 2009a. Addendum to the Response Action Work Plan Libby Creek Removal Plan. Libby,

Montana. June 5.

. 2009b. Addendum to the Response Action Work Plan Pipe Creek Removal Plan. Libby, Montana.

June 5.

EPA. 1997. Ecological Risk Assessment Guidance for Superfund: Process for Designing and Conducting
Ecological Risk Assessments. Interim Final. U.S. Environmental Protection Agency, Environmental
Response Team, Edison, NJ.

. 1998. Guidelines for Ecological Risk Assessment. U.S. Environmental Protection Agency.

EPA/63 0/R-95/002F.

. 1999. Issuance of Final Guidance: Ecological Risk Assessment and Risk Management Principles

for Superfund Sites. U.S. Environmental Protection Agency, Office of Solid Waste and Emergency
Response Directive. Washington DC, EPA-9285-7-28-P.

. 2008. Performance Evaluation of Laboratory Methods for the Analysis of Asbestos in Soil at the

Libby, Montana Superfund Site. Prepared by US Environmental Protection Agency with Technical
Assistance from Syracuse Research Corporation. October 7, 2008.

ISk

smitn

Non-OU3 Asbestos BERA FINAL.docx

9-1

-------
Section 9 • References

. 2011. Water Source Identification Study - Phase I Sampling and Analysis Plan. Libby Asbestos

Superfund Site. Libby, Montana. Prepared for the U.S. Environmental Protection Agency by CDM
Federal Programs Corporation. November 1.

. 2012a. Water Source Identification Study - Phase II Sampling and Analysis Plan/Quality

Assurance Project Plan. Libby Asbestos Superfund Site. Libby, Montana. Prepared for the U.S.
Environmental Protection Agency by CDM Federal Programs Corporation. April.

. 2012b. Nature and Extent of LA Contamination in Surface Water and Sediment - Sampling and

Analysis Plan/ Quality Assurance Project Plan Libby Asbestos Superfund Site, Operable Unit 4. Libby,
Montana. Prepared for the U.S. Environmental Protection Agency by CDM Federal Programs
Corporation. May 4.

. 2013a. Data Summary Report: Water Source Identification Study. Libby Asbestos Superfund

Site. Libby, Montana. Prepared for the U.S. Environmental Protection Agency by CDM Smith. June.

. 2013b. Data Summary ReportNature and Extentof LA Contamination in Surface Water and

Sediment. Libby Asbestos Superfund Site, Operable Unit 4. Libby, Montana. Prepared for the U.S.
Environmental Protection Agency by CDM Federal Programs Corporation. March.

. 2013c. Sampling and Analysis Plan/ Quality Assurance Project Plan: Sediment Porewater Study

of Kootenai River Tributaries. Libby Asbestos Superfund Site. Libby, Montana. Prepared for the U.S.
Environmental Protection Agency by CDM Smith. May.

. 2014. Quality Assurance and Quality Control Summary Report2010-2013.Libby Asbestos

Superfund Site. Libby, Montana. U.S. Environmental Protection Agency, Region 8. Prepared for the U.S.
Environmental Protection Agency by CDM Smith. May.

LibbyMT.com. 2013. Libby, Montana and Kootenai River Country, Kootenai River. Accessed at:
http://www.libbymt.com/areaattractions/kootenairiver.htm, on November 15, 2013.

Meeker G.P., Bern A.M., Brownfield I.K., Lowers H.A., Sutley S.J., Hoeffen T.M., and Vance J.S. 2003. The
composition and morphology of amphiboles from the Rainy Creek Complex, Near Libby, Montana.
American Mineralogist 88:1955-1969.

Toll J., Garber K., Deforest D., Brattin W. 2013. Assessing population-level effects of zinc exposure to
brown trout (Salmo trutta) in the Arkansas River at Leadville, Colorado. Integrated Environmental
Assessmentand Management 9(l]:50-62. doi: 10.1002/ieam.l325. Epub 2012 Sep 18.

smith

Non-OU3 Asbestos BERA_FINAL.docx

-------
Tables

-------
This page left blank intentionally

-------
Table 2-1. State Species of Concern

Libby Asbestos Superfund Site

Group

Rank

Amphibians

Coeur d'Alene Salamander (Plethodon idahoensis)

Boreal Toad, Green (also known as Western Toad ) (Bufo boreas)

Fish

Bull Trout (Salvelinus confluentus)

Torrent Sculpin (Cottus rhotheus)

Westernslope Cutthroat Trout (Oncorhynchus clarkii lewisi)

Invertebrates

Stonefly (Utacapnia columbiana)

Slug, Magnum Mantleslug (Magnipelta mycophaga )

S1S3

Slug, Pygmy Slug (Kootenaia burkei)

S1S2

Land Snail, Robust Lancetooth (Haplotrema vancouverense)

S1S2

Slug, Sheathed Slug (Zacoleus idahoensis)

S2S3

Land Snail, Smoky Taildropper (Prophysaon humile)

S1S3

Land Snail, Striate Disc (Discus shimekii)

51 = At high risk because of extremely limited and potentially declining numbers, extent and/or habitat, making it highly
vulnerable to global extinction or extirpation in the state.

52 = At risk because of very limited and potentially declining numbers, extent and/or habitat, making it vulnerable to global
extinction or extirpation in the state.

53 = Potentially at risk because of limited and potentially declining numbers, extent and/or habitat, even though it may be
abundant in some areas.

-------
Table 2-2. Description of Surface Water, Sediment, and Porewater Sampling Locations

Libby Asbestos Superfund Site

Water Body

Location #

Location Description

Downstream of last residence on Callahan Creek 1/4 mile up from Hwy 2

Callahan Creek

Hwy 2 bridge

Prior to confluence with Kootenai River

Cedar Creek

Upstream of Hwy 2 bridge, near standpipe

Cherry Creek

Northeast of Granite Creek Road bridge

Fisher River

Prior to confluence with Kootenai River

Small camp area approximately mile 8.8 Fisher River Road

Upstream of Balsam Street bridge

Outlet of Flower Creek Reservoir

Flower Creek

Upstream from regulating resevoir

Near Balsam Street bridge

Prior to confluence with Kootenai River at 2nd Street Ext bridge

West side of Hwy, South side of creek

Granite Creek

Near Granite Creek/Cherry Creek junction

Prior to confluence with Libby Creek

Upstream of operable unit 5 fire pond flume

Libby Creek

Northeast of Hammer Cutoff bridge

Libby Creek at Hwy 2 bridge

Prior to confluence with Kootenai River at 5th Street Ext bridge

O'Brien Creek

Prior to Kootenai River confluence

Near Rabbit O'Brien Creek Road bridge

Parmenter Creek

Northwest corner of bridge on Dome Mountain Avenue

Upstream of Kootenai River Road bridge, near stand pipe

Upstream of Bobtail cut off road bridge

Pipe Creek

Pipe Creek at Kootenai River Road

Prior to confluence with Kootenai River at Bothman Drive bridge

Pipe Creek north of Red Dop Saloon

Quartz Creek

Upstream of Kootenai River Road bridge

Ext = extension
Hwy = highway

-------
Table 2-3. Summary Statistics for LA in Surface Water

Libby Asbestos Superfund Site

Operable
Unit

Water Body

Number of
Samples

Number of
Detected
Samples

LA Detection
Frequency

Range of Detected
LA Results (MFL)

Mean
Concentration
(MFL)

Cedar Creek

17%

0.026

0.0043

Cherry Creek

All ND

Fisher River

All ND

Flower Creek

All ND

Granite Creek

25%

0.086-1

0.084

Libby Creek

27%

0.013-0.39

0.035

Parmenter Creek

All ND

Pipe Creek

17%

0.043 - 0.2

0.016

Quartz Creek

0.26

0.021

Fishing Pond

Data not available*

Callahan Creek

All ND

O'Brien Creek

All ND

*Surface water data have not been collected for the fishing pond in operable unit 5, surface water from Libby Creek will serve as a
surrogate. Libby Creek will be the water source for the fishing pond when it is filled in.

Notes:

LA = Libby amphibole asbestos

ND = non-detect

MFL = million fibers per liter

-------
Table 2-4. Summary Statistics for LA in Sediment

Libby Asbestos Superfund Site

Panel A: PLM-Results for Fine, Ground Fraction

Sediment LA Bin Results (PLM-VE)

Unit

Water Body

Samples

Analyses*

Number of
Detected

Detection
Frequency

Number of
Bin A(ND)

Number of
Bin B1 (Tr)

Number of
Bin B2 (<1%)

Number of
Bin C (>1%)

Fisher River

25%

Flower Creek

50%

Granite Creek

40%

Libby Creek

Pipe Creek

33%

Callahan Creek

67%

O'Brien Creek

Panel B: PLM-Grav Results for Coarse Fraction

Sediment LA Results (PLM-Grav)

Unit

Water Body

Samples

Analyses*

Number of
Detected

Detection
Frequency

Number of
ND

Number of
Tr

Number of
> Tr

Fisher River

Flower Creek

Granite Creek

Libby Creek

Pipe Creek

Callahan Creek

O'Bbrien Creek

*Preparation and laboratory quality control analyses (i.e., preparation and laboratory duplicates) are included in this table; thus, the
number of analyses may be greater than the number of samples.

Notes:

% = percent
< = less than

> = greather than or equal to

LA = Libby amphibole asbestos

PLM-Grav = polarized light microscopy - gravimetric

PLM-VE = polarized light microscopy - visual estimation

ND = non-detect

Tr = trace

-------
Table 2-5. Summary Statistics for LA in Porewater

Libby Asbestos Superfund Site

Operable
Unit

Water Body

Number of
Samples

Number of
Detected
Samples

LA Detection
Frequency

Range of Detected
LA Results (MFL)

Mean
Concentration
(MFL)

Fisher River

All ND

Flower Creek

All ND

Granite Creek

All ND

Libby Creek

50%

0.3

0.15

Pipe Creek

All ND

Callahan Creek

All ND

O'Brien Creek

All ND

Notes:

LA = Libby amphibole asbestos

ND = non-detect

MFL = million fibers per liter

-------
This page left blank intentionally

-------
Figures

-------
This page left blank intentionally

-------
Morart
Lake}

RexforcTS^ \
Othorn Eureka

Lake

Alkali
Lake
Frank
Lake

Murphy
1 Lake

Lake
Koocanusa

Upper ¦
¦StiUwate'i
Lake

Savage
Lake

Whitefish
, Lake

Whitefish

Columbia
FallsS*^

Ashley
Lake

Kalispel

Little
Bitterroot
Lake

Thompson
Lakes

Flathead
Lake \-J'

Noxort
Rapids

iilikOY

Figure 2-1

Site Location Map
Libby Asbestos Superfund Site | Libby, MT

A 0 4 8 16 SDIMI _
U ii " ' 1 i Smith

N Miles

Libby Asbestos Superfund Site

•EIBBY

Highway

MONTANA

River

Background Terrain Sources: Esri, USGS, NOAA
Road and Railroad Source: US Census Tiger/Line

Waterways and Waterbodies Source: National Hydrography Dataset - USGS

Waterbody

WYOMING

-------
Screening Plant
Former Buildings

BNSF Lfbby
Railyard
Former Export
Plant

Screening Plant
Former Buildings
Mine Disturbance
Area

Miles

Mining ¦ Related

Site Features

Smith FIGURE 2-2

-------
F jliakem
'Koocanusa,

FORMER LIBBY
/VERMICULITE MINE

OU3 (Study Area)

I OU3 (Study Area)

OU7

(Study Area)

OU6

CANADA

MONTANA

WYOMING

OU1 - Former Export Plant

OU2 - Former Screening Plant

OU3 - Former Libby Vermiculite Mine and
Kootenai River (Study Area)

[J OU4 - City of Libby

OU5 - Former Stimson Lumber Company

hazardous substances and/or pollutants or contaminants from such property, including, but not limited to, the mine
OU7 - Town of Troy property, the Kootenai River and sediments therein, Rainy Creek, Rainy Creek Road and areas in which tree bark is

contaminated with such hazardous substances and/or pollutants and contaminants.

OU8 - U.S. and Montana State Highway Aeriaisources: Esri, usgs, noaa

Corridors Source: Esri, DigitalGlobe, GeoEye, i-cubed, USDA, USGS, AEX, Getmapping, Aerogrid,

IGN, IGP, swisstopo, and the GIS User Community
Road and Railroad Source: US Census Tiger/Line

Waterways and Waterbodies Source: National Hydrography Dataset - USGS

Figure 2-3
Operable Unit Boundaries
Libby Asbestos Superfund Site | Libby, MT

Miles

<£K!th

-------
Figure 2-4. Wind Rose for LBBM8 (2007-2013)

WIND SPEED
(Knots)

>= 22
17-21
11-17
7-11
4-7
1 -4
Calms: 73.67%

WIND ROSE PLOT:

Station # 8 - Libby, MT

DISPLAY:

Wind Speed

Direction (blowing from)

COMMENTS:

DATA PERIOD:

COMPANY NAME:

Start Date: 1/1/2007 - 00:00
End Date: 12/31/2013-23:00

MODELER:

CALM WINDS:

73.67%

TOTAL COUNT:

54766 hrs.

AVG. WIND SPEED:

0.49 Knots

DATE:

1/20/2014

PROJECT NO.:

WRPLOT View - Lakes Environmental Software

Data obtained from Meso West

-------
txpanded ("exfoliated") vermiculite

Figure 2-5. Photographs of Vermiculite and Asbestos

Libby Asbestos Superfund Site

-------
^tmkeWJ
Koocanusa

Former Libby
Vermiculite Mine

Gmnitel
Creek

Cedar Creek ^8

IW.aTila iidlbs^lM

-MF

River
Railroad
Highway
Waterbody

LJ^

, Double V
N Lake \

I BR

BL i

I -
_

ILosf *7
: -LaifeL

' in

- •

•\N^«k

9^pHI

r n\

—X M

jM /J"

' -m

at> v

V7A 0U3 - Former Libby Vermiculite Mine
and Kootenai River (Study Area)

Note(s): EPA established the preliminary study area for the purposes of planning and developing the scope of the RI/FS
for Oil3. This study area may be revised as data are obtained during the Rlfor Oil3 on the nature and extent of
environmental contamination associated with releases that may have occurred from the mine site and any area
(including anystructure, soil, air, water, sediment or receptor) impacted by the release and subsequent migration of
hazardous substances and/or pollutants or contaminants from such property, including, but not limited to, the mine
property, the Kootenai River and sediments therein, Rainy Creek, Rainy Creek Road and areas in which tree bark is
contaminated with such hazardous substances and/or pollutants and contaminants.

Aerial Image Source: 2002 - Visual Intelligence Systems, Inc.

Road and Railroad Source: US Census Tiger/Line

Waterways and Waterbodies Source: National Hydrography Dataset - USGS

Figure 2-6

Surface Water Sampling Locations in 0U4
Libby Asbestos Superfund Site | Libby, MT

Miles

-------
^tmkeWJ
Koocanusa

Former Libby
Vermiculite Mine

Granite!
Creek

KPw^PiaBBIIia

SHoy

Cedar Creek

j" - "iwr' .

Libbyj

®T®1 T

Ij-Sj

IP' -. *jmB

BkjI

f¥

\ "*j

BMa DoublelEffttrX

' in

9^pHI

¦131

' -m

•\N^«k

River
Railroad
Highway
Waterbody

\7) 0U3 - Former Libby Vermiculite Mine
and Kootenai River (Study Area)

Note(s): EPA established the preliminary study area for the purposes of planning and developing the scope of
the RI/FS for OU3. This study area may be revised as data are obtained during the Rl for OU3 on the nature
and extent of environmental contamination associated with releases that may have occurred from the mine
site and any area (including anystructure, soil, air, water, sediment or receptor) impacted by the release and
subsequent migration of hazardous substances and/or pollutants or contaminants from such property,
including, but not limited to, the mine property, the Kootenai River and sediments therein, Rainy Creek, Rainy
Creek Road and areas in which tree bark is contaminated with such hazardous substances and/or pollutants
and contaminants.

Aerial Image Source: 2002 - Visual Intelligence Systems, Inc.

Road and Railroad Source: US Census Tiger/Line

Waterways and Waterbodies Source: National Hydrography Dataset - USGS

Figure 2-7

Sediment Sampling Locations in 0U4
Libby Asbestos Superfund Site | Libby, MT

Miles

-------
^tmkeWJ
Koocanusa

Former Libby
Vermiculite Mine

Granite!
Creek

W ' *jm, \

y^*v i

¦f; . -IV Vr

V ¦> ^e^aaldMjLcSsifocSBi

¦s^Wm
TSBM

IW.aTila iidlbs^/M

WmSmsHSM

Railroad
Highway
River

Porewater Sampling
Locations

Waterbody

Y7A OU3 - Former Libby Vermiculite Mine
and Kootenai River (Study Area)

Aerial Image Source: 2002 - Visual Intelligence Systems, Inc.

Road and Railroad Source: US Census Tiger/Line

Waterways and Waterbodies Source: National Hydrography Dataset - USGS

Figure 2-8

Porewater Sampling Locations in OU4
Libby Asbestos Superfund Site | Libby, MT

2.5

Miles

Smith

-------
PS

•v S\

\ 1 ¦

biv

,*>w:

jfwJralik^ ^

v: • : m§/

\v^ - - '^pfea *'

Railroad
Highway
River

Waterbody

Surface Water, Sediment and Porewater
Sampling Locations

Aerial Image Source: 2002 - Visual Intelligence Systems, Inc.

Road and Railroad Source: US Census Tiger/Line

Waterways and Waterbodies Source: National Hydrography Dataset - USGS

¦Plv%

rSJt -'X r'
&

tern

^'. V ^

is 1
¦ Ri ¦ I

ihi

S- *

r#*&M

- c<

A ?V A,

St'V

,;'\v /

' & t* f.

-"fc*

gfc m

mip • vi

. !p #

• ._•/»/ ...c^

¦-¦¦¦' • BE

, kl'5 W

sSlnl it 4<

*a • , iajH'

¦* •••-: * ™»,W>T®r

r* v

t "PBS? *rf
/ •

¦ *5®

Figure 2-9

Sampling Locations in 0U7
Libby Asbestos Superfund Site | Libby, MT

1.25

Miles

3DM
-J Smith

-------
Figure 3-1. Conceptual Site Model for Ecological Receptors to Asbestos

Past Mining
Operations

Historic
Instream
Riprap
Material

Historic
airborne
emissions

Current airborne
emissions

Surface Soil/Duff

Runoff,

Stream Bank Material

Surface Water

Sediment

Terrestrial food items

Aquatic food items

Ingestion

Inhalation

Direct Contact

Ingestion

Inhalation

Ingestion

Direct Contact

Ingestion

Direct Contact

Fish

Benthic Macro-

Amphibians

Birds/

Plants/Soil

invertebrates

Mammals

Invertebrates

Pathway is believed to be complete, and might be significant
Pathway is believed to be complete, but is probably minor
Pathway is incomplete or believed to be neg

-------
Appendices

I
I

-------
This page left blank intentionally

-------
Appendix A

Responses to Comments

EPA has compiled comments received from the U.S. Fish and Wildlife Service (FWS] and Montana
Department of Environmental Quality (MDEQ], Responses are provided below.

Responses to FWS Comments f!2 710 7141

Section 2.3.1:

1. Since the BERA covers all the other 0U, the species list should include species found in the
Kootenai River as well as other drainages. FWP would have a species list, but at the very least,
ESA listed fish species should be listed here. Brook trout were not found in Rainy Creek, I think
they were in the reference stream Bobtail Creek.

Response: Brook trout have been remove:! from the text Because the Kootenai River is being
evaluated as part of the 0U3 BERA, the other drainages are the only ones that ESA species are
needed for.

Section 2.3.2:

2. Most of these OUs will have numerous terrestrial receptors. Deer and many other mammals will
use these habitats more frequently, and many different species of birds protected by the
migratory bird treaty act will also be present. Instead of saying no habitat exists, the Service
suggests acknowledging that ecological receptors will use the site but risks should be lower than
the 0U3 site because concentrations are lower.

Response: The text does not indicate that there is no use, only that use by terrestrial receptors will
he limited due hy the quantity and/or quality of habitat and proximity to human disturbance,
Some use will occur, hut it is anticipated to he minor, and therefore exposure will he minor. The
text has been modified in several places to clarify this point

3. Some species will be more prevalent in these areas vs. forested areas

Response: See response to comment #2,

Section 2.3.3:

4. Suggest including this table and amending the text above to reflect this list.

Response: Because habitat is limited for terrestrial receptors, the table containing federally listed
terrestrial species has not been included however the presence of these species in the forested areas
are captured in the OUS BERA,

A-l

-------
Section 2.4.2:

5. Many of these creeks are listed critical habitat for bull trout and should be acknowledged
somewhere in this document. I inserted possible language above.

Response: Revisions to text have been accepted
Section 3.1:

6. Figure 3-1 should at the very least have an open circle air inhalation for mammals and birds.

Response: Rig nee 3-1 has been revised accordingly.

Section 3.1.3:

7. Habitat is present in these OUs that provide habitat for numerous birds protected by the
Migratory Bird Treaty Act, the Bald and Golden Eagle Protection Act, and the Endangered
Species Act. If you don't want to address the ecological risks in this document, at least
acknowledge that risks to terrestrial receptors is presented in the 0U3 BERA, and that risks at
these sites are expected to be lower.

Response: See response to comment #2,

8. Will cleanup recommendations for protection of critters at 0U3 be implemented at the other
OUs?

Response: No unacceptable risks to ecological receptors were observed in OUS, therefore no

cleanup recommendation will be made to protect ecological receptors in OUS or the other OUs,

Section 8:

The Service believes that terrestrial receptors should be addressed, and that they will be present
in these OUs. The work completed at 0U3 can be used to characterize risk as done for aquatic
receptors.

Response: See response to comment #2,

Responses to MDEO Comments f!2 715 7141
Section 1:

1. General Comment - DEQ appreciates being directed to the figures, tables, and appendices so
quickly.

Response: No response is necessary, the figures, tables, and appendices have been provided for
review.

Section 2.2:

2. Please make the description of 0U7 analogous to that of 0U4.

Response: The OU definition will not he revised as these are standard and are included throughout
other RPA documents.

A-2

-------
Section 2.3.2:

3. Does this mean nothing lives in the forested area? Or that only non-receptors live there? Please
clarify.

Response: The text has been revised to clarify the anticipated use of this OU hy terrestrial receptors,
Section 3.1.1:

4. Please provide a reference for this statement.

Response: The text has been revised to reference Figure 2-4, the wind rose, This presents the
prevailing wind direction,

Section 3.1.4.1:

5. Please give reasoning or reference for this suspicion.

Response: The text has been revised simitar to Part 1 (0U3) to provide further explanation
Section 3.2.4:

6. Could a hazard quotient derivation approach be used now that the LA toxicity values are in IRIS,
or is this a different toxicity benchmark?

Response: The toxicity values for LA recently made available in IRIS are applicable to human health
risk, not ecological risk and unfortunately are not applicable to this investigation.

Section 7.2:

7. Please provide a notation of the section where this discussion is located.

Response: Section 3,1,4 has been referenced.

Figures:

8. Please add the note from Figure 2-3 to Figures 2-6. 2-7, and 2-8.

Response: Figures have been adjusted as requested, with minor modification to the text describing
the 0U3 study area,

A-3

-------
This page left blank intentionally

-------
Appendix B

Detailed Sample and Analytical Information

Provided electronically

-------
This page left blank intentionally

-------
Appendix C

0U5 Confirmation Soil Samples from the
Fishing Pond

Provided electronically

-------
This page left blank intentionally

-------