DRAFT
DO NOT CITE OR QUOTE
£EPA
EP A/600/R-09/13 8A
December 2009
External Review Draft
The Effects of Mountaintop Mines and Valley
Fills on Aquatic Ecosystems of the Central
Appalachian Coalfields
NOTICE
This information is distributed solely for the purpose of predissemination peer review under
applicable information quality guidelines. It has not been formally disseminated by the U.S.
EPA. It does not represent and should not be construed to represent any Agency determination
or policy.
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC 20460
-------
DISCLAIMER
This information is distributed solely for the purpose of predissemination peer review
under applicable information quality guidelines. It has not been formally disseminated by the
U.S. EPA. It does not represent and should not be construed to represent any Agency
determination or policy.
ABSTRACT
This report assesses the state of the science on the environmental impacts of mountaintop
mines and valley fills (MTM-VF) on streams in the Central Appalachian Coalfields. These
coalfields cover about 48,000 square kilometers (12 million acres) in West Virginia, Kentucky,
Virginia and Tennessee, USA. Our review focused on the impacts of mountaintop removal coal
mining, which, as its name suggests, involves removing all or some portion of the top of a
mountain or ridge to expose and mine one or more coal seams. The excess overburden is
disposed of in constructed fills in small valleys or hollows adjacent to the mining site.
Our conclusions, based on evidence from the peer-reviewed literature and from the U.S.
Environmental Protection Agency's Programmatic Environmental Impact Statement released in
2005, are that MTM-VF lead directly to five principal alterations of stream ecosystems:
(1) springs, intermittent streams, and small perennial streams are permanently lost with the
removal of the mountain and from burial under fill, (2) concentrations of major chemical ions are
persistently elevated downstream, (3) degraded water quality reaches levels that are acutely
lethal to standard laboratory test organisms, (4) selenium concentrations are elevated, reaching
concentrations that have caused toxic effects in fish and birds and (5) macroinvertebrate and fish
communities are consistently and significantly degraded.
Preferred citation: U.S. EPA (Environmental Protection Agency). 2009. The Effects of Mountaintop Mines and
Valley Fills on Aquatic Ecosystems of the Central Appalachian Coalfields. Office of Research and Development,
National Center for Environmental Assessment, Washington, DC. EPA/600/R-09/138A.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 ii DRAFT—DO NOT CITE OR QUOTE
-------
CONTENTS
LIST OF TABLES vi
LIST OF FIGURES vii
LIST OF ABBREVIATIONS AND ACRONYMS viii
FOREWORD ix
AUTHORS AND REVIEWERS x
ACKNOWLEDGMENTS xi
1. EXECUTIVE SUMMARY 1
2. INTRODUCTION 3
2.1. OPERATIONS USED IN MTM-VF 4
2.2. REGULATORY CONTEXT 5
3. LOSS OF HEADWATER AND FOREST RESOURCES 12
3.1. ESTIMATING EXTENT OF HEADWATER ECOSYSTEM LOSS 12
3.2. LOSS OF HEADWATER ECOSYSTEM BIOTA 13
3.3. LOSS OF HEADWATER ECOSYSTEM FUNCTIONS 15
3.4. CUMULATIVE IMPACTS ON FOREST RESOURCES 16
3.4.1. Quantity of Forest Habitat Lost to Mountaintop Removal and Mining
Infrastructure 17
3.4.2. Quality and Connectivity of Forest Habitat Lost 17
3.4.3. Riparian Habitat Lost 17
4. IMPACTS ON WATER QUALITY 21
4.1. ALTERATION OF FLOW 21
4.2. CHANGES IN SEDIMENTATION 23
4.3. CHANGES IN CHEMICAL TRANSPORT AND BASIC WATER QUALITY
PARAMETERS 23
4.3.1. pH, Matrix Ions and Metals 23
4.3.2. Water Temperature 25
4.3.3. Nutrients 25
4.3.4. Dissolved Oxygen 26
4.4. CHANGES IN SEDIMENT CHEMISTRY .26
4.5. CUMULATIVE IMPACTS 26
5. TOXICITY TESTS 35
5.1. TOXICITY TESTS ON WATER COLUMN AND SEDIMENTS
DOWNSTREAM OF MTM-VF 35
5.2. TOXICITY TESTS ON WATER FROM OTHER ALKALINE COAL MINING
EFFLUENTS 36
5.3. TOXICITY OF MAJOR IONS: K+, HC03~, Mg2+, CI", S042", Na+, Ca2+ 37
5.3.1. Mount etal., 1997 37
5.3.2. Soucek (2007a, b); Soucek and Kennedy (2005) 38
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 iii DRAFT—DO NOT CITE OR QUOTE
-------
CONTENTS (continued)
5.3.3. Meyer et al., 1985 39
5.4. COMPARING TOXICITY TESTS ON MAJOR IONS TO OBSERVATIONS
DOWNSTREAM OF MTM-VF 39
5.5. TOXICITY OF TRACE METALS IN WATER ZZZZZZZZZ.41
5.5.1. Selenium
5.5.1.1. Invertebrates 41
5.5.1.2. Fish
5.5.1.3. Birds
5.5.2. Manganese and Iron 42
5.6. TOXICITY OF TRACE METALS IN SEDIMENT ZZZZZZZZZZZZ43
6. IMPACTS ON AQUATIC ECOSYSTEMS 46
6.1. EFFECTS ON BIOLOGICAL COMPOSITION ZZZZZZ.46
6.1.1. Benthic Macroinvertebrates 46
6.1.1.1, Benthic Macroinvertebrate Indices 46
6.1.1.2, Benthic Macroinvertebrate Diversity 47
6.1.1.3, Benthic Macroinvertebrate Density 48
6.1.1.4, Benthic Macroinvertebrate Functional Groups 48
6.1.1.5, Benthic Macroinvertebrate Taxa 48
6.1.2. Fish
6.2. EFFECTS ON ECOLOGICAL FUNCTION ZZZZZZZ51
6.3. BIOLOGICAL CONDITION ZZZ.51
6.4. RELATIONSHIP BETWEEN BIOLOGICAL METRICS AND
ENVIRONMENTAL FACTORS 52
6.4.1. Ion Concentration
6.4.2. Specific Metals
6.4.3. Organic and Nutrient Enrichment 53
6.4.4. Instream Habitat 53
6.4.5. Disturbance and Loss of Upland Habitat 54
6.5. CUMULATIVE EFFECTS ZZZZZZZZZZZZ.'.SS
7. RECLAMATION, MITIGATION AND RECOVERY 60
7.1. OVERVIEW OF RECLAMATION AND MITIGATION PRACTICES !Z"60
7.1.1. Reclamation 60
7.1.2. Reclamation Bonds 61
7.1.3. Mitigation 61
7.2. EVIDENCE OF RECOVERY: VEGETATIVE COVER Z! 63
7.3. EVIDENCE OF RECOVERY: MINESOILS * 64
7.4. EVIDENCE OF RECOVERY: WATER QUALITY, STREAM ECOLOGICAL
FUNCTION, AQUATIC BIOTA 66
7.4.1. Long-Term Effects on Downstream Water Quality 66
7.4.2. Created Headwater Streams 66
This document is a draft for review purposes only and does not constitute Agency policy.
12/1 iv DRAFT—DO NOT CITE OR QUOTE
-------
CONTENTS (continued)
8. SUMMARY, INFORMATION GAPS AND RESEARCH OPPORTUNITIES 70
8.1. A CONCEPTUAL MODEL OF THE IMPACTS OF MTM-VF 70
8.2. CONCLUSIONS 72
8.2.1. Loss of Headwater and Forest Resources 73
8.2.2. Impacts on Water Quality 73
8.2.3. Toxicity Impacts on Aquatic Organisms 74
8.2.4. Impacts on Aquatic Ecosystems 75
8.2.5. Cumulative Impacts of Multiple Mining Operations 75
8.2.6. Effectiveness of Mining Reclamation and Mitigation 75
8.3. INFORMATION GAPS AND RESEARCH OPPORTUNITIES 76
8.3.1. Update the MTM-VF Inventory and Surveys of Impact Extent 76
8.3.2. Quantify the Contributions of Headwater Streams 77
8.3.3. Improve Understanding of Causal Linkages 77
8.3.4. Develop Tests Using Sensitive Taxa 78
8.3.5. Further Investigate Selenium and Sediments 79
8.3.6. Quantify Cumulative Effects 80
8.3.7. Quantify Functionality of Constructed Streams 80
LITERATURE CITED 82
APPENDIX: A LITERATURE SEARCHES 94
APPENDIX: B REGULATORY ISSUES RELATED TO MTM-VF OPERATIONS 98
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 v DRAFT—DO NOT CITE OR QUOTE
-------
LIST OF TABLES
1. Watershed areas above the toe of valley fills approved from 1985 to 2001 18
2. Substrate measures 28
3. Proportion of sediments that were sand and fines (mean [standard error]) in paired sites 28
4. Water quality variables in unmined streams versus streams below valley fills 29
5. Water quality parameters for unmined or reference streams or streams downstream
from mined, filled, or filled and residential watersheds in West Virginia 31
6. Alkalinity, pH and metals in control streams and streams downstream from filled
watersheds in West Virginia 32
7. Dissolved oxygen, pH and conductivity for sites in eastern Kentucky 32
8. Seasonal mean (standard deviation) of conductivity (|iS/cm) for the four classes of
streams 33
9. Range of sediment concentrations of metals and arsenic (mg/kg) in streams down
stream from the sedimentation ponds below valley fills in 2002 and 2004 and from a
reference site in 2002 33
10. Polycyclic aromatic hydrocarbons, arsenic and metals detected in sediments of larger streams
in the Kanawha Basin 34
11. Field-based 20% effect concentrations (FEC20) (Linton et al., 2007) 44
12. Comparison of measured sediment concentrations with probable effects levels 44
13. Summary of research examining the relationship between mountaintop mining and ecological
characteristics in downstream habitats 56
14. Average ion concentration (reported as specific conductance) in MTM-VF and
reference streams reported in conjunction with biological data 58
15. Drainage areas for headwaters in the PEIS study area (KY, TN, WV, VA) 68
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 vi DRAFT—DO NOT CITE OR QUOTE
-------
LIST OF FIGURES
1. The Central Appalachian coalfields 7
2. A watershed view of a mountaintop mine and valley fill (no consistent scale) 8
3 Small stream watershed before and after mountaintop mining and creation of a valley
fill (simplified view, scales differ between upper and lower halves of diagram) 9
4. Satellite images of the 40-km2 Hobet 21 mine (Boone County, WV) (Panel A), and the
Washington DC area (Panel B), at the same scale 10
5. Permit boundaries for surface and underground mines in southwestern West Virginia 11
6. Headwater types 18
7. Map showing loss of headwater streams to MTM-VF 19
8. Hot spots of rarity-weighted richness for critically imperiled and imperiled species 20
9. Mn (black) and Fe (orange) deposits on a caddisfly collected downstream of a
mountaintop mine and valley fill 45
10. Conceptual model depicting stages of change in biological conditions in response to an
increasing stressor gradient 59
11. Earth movement by humans and streams 69
12. A conceptual model of the impacts of mountaintop mines and valley fills on aquatic
ecosystems 81
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 vii DRAFT—DO NOT CITE OR QUOTE
-------
LIST OF ABBREVIATIONS AND ACRONYMS
AMD
acid mine drainage
AOC
approximate original contour
ARRI
Appalachian Regional Reforestation Initiative
AWQC
ambient water quality criterion
BCG
biological condition gradient
CaC03
calcite
CaMg(C03)
dolomite
CaS04
calcium sulfate
CWA
Clean Water Act
DO
dissolved oxygen
EC 10
effect concentration for 10% of the tested organisms
EMAP
Environmental Monitoring and Assessment Program
EPA
U.S. Environmental Protection Agency
EPT
Ephemeroptera, Plecoptera and Trichoptera
FEC20
field-based effect concentration for 20% of the tested organisms
GIS
geographic information system
GLIMPSS
genus-level index of most probable stream status
HBI
Hilsenhoff Biotic Index
IBI
Index of Biotic Integrity
KS04
potassium persulphate
LC50
lethal concentration for 50% of the tested organisms
LOEC
lowest-observed-effect concentration
MBI
macroinvertebrate bioassessment index
MgS04
magnesium sulfate
MHRW
moderately hard reconstituted water
MTM-VF
mountaintop mines and valley fills
NPDES
National Pollutant Discharge Elimination System
OSM
Office of Surface Mining
PAH
polycyclic aromatic hydrocarbons
PEIS
programmatic environmental impact statement
SMCRA
Surface Mining Control and Reclamation Act
SOC
soil organic carbon
TDS
total dissolved solids
USACE
United States Army Corps of Engineers
WVSCI
West Virginia Stream Condition Index
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 viii DRAFT—DO NOT CITE OR QUOTE
-------
FOREWORD
Headwater streams and watersheds in Appalachia play a disproportionately large role in
the region's ecology. They are sources of clean, abundant water for larger streams and rivers,
are active sites of the biogeochemical processes that support both aquatic and terrestrial
ecosystems, and are characterized by exceptional levels of plant and animal endemism
(i.e., biodiversity hotspots). The benefits of healthy headwaters are cumulative as the critical
ecological functions of many small streams flowing into the same river system are necessary for
maintaining ecological integrity.
The practice of mountaintop mining and valley fills, which has become increasingly
common in Appalachian states, can have major environmental consequences for the mountain
ecosystem, the nearby valleys and downstream water quality. There is a growing body of
evidence in the scientific literature that valley fills from mountaintop mining are having
deleterious ecological effects. Recent published reports show that as water quality deteriorates
downstream of a valley fill, the biota within the stream are likewise affected.
The mining of coal in the United States is highly regulated. Mountaintop mining, in
particular, involves multiple statutes and agencies at both the federal and state levels. The two
key federal laws are the Surface Mining Control and Reclamation Act (SMCRA, 25 U.S.C. §
1201) and the Clean Water Act (CWA, 33 U.S.C. § 1252). The key entities at the federal level
are the Office of Surface Mining (OSM), the Environmental Protection Agency (EPA) and the
U.S. Army Corps of Engineers (USACE). On June 11, 2009, in a Memorandum of
Understanding, these agencies committed to a series of activities to improve the regulation of
mining practices under existing statutory authorities.
This assessment report is one of several actions EPA has initiated to better understand the
ecological impacts of mountaintop mining. For this report, the EPA Office of Research and
Development has reviewed and assessed the published peer-reviewed literature on the aquatic
impacts associated with mountaintop mining. This version of the assessment will undergo an
external peer review by EPA's Science Advisory Board. The final peer-reviewed assessment
will inform the EPA as it continues to implement its regulatory duties under the Clean Water
Act.
Michael W. Slimak, Ph.D., Associate Director
National Center for Environmental Assessment, Office of Research and Development
U.S. Environmental Protection Agency, Washington, DC
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 ix DRAFT—DO NOT CITE OR QUOTE
-------
AUTHORS AND REVIEWERS
AUTHORS
Susan B. Norton1
Michael Griffith2
Laurie Alexander1
Amina Pollard1
Glenn W. Suter II2
'U.S. Environmental Protection Agency
National Center for Environmental Assessment
Washington, DC 20460
2U.S. Environmental Protection Agency
National Center for Environmental Assessment
Cincinnati, OH 45268
REVIEWERS
R. Hunter Anderson1
Theodore R. Angradi2
Paolo D'Odorico3
Ken M. Fritz4
Kyle J. Hartman5
Brent R. Johnson4
Teresa Norberg-King2
J. Todd Petty5
Bernard W. Sweeney6
'U.S. Environmental Protection Agency, Office of Research and Development, National Center
for Environmental Assessment, Cincinnati, OH
2U.S. Environmental Protection Agency, Office of Research and Development, National Health
and Environmental Effects Research Laboratory, Mid-Continent Ecology Division, Duluth, MN
department of Environmental Sciences, University of Virginia, Charlottesville, VA
4U.S. Environmental Protection Agency, Office of Research and Development, National
Exposure Research Laboratory, Cincinnati, OH
5Division of Forestry, West Virginia University, Morgantown, WV
6Stroud Water Research Center, 970 Spencer Road, Avondale, PA
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 x DRAFT—DO NOT CITE OR QUOTE
-------
AUTHORS AND REVIEWERS continued
ACKNOWLEDGMENTS
We gratefully acknowledge Stephania Shamet of EPA Region 3 who wrote Appendix B.
Kate Schofield of the National Center for Environmental Assessment produced the conceptual
model diagram. Chris Russom, Chuck Stephan and Dave Mount of the National Health and
Environmental Effects Research Laboratory, Mid-Continent Ecology Division provided valuable
assistance on literature reviews and interpretation of the toxicity of ions. Contract support was
provided by Tetra Tech under contract EP-C-07-068.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 xi DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
1. EXECUTIVE SUMMARY
Mountaintop mines and valley fills (MTM-VF) lead directly to five principal alterations
to stream ecosystems: (1) springs, intermittent streams, and small perennial streams are
permanently lost with the removal of the mountain and from burial under fill, (2) concentrations
of major chemical ions are persistently elevated downstream, (3) degraded water quality reaches
levels that are acutely lethal to standard laboratory test organisms, (4) selenium (Se)
concentrations are elevated, reaching concentrations that have caused toxic effects in fish and
birds, and (5) macroinvertebrate and fish communities are consistently and significantly
degraded. These conclusions are based on evidence, described in this report, from the
peer-reviewed literature and from the U.S. Environmental Protection Agency (EPA)
Programmatic Environmental Impact Statement (PEIS) released in 2005. Our review focused on
the impacts on mountaintop removal coal mining, which as its name suggests, involves removal
of all or some portion of the top of a mountain or ridge to expose and mine one or more coal
seams. The excess overburden is disposed of in constructed fills in small valleys or hollows
adjacent to the mining site.
Evidence shows that concentrations of chemical ions are, on average, about 10 times
higher downstream of MTM-VF than in streams in unmined watersheds. Sulfate (SO4 ~),
bicarbonate (HCO3 ), calcium (Ca2+) and Magnesium (Mg2+) are the dominant ions in the
mixture, but potassium (K+), sodium (Na+), and chloride (CP) are also elevated. These ions all
contribute to the elevated levels of total dissolved solids (TDS), typically measured as specific
conductivity, observed in the effluent waters below valley fills. These ions do not degrade or
precipitate out of the water column. Concentrations decrease only when diluted by another,
cleaner, source of water.
Water from sites having high chemical ion concentrations downstream of MTM-VF is
acutely lethal to invertebrates in standard aquatic laboratory tests, and models of ion toxicity
based on laboratory results predict that acute toxicity would be expected from the ions alone.
Benthic macroinvertebrate assessments of condition frequently score "poor quality" at sites
downstream of MTM-VF that have high ion concentrations.
Selenium concentrations are also elevated downstream of MTM-VF. Selenium can
bioaccumulate through aquatic food webs, and elevated levels have been found in fish in this
mining region. More than half of the sites surveyed downstream of MTM-VF exceeded the
chronic Ambient Water Quality Criterion (AWQC) for selenium. Selenium has been associated
with increased death and deformities in fish and reduced hatching in birds in studies of coal
overburden effluents in other regions.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 1 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Permits already approved from 1992 through 2002 are projected, when fully
implemented, to result in the loss of 1,944 km of headwater streams. This represents a loss of
almost two percent of the stream miles in the focal area (KY, TN, WV, and VA), a length that is
more than triple the length of the Potomac River, just during this 10-year-period. We found no
studies that updated the MTM-VF inventory conducted as part of the PEIS in 2002, but both
mine footprint and stream losses were projected to double by 2012. An updated inventory would
allow statistically sound estimates of cumulative stream loss and is a critical information need.
Reclamation practices (e.g., contouring and revegetation) were common in all of the
reviewed studies. The data indicate that reclamation partially controls the amount of soil erosion
and fine sediments transported and deposited downstream. The acidic drainage that is often
associated with coal mining is largely neutralized through reactions with carbonate minerals
within the valley fills or treatment in the sediment retention ponds. Yet, because ions, metals,
and selenium below MTM-VF were elevated in the reviewed studies, we conclude that current
management efforts do not improve all aspects of water quality. Additionally, there is no
substantive evidence in the literature or PEIS that onsite mitigation by constructed channels or
wetlands has replaced or will replace the lost ecosystem functions and biodiversity.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 2 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
2. INTRODUCTION
The purpose of this report is to assess the state of the science on the environmental
impacts of MTM-VF on streams in the Central Appalachian Coalfields.1 The coalfields cover
about 48,000 square kilometers (12 million acres) in West Virginia, Kentucky, Virginia and
Tennessee, USA (see Figure 1) (U.S. EPA, 2003, 2005).
The Central Appalachian Coalfields have a long history of mining. Current mining
methods, including MTM-VF, employ methods to control the acid mine drainages (AMD) that
have been a historic and continuing source of water quality degradation. The purpose of this
report is to evaluate evidence of the impacts of MTM-VF on headwater and downstream systems
despite improvements in acidic discharges. It is prompted by EPA's re-examination of how best
to implement environmental laws, especially the Clean Water Act (CWA), that are relevant to
surface mining (see Section 2.2).
We evaluated six potential consequences of MTM-VF:
• Loss of headwater and forest resources (see Section 3)
• Impacts on water quality (see Section 4)
• Impacts from aquatic toxicity (see Section 5)
• Impacts on aquatic ecosystems (see Section 6)
• The cumulative impacts of multiple mining operations
(see subsections of Sections 3, 4, and 6)
• Effectiveness of mining reclamation and mitigation (see Section 7)
We did not evaluate the impacts of MTM-VF on cultural or aesthetic resources.
We used two sources of information for our evaluation: (1) the peer-reviewed, published
literature and (2) the PEIS and its associated appendices (U.S. EPA, 2003, 2005). Only a few
peer-reviewed papers have studied water quality or stream ecosystems in headwaters directly
affected by or downstream of MTM-VF in the Central Appalachian Coalfields (Appendix A).
This report draws from these papers and from the relevant research findings of laboratory studies
and observational studies from other locations and mining activities. We also discuss the
findings published in the PEIS, which was published as two separate documents; the Draft,
published in 2003, and the Final, published in 2005. The final PEIS included responses to
comments on the draft and newer research results but did not include a revision of the original
'The derivation of the study boundary is described further in Chapter 4 of the PEIS (U.S. EPA, 2003, 2005).
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 3 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
material. When citing results from the many appendices of the PEIS, we specified the source to
make it easier for readers to find the original material. Finally, authoritative text books were
used as a source of background information and general scientific knowledge.
2.1. OPERATIONS USED IN MTM-VF
Mountaintop removal mining, like other surface mining practices, removes the soil and
rock over a coal seam (i.e., the overburden) to expose the coal. It is distinct from other types of
surface coal mining (e.g., contour mining) in its scale. This overview of the processes used in
MTM-VF summarizes the description in the PEIS (U.S. EPA, 2003, 2005). The mountain or
ridge top is prepared for mining by building access roads, clearing all trees and stockpiling
topsoil for future use in reclamation. Then, explosives are used to blast the entire top of the
mountain or ridge to expose and mine one or more coal seams (see Figure 2). As much as
300 vertical meters (1,000 feet) of overburden are removed.
The overburden removed during mountaintop mining cannot all safely be put back into
place because of the overall volume of the material and because the volume increases when the
rock is broken up. Some of the overburden is used to recontour the mine surface. The excess
overburden is disposed of in constructed fills in valleys or hollows adjacent to the mined site.
These fills bury the intermittent streams, springs and small perennial streams that comprise the
headwaters of rivers.
Both water flow and sediment discharges are altered by MTM-VF (see Figure 3). The
heavy equipment used to mine and move the overburden compacts the bare soils, forming a
large, relatively impervious surface that increases surface runoff. On the mined site, surface
runoff is diverted into ditches and sediment ponds, replacing natural subsurface flow paths.
Water flows out of the ditches through notches, or is directed toward the valley fill. Depending
on the construction and degree of compaction of the valley fill, the water then either percolates
through porous fill material or flows through ditches and coarser rock drains within, under, or
beside the fill. The effluent that emerges downstream of the ditches and below the downgradient
edge (i.e., the toe) of the valley fill is discharged into constructed channels and then to ponds that
are also used as treatment basins, for example, to settle solid particles, precipitate metals, or
regulate pH.
After the coal is removed, the extraction area is graded and planted to control sediment
runoff. The sediment retention pond may be eventually removed, and the stream channel is
recreated under the footprint of the pond.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 4 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
The coal is transported from the mine using trucks, conveyers or rail to a processing site,
where it is washed prior to transport to market. The impacts of coal processing, slurry ponds and
transport are not discussed in this report.
Mines can be as large as some cities (see Figure 4) and may use several different types of
mining, including underground methods such as room and pillar or long-wall mining and surface
methods such as contour, area and high-wall mining, in addition to mountaintop removal.
Though these other forms of mining can also produce small fills, valley fills resulting from
mountaintop removal are by far the largest. The active life of a mine increases with size; larger
mines can be active between 10 and 15 years.
The density of all coal mining activity (surface and underground) can be quite high in
some parts of the region (see Figure 5). Current statistics on the spatial extent of MTM-VF are
unavailable. As of 2002, the footprint of surface mine permits was estimated at 1,634 km2
(U.S. EPA, 2002) or about 3.3% of the land cover in the Central Appalachian Coalfields. As of
2001, permits for 6,697 valley fills were approved. Between removal with the mountain or
burial under fill, over 1,900 km of stream were scheduled to be lost through these existing
permits (U.S. EPA, 2002). The streams lost represent 2% of the streams in the study area, a
length that is more than triple the length of the Potomac River. More current statistics were
unavailable at the time this report was written, but both mine footprints and stream losses were
projected to double by 2012 (U.S. EPA, 2002).
2.2. REGULATORY CONTEXT
MTM-VF are permitted by state and federal surface mining and environmental protection
authorities. Individual mines are regulated under the Surface Mining Control and Reclamation
Act (SMCRA) by the Office of Surface Mining (OSM) and by delegated States under OSM
oversight. In addition, several specific sections of the CWA apply. These are implemented by
the EPA, the U.S. Army Corps of Engineers (USACE) and individual states authorized to
implement portions of the CWA. Although a complete listing and interpretation of the
regulations that affect MTM-VF operations are beyond the scope of this paper, Appendix B
provides a brief discussion of how water quality standards are implemented through the CWA in
the context of MTM-VF.
Two CWA permits are relevant to MTM-VF. The USACE issues a permit pursuant to
Section 404 of the CWA (33 U.S.C. § 1344) for the discharge of dredged and/or fill material.
This permit includes the valley fill itself and the fill necessary to create a sediment pond below
the valley fill. The second permit is issued by either the EPA or an authorized state pursuant to
Section 402 of the CWA (33 U.S.C. § 1342). The Section 402 program is also known as the
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 5 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
National Pollutant Discharge Elimination System (NPDES). The NPDES permit includes the
discharge from the sediment pond and any storm water associated with the mining activity.
Both permitting programs prohibit activities or discharges that cause or contribute to
violations of numeric or narrative state water quality criteria. While numeric criteria protect a
water body from the effects of specific chemicals, narrative criteria protect a water body from the
effects of pollutants that are not easily measured, or for pollutants that do not yet have numeric
criteria, such as chemical mixtures, or suspended and bedded sediments. Examples of narrative
standards that are particularly relevant to evaluating MTM-VF impacts include
• From West Virginia: No significant adverse impact to the chemical, physical, hydraulic,
or biological components of aquatic ecosystems shall be allowed (WV § 47-2-3).
• From Kentucky: Total dissolved solids or conductivity shall not be changed to the extent
that the indigenous aquatic community is adversely affected (401 KAR 10:031,
Section 4(f)).
"Adversely affect" or "adversely change" means to alter or change the community
structure or function, to reduce the number or proportion of sensitive species, or
to increase the number or proportion of pollution tolerant aquatic species so that
aquatic life use support or aquatic habitat is impaired (401 KAR 10:001,
Section 1(5)).
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 6 DRAFT—DO NOT CITE OR QUOTE
-------
wv
KY
TN
MILES
Central Appalachian Coalfields
State Boundaries
County Boundaries
2
3 Figure 1. The Central Appalachian coalfields.
4
5 Source: EPA (U.S. EPA, 2003, 2005).
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 7 DRAFT—DO NOT CITE OR QUOTE
-------
Sediment Pond
Former
Mountain Contour
Coal Seam
u V ¦
:: : ¦
Valley
Fill
Figure 2. A watershed view of a mountaintop mine and valley fill (no
consistent scale).
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 8 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
A.
B.
Secondary
Porosity
I I I
Primary
Porosity
i - i /°
l ,1
y
i i i
v yj,
Onnrifii /a tnwn
Tnlorant taYa
Water Table
Soil Horizon (organic/mineral)
Coal Layer
Overburden
Figure 3. Small stream watershed before and after mountaintop mining and
creation of a valley fill (simplified view, scales differ between upper and
lower halves of diagram). Photographs of macroinvertebrates by Greg Pond.
A. Before mining. The figure on the left side of the diagram illustrates the natural topography, geologic
strata and soil layers associated with small mountain streams in eastern coalfields. Stream valleys (natural
depressions in the landscape that conduct channelized streamflow) are the most obvious topographic feature
of the watershed. However, most of the water in small watersheds flows underground though a complex
system of aquifers (a), soil layer interflows (b) and slow moving trickles through minute stress fractures in
geologic strata of the parent mountain (c). Overland flow and subsurface flows (indicated by arrows) form
channelized flows (d) that integrate features of the entire landscape, including riparian vegetation and
diverse, instream biological communities.
B. After mining. On the right side, the same watershed is shown after the mountain rock layers have been
removed, crushed and deposited in the stream valley. Flat surfaces of remaining rock layers are less
permeable, producing higher surface runoff into a flood control channel (e) and valley fill (f, height is
approximate). Infiltration though valley fills of water exposed to larger total surface area of porous
unweathered rock (g) produces higher channelized flows and higher concentrations of dissolved ions and
trace metals downstream, where biological communities shift towards tolerant taxa (h). Subsurface
flowpaths in the intact geologic strata vary, depending on the types of rock in them, but water tables may
'back up' against the valley fill as shown here (i), increasing baseflows and exposure to valley fill materials.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 9 DRAFT—DO NOT CITE OR QUOTE
-------
1
B
2
3
4 Figure 4. Satellite images of the 40-km2 Hobet 21 mine (Boone County, WV)
5 (Panel A), and the Washington DC area (Panel B), at the same scale.
6
7 Source: Google Maps (2009).
k»«A*h Of SfKVf-m JistvA r•
im> g'll'S" r~- ««
r/iw document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 10 DRAFT—DO NOT CITE OR QUOTE
-------
W«2«t VMtjliiiafcwIromtwtfal P< unction
T -Atki
V-^« \ * SMi
«»»« «1(415®. WW J
County Boundaries
Mining Perm* Bowndarses
I i ¦coMnty
Not Started
Inactive
Actio*
R«c lariat ion
<*«t«r treatment
Bond Forf«it
Unknown/Other
1
2 Figure 5. Permit boundaries for surface and underground mines in
3 southwestern West Virginia. The Hobet 21 is shown in middle left near Point a.
4
5 Source: WV DEP (2009). Colors modified to improve legibility.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 11 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
3. LOSS OF HEADWATER AND FOREST RESOURCES
Headwater streams dominate surface flows in the United States and comprise 70-80% of
the total stream miles in the eastern coal mining states (Leopold, 1964; U.S. EPA, 2003, 2005).
Headwater stream ecosystems occur on all mountains in the eastern coalfields and in all valleys
that receive the excess overburden from mountaintop mining. Impacts include the loss of
headwater streams and forests on the removed mountaintops; burial of streams in the actual
footprint of the valley fills; and fragmentation of adjacent forests.
3.1. ESTIMATING EXTENT OF HEADWATER ECOSYSTEM LOSS
Estimating the extent of headwater ecosystems lost from MTM-VF begins with defining
where streams begin. The term, headwaters, refers to the springs, seeps, creeks and seasonal and
temporary flows that collectively form the origins of large river networks (see Figure 6).
Headwater streams are formed by leakages that in turn form linkages. Headwaters form where
groundwater breaks through (leaks) to the surface. At the surface, erosional processes create
channels that form small (first order) streams that link groundwater, stormwater, upland and
riparian ecosystems (see Figure 6) (Paybins, 2003; Freeman et al., 2007; Nadeau and Rains,
2007).
Headwater streams are classified as perennial, intermittent, or ephemeral based on the
relative contributions of groundwater and stormwater inputs, the timing and duration of
channelized (surface) flow, the drainage area, the channel's morphology and the underlying rock
types (Hewlett, 1982). Perennial headwaters flow year-round and are predominantly
groundwater-fed; intermittent streams flow seasonally (winter-spring) when groundwater levels
are elevated; and ephemeral streams receive no groundwater input and flow only in response to
precipitation events (e.g., rainfall, snowmelt) (Johnson et al., 2009).
OSM inventoried valley fills in the Central Appalachian Coalfields to estimate the
number of stream miles lost to mountaintop mining and valley fills, based on a 0.12-km2
(30-acre) minimum watershed size. This study found that in the 17-year period from 1985 to
2001, approximately 1,165 km (724 miles) of headwater streams were permanently buried under
valley fills in West Virginia, Kentucky, Virginia, and Tennessee (U.S. EPA, 2003, 2005). In a
cumulative impact study, the EPA (U.S. EPA, 2002) reassessed the number of stream miles lost
by including streams that were lost to other mining activities (blasting, backfilling, etc.) in
addition to valley fill footprints. In the revised estimate, 1,944 km (1,208 miles) of streams were
approved to be lost due to valley fills and associated activities from 1992 to 2002 (U.S. EPA,
2003, 2005). This means that more than 2% of the total stream miles and 4% of first- and
This document is a draft for review purposes only and does not constitute Agency policy,
12/17/09 12 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
second-order stream miles in the PEIS study area were approved for permanent burial during this
10-year period.
Estimating headwater stream loss in terms of miles of stream impacted in watersheds
above a size threshold is a useful beginning but does not address the loss of other headwater
ecosystems. For example, the estimate does not include the springs, seeps and wet areas that
may occur in places other than the stream channel and in watersheds less than 0.12 km
(30 acres) in size. Headwater stream burial estimated as the watershed area above the toe of
valley fills on permits approved from 1985 through 2001 is shown in Table 1. The total area
impacted shown here does not include valley fill permits approved prior to 1985 or after 2001.
In a study of 36 first-order streams for which valley fill permits were pending or
approved, Paybins (2003) estimated that the median watershed area for intermittent flows was
0.1 km2 (14.5 acres) and the median watershed size for perennial flows was 0.2 km2 (40.8 acres).
The average area of a valley fill shown in Table 1 is 0.3 km2. These data suggest that
intermittent and perennial streams are being buried by valley fills. The relationship of permits to
valley fills is not one-to-one. Multiple permits may refer to the same valley fill, and large valley
fills may cover more than one headwater basin (Paybins, 2003; see Figure 7).
3.2. LOSS OF HEADWATER ECOSYSTEM BIOTA
The biodiversity of the Central Appalachians is of national and even global significance.
The Southern Appalachian and most of the Central Appalachian Mountains were a refuge for
organisms during the last glacial maximum, which ended 10,000 years ago. The area includes
one of the most prominent hot spots for high biodiversity measured as rarity-weighted species
richness identified by NatureServe (see Figure 8). For example, nearly 10% of global
salamander diversity is found within streams of the Southern Appalachian Mountains (Green and
Pauley, 1987).
Evidence relevant to evaluating the loss of headwater ecosystem biota comes from
surveys conducted as part of the PEIS and reports from headwaters in other temperate regions.
We assume that most of the organisms inhabiting these systems are eliminated when the
headwater is buried or blasted during the mining process.
Headwater habitats are spatially and temporally dynamic and support diverse biological
communities (Gomi et al., 2002; Meyer et al., 2007; Clarke et al., 2008). Small but
biologically significant—differences in light, hydrology, water chemistry, substrate, sediments,
food resources, gradient and precipitation across small streams within the same river network
offer a wide variety of habitats and niches for aquatic and semiaquatic plants, animals and
ecologically beneficial fungi and microbes (Meyer et al., 2007).
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 13 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
The effects of the loss of headwater biota on regional biodiversity would be expected to
be most severe for taxa that occur only in headwater ecosystems, including intermittent streams
(Morse et al., 1993; Morse et al., 1997; Hakala and Hartman, 2004). Intermittent streams can
support diverse and abundant invertebrate assemblages (Feminella, 1996; Williams, 1996; Stout
and Wallace, 2003). Some taxa inhabit only intermittent or only perennial streams, but many
inhabit both types of streams (Feminella, 1996; Stout and Wallace, 2003; Hakala and Hartman,
2004). Populations of these taxa may be sustained by immigrants from nearby headwaters:
intermittent streams can flow in all seasons in wet years; perennial streams may dry down
seasonally during periods of drought and organisms may move freely between them. Stout and
Wallace (2003) sampled 36 intermittent streams in West Virginia and Kentucky that were
scheduled for burial by MTM-VF and collected approximately 73 genera and 41 families of
aquatic invertebrates, many of which are found in perennial streams as well. In a stream with
only subsurface flows over 70% of its length in summer, Collins et al. (2007) found that
subsurface invertebrate community composition and water chemistry in the intermittent reaches
was comparable to those in reaches with perennial surface flows.
In studies of two Appalachian headwater streams, more than 30 species of diatoms and
more than 40 species of beneficial fungi were recorded (Gulis and Suberkropp, 2004;
Greenwood and Rosemond, 2005). Diatoms and fungi are important food sources for fish and
aquatic insects. In addition, fungi produce enzymes that are essential to the rapid decomposition
of organic matter (e.g., wood and leaf litter). The breakdown of plant matter by fungi and other
microbes make nutrients in difficult-to-digest vegetation accessible to fish and invertebrates
(Gulis et al., 2006).
Headwater streams also support diverse and abundant assemblages of amphibians.
Salamanders are the most common vertebrates in headwaters (Davie and Welsh, 2004). Many
stream salamanders require headwater seeps and intermittent streams in forested habitats to
maintain viable populations (Petranka, 1998; Davie and Welsh, 2004). High levels of genetic
diversity in geographically distinct lineages of the spring-endemic Brownback Salamander
(Eurycea aquatica, Plethodontidae) have recently been described in the Southern Appalachian
Mountains (Timpe et al., 2009). Among the Appalachian plethodontids, species vary in their
preferences for ephemeral, intermittent, or perennial headwaters to the extent that life stage and
taxonomic information could be used to estimate hydroperiod at the collection sites (Johnson
et al., 2009). Many amphibian species are most abundant in intermittent streams, perhaps
because periodic drying offers freedom from predatory fish (Davie and Welsh, 2004).
Some species of salamanders split their lives between forests and headwaters and depend
on a close connection in order to move between the two (Petranka, 1998). Cool, moist soils and
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 14 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
large woody debris in the forested riparian zones of small streams provide suitable habitat for
salamanders (Petranka, 1998). Forest clearing increases the dispersal distance between the two
ecosystems and is expected to decrease the abundance of salamanders in small streams that
remain at a site (Maggard and Kirk, 1998). Changes to the dendritic structure and terrestrial
connectivity of natural headwater streams also decrease the number of salamander species found
(Grant et al., 2009).
3.3. LOSS OF HEADWATER ECOSYSTEM FUNCTIONS
As with the loss of biota, we assume that all of the ecosystem functions performed by
headwaters are lost when the headwater stream is buried or removed. These functions are lost
not only to the headwater stream itself, but also to ecosystems downstream of the MTM-VF
(discussed in Sections 5 and 6).
Because they are small, the contributions of headwater streams to ecosystem function at
the watershed scale are often overlooked (Meyer and Wallace, 2001; Meyer et al., 2007).
Although it is well known that a stream's ecological integrity depends on the functioning of its
smallest tributaries, we do not know how to measure the incremental effects of stream loss on
downstream functions.
Nutrient uptake and transformation occurs more rapidly in headwaters, where
slower-moving waters have longer contact times with biologically and chemically reactive
benthic substrates and hyporheic zones2 of small, shallow channels (Alexander et al., 2000;
Bernhardt et al., 2005). Peterson et al. (2001) estimated that 50-60% of the inorganic nitrogen
entering a stream is retained or transformed in the headwaters, reducing downstream nutrient
loads by half. This estimate is likely conservative because denitrification, a process that is
known to occur in natural stream channels and riparian zones (Payne, 1981), eliminates nitrogen
as N2 gas and is not included in the estimate by Peterson et al. (2001). Riparian buffers have a
central role in nitrogen removal, which is affected not only by buffer width and riparian
vegetation, but also by soil type, subsurface hydrology, chemistry and interstitial biofilm
communities in the riparian-hyporheic zone (Pusch et al., 1998; Mayer et al., 2007).
In addition to purifying water of nutrients, natural headwaters detoxify water of other
contaminants including the metals copper (Cu), zinc (Zn), manganese (Mn), and iron (Fe)
(Schorer and Symader, 1998). In contrast, outflows from filled headwaters typically are net
exporters of toxicants to downstream segments (see Section 4). The loss of natural ecosystem
2Hyporheic zone: the subsurface ecotone below and adjacent to the stream channel, where surface water and ground
water mix and exchange solutes. Much of the streamflow and biogeochemical processing in streams occur
underground. The hyporheic zone also supports a rich variety of aquatic fauna (Boulton et al., 1998).
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 15 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
function and the export of toxicants act in combination to increase risk to water quality below
MTM-VF.
In their natural state, forested headwaters typically transport little sediment or large
woody debris by fluvial processes and act as sediment reservoirs for periods spanning decades to
centuries (Benda et al., 2005). Substrate and organic debris dams slow the flow of water through
headwaters, creating more contact time for processing organic matter, nutrients, and toxicants
and regulating runoff in normal rain events. Recent evidence indicates that the number and
distribution of small tributaries and the presence of forest cover, are the primary controls of
runoff in high gradient watersheds (McGlynn et al., 2003; McGuire et al., 2005; Laudon et al.,
2007).
Forested headwaters also receive and process large volumes of organic matter from
upland and riparian vegetation (Wipfli et al., 2007). This terrestrial subsidy supports the biomass
of animals, plants and fungi found in headwaters and downstream segments.
Headwaters and associated interstitial habitats provide refugia for macroinvertebrates
during floods or spates and speed the recovery of aquatic communities when flow conditions
improve (Angradi, 1997; Angradi et al., 2001). Headwaters also serve as nurseries and spawning
grounds for amphibians and fish, including the brook trout (Salvelinus fontinalis), the only trout
native to West Virginia. Brook trout live and spawn in headwaters and often are the only fish
present in Appalachian first-order and second-order streams (Hakala and Hartman, 2004). In a
study of one West Virginia watershed, Petty et al. (2005) estimated that >80% of all brook trout
spawning occurred in small streams (<3 km2), including headwaters draining areas less than
0.25 km2.
3.4. CUMULATIVE IMPACTS ON FOREST RESOURCES
The EPA's (2002) cumulative impact study evaluated ecological condition, biodiversity,
forest loss and forest fragmentation in the Central Appalachian Coalfields. The 48,562 km2 of
the study area are dominated by 92% forest cover and contains roughly 95,000 km of streams,
including 67,600 km (71%) of headwater streams. The total mining permit land area estimated
from mine permit geographic information system (GIS) layers obtained from OSM was
1,634 km2: 1,100 km2 in Kentucky, 365 km2 in West Virginia, 131 km2 in Virginia and 38 km2
in Tennessee.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 16 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
3.4.1. Quantity of Forest Habitat Lost to Mountaintop Removal and Mining
Infrastructure
Surface mining deforested 1,540 km2 (3.4%) of the study area during the 10 years
between 1992 and 2002. An estimated 5,700 km2 (11.5%) of the PEIS study area was projected
to be deforested by 2012, an area 1.4 times the size of the state of Rhode Island. This estimate
does not recognize any reforestation efforts following mining and logging. The estimated
cumulative habitat loss included a 3-fold increase in the area in former headwater stream
watersheds classified in land use/land cover databases as "surface mining/quarries/gravel pits"
(U.S. EPA, 2002).
3.4.2. Quality and Connectivity of Forest Habitat Lost
In its natural condition, the Appalachian landscape is dominated by interior forest. A
decrease in forest cover followed by conversion to grasslands or other land cover has the
potential to shift the fauna of the region from that found in intact, high elevation forests to one
dominated by grassland and edge dwelling species.
Wickham et al. (2007) found that the pattern of deforestation from MTM-VF is
destroying interior forests at a greater rate than would be expected from the overall rate of
deforestation. Because of fragmentation, the area of interior forest lost was 1.75-5 .0 times
greater than the direct forest lost between 1992 and 2001. An increase in habitat fragmentation
has the potential to isolate natural populations, reduce population sizes, reduce gene flow,
increase the risk of extirpation or extinction of rare species and increase the rate of invasion by
exotic species, especially plants (Harper et al., 2005; Ewers and Didham, 2006). Fragmentation
of the terrestrial environment due to mining, projected from land cover data in the West Virginia
Gap Analysis Program and the permit rates observed during the 10 years preceding the
publication of the PEIS, indicates
• a 40% increase in the number of isolated forest habitat fragments,
• a 41% decrease in the average size of habitat fragments from 24.64 to 14.3 acres and,
• a 2.7% increase in the amount of edge habitat, caused by fragmentation of interior forests
(U.S. EPA, 2002).
3.4.3. Riparian Habitat Lost
In the West Virginia portion of the study area, the projected loss of riparian habitat from
MTM-VF is 30.72 km2, 3.2% of the baseline. Approximately 42% of these projected losses
occur in headwater (first- and second-order) streams (U.S. EPA, 2002).
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 17 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 1. Watershed areas above the toe of valley fills approved from 1985 to
2 2001
3
Watershed area
Description
0.3 km2 (71 acres)
Average size of watershed above valley fill toe
15.3 km2 (3,774 acres)
Largest size watershed above valley fill toe
1,774.4 km2 (438,472 acres)
Total watershed area impacted by valley fill construction
5 Source: EPA (U.S. EPA, 2003, 2005), Section III.
formation
precipitation
evaporation
f jfc < infiltration
w<*Sjdeep percolation
percolation
groundwater
Figure 6. Headwater types: ephemeral, intermittent and perennial streams.
Hillslopes erode to form ephemeral stream channels (dotted lines) that flow into
intermittent (dashed lines) and perennial (solid line) streams.
Source: Jennings and Harmon (2002).
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 18 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
Key:
Stream
^ Mined area
Actual Valley Fill Area
Valley Fill Permit Area
Figure 7. Map showing loss of headwater streams to MTM-VF. This diagram
depicts the loss of stream miles and channel complexity that results from
extensive mountaintop mining and valley filling. Blue lines inside valley fill
areas represent buried streams. Note that the actual area of valley fill may exceed
the area permitted for fill and that stream loss based on permit area may
underestimate the full extent of buried streams.
Source: Modified from Shank (2004; Figure 12).
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 19 DRAFT—DO NOT CITE OR QUOTE
-------
3
cc
§-
C}
ss
2
On
£
"o>
>S
<5*
^3
SS
•a
o
On
TO
Zc
O
a
<>
£5
3
a.
§-
<*>
Cc
3
^ ¦ HKih Source NatureServe and its Natural Heritage
"**• iss "¦s member programs Jbly 2008
Ci Produced by National Geographic Maps and NatureServe
O December 2008
S
S- Figure 8. Hot spots of rarity-weighted species richness in the United States. The Central Appalachian Mountains,
?? including the Central Appalachian Coalfields, have been identified as one of the most significant hot spots for
^ biological diversity in the United States
s Source: NatureServe and its Natural Heritage member programs, July 2008 (National Geographic Maps and NatureServe,
^ 2008).
o
£
NatureServe
Rarity-VWeighted
Richness Index
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
4. IMPACTS ON WATER QUALITY
In this section, we report the results of a number of studies that have assessed the changes
in the physico-chemical attributes of streams downstream of MTM-VF. Although much of this
information may also apply to constructed channels and other water-containing structures on the
valley fills and the mined site, there are practically no data in the PEIS or the peer-reviewed
literature on these constructed systems. The physico-chemical attributes we review below
include alteration of stream flow, sedimentation of stream substrates, water chemistry, and
sediment chemistry. Alterations of these attributes are the potential causes of the effects
observed downstream of MTM-VF, which are described in Sections 5 and 6 of the report.
4.1. ALTERATION OF FLOW
Four factors may affect stream flow below valley fills. First, trees and other vegetation
are removed from both the mined area and the area of the valley fill, and trees are generally slow
to regrow on the mined area and valley fill. This reduces evapotranspiration rates from the
watershed because transpiration is a function of the active vegetation (Dickens et al., 1989;
Messinger, 2003). Second, the valley fill forms an unconsolidated aquifer in the watershed that
stores a portion of any water that infiltrates into it (Dickens et al., 1989; Wunsch et al., 1999).
This water comes from recharge along the periphery of the spoil body where surface-water
drainage may be caught, from groundwater intercepted from adjacent bedrock aquifers, or from
precipitation falling on the fill. Third, compaction of the fill surface by heavy equipment can
reduce infiltration of precipitation and increase overland runoff (Negley and Eshleman, 2006).
Fourth, when headwater streams are lost (see Section 3), attributes that influence surface flow
(e.g., woody debris, surface water/ground water connections) are also lost.
Valley fills may act like a headwater aquifer and provide a more constant source of flow
during the dry parts of the year. Comparing adjacent mined and unmined watersheds, monthly
mean unit flow was relatively similar between the mined and unmined watersheds when soil and
aquifer moisture levels were at their maximums in late winter and spring (February to May), but
monthly mean flow in the mined watershed was greater than that in the unmined watershed
during summer, autumn and early winter, when soil and aquifer moisture levels were reduced
(Messinger and Paybins, 2003). Wiley et al. (2001) found the 90% duration flows3 at sites
below valley fills were 6 to 7 times greater than the 90% duration flows found at unmined sites.
Moreover, daily streamflows from sites below valley fills were generally greater than those in
3The 90% duration flow is the stream flow (m3/sec) equaled or exceeded at a site 90% of the time, a measure of the
baseflow.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 21 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
unmined watersheds during periods of low streamflow (Wiley et al., 2001). Green et al. (2000)
observed that several of their unmined sites did not have surface flows during the summer and
fall of a year when a drought occurred, but the streams below valley fills continued to have
surface flows.
Storm intensity changes the relative effect of the valley fill on downstream flows.
Intense storms can produce greater stormflows in watersheds with MTM-VF compared to
unmined watersheds, but stormflows associated with precipitation from lower intensity storms
may be ameliorated by valley fills. Messinger and Paybins (2003) found that a mined watershed
had greater peak flows during severe storms than an unmined watershed. Unit peak flow4 was
greater in the mined watershed following summer thunderstorms when rainfall intensity
exceeded 2.5 cm/hour (Messinger, 2003). In contrast, unit peak flow was lower in the mined
watershed following low-intensity, long-duration rainfall events—particularly in the winter.
Wiley and Brogan (2003) found that peak discharges after an intense storm were greater
downstream of valley fills than in unmined watersheds. Peak discharges were estimated by
applying the slope-area method5 (Benson and Dalrymple, 1967) to measurements of high water
marks observed after flooding associated with a thunderstorm complex, which resulted in 7.6 to
15 cm of rainfall in southeastern West Virginia over a 5- to 6-hour period. Six sites were
studied; three below valley fills and three in unmined watersheds. At two of the three sites
downstream of valley fills, the estimated peak discharges were equivalent to floods that would
naturally occur only once every 50 to >100 years. Peak discharges at the sites in unmined
watersheds had less severe estimated flood recurrence intervals of 10 to 25 years (Wiley and
Brogan, 2003). Estimates for the third site downstream of a valley fill were more difficult to
interpret, as the peak discharge had an estimated flood recurrence interval of <2 years, much less
severe than the other two mined sites. The differences might be due to differences in rainfall
among the watersheds or differences in mine and valley fill attributes. Comparisons among sites
assume that rainfall was similar among the watersheds, but there were differences among the
sites from unmined watersheds. Moreover, thunderstorms can cause locally variable rainfall,
particularly in mountainous terrains (Barros and Lettenmaier, 1994; Roe, 2005). Also, the third
site was downstream from only a single reclaimed valley fill; there was no active surface mining
in the basin, and the valley fill was larger than those of the other two sites, which both had active
surface mining.
4Unit peak flow is discharge per unit area of watershed, m3/sec/km2.
5With the slope-area method, the maximum flood height is estimated from the physical evidence left by the flooding,
the high water marks. Then the cross-sectional area and wetted perimeter (i.e., the length of the part of the perimeter
of the channel cross-section [stream bed and banks] below the water surface) of the stream channel is measured at
that flood height. The slope of the stream bed is also measured, and Manning's n, an index of the roughness of the
stream bed, is estimated. The peak discharge is then calculated using these variables.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 22 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
4.2. CHANGES IN SEDIMENTATION
All valley fills are built with a sediment retention pond, which is intended to capture sand
and finer-sized particles that are produced by the fragmentation of the overburden and may be
washed downstream from the toe of the valley fill (U.S. EPA, 1979). Despite this, Wiley et al.
(2001), using a modified Wolman (1954) pebble count for the bankfull channel,6 found that the
percentage of particles less than 2 mm (i.e., sand and fines) was elevated in stream reaches
downstream from a valley fill (i.e., median = 60%, interquartile range = 56-65%) when
compared to unmined streams (i.e., median = 24%, interquartile range = 15-34%).
Similarly, Green et al. (2000), using methods from EPA's Environmental Monitoring and
Assessment Program for the wetted channel7 (Kaufmann and Robison, 1998), found that mean
substrate sizes were less in filled or filled/residential streams compared to unmined streams,
while mean percentage of sand and fines was greater. However, mean substrate sizes were
largest at sites described as being downstream of other types of mining without valley fills (i.e.,
generally older contour mines) (see Table 2).
Hartman et al. (2005) did not find any clear pattern of sediments in a study that compared
pairs of sites using samples taken in December with a scoop sample separated with modified
Wentworth sieves (McMahon et al., 1996) (see Table 3). In two cases, the proportions of sand
and fines were similar; in the fourth case, it was greater in the filled site; and in the third case, it
was greater in the reference site. However, there appears to have been a significant nonmining
disturbance in this last control site, Big Buck Fork.
4.3. CHANGES IN CHEMICAL TRANSPORT AND BASIC WATER QUALITY
PARAMETERS
4.3.1. pH, Matrix Ions and Metals
Almost invariably, coal mining exposes pyrite, a ferric sulfide mineral formed in
association with coal (Caruccio et al., 1977; Altschuler et al., 1983; Casagrande, 1987; Younger,
2004). In the presence of water and oxygen (O2), pyrite is oxidized (i.e., a reaction catalyzed by
autotrophic bacteria) to form the strong acids characteristic of acid mine drainage (Stumm and
Morgan, 1996):
Fe(2+)S2 + 3.75 02 + 3.5 H20 Fe(3+)(OH)3 + 2 SO42- + 4 H+ (Eq. 1)
6The bankfull channel is the entire channel, which is submerged at bankfull discharge—the point just before the
stream flow begins to spread out onto the stream's flood plain at high flows. As a result, this approach measures
some substrate that is dry during baseflow, which is when these channel characteristics are usually measured.
7The wetted channel is the portion of the channel that was submerged at the time these channel characteristics were
measured.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 23 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
However, in the presence of sufficient carbonate minerals, such as calcite (CaC03) and
dolomite [CaMg(C03)], the acidity can be neutralized (Rose and Cravotta, 1998):
2 CaC03 + 2 H —> 2 Ca2+ + 2 HCOf (Eq. 2)
2 CaMgC03 + 2 H+ -> Ca2+ + Mg2+ + 2 HC03~ (Eq. 3)
The effluent waters from valley fills are generally not acidic and may be somewhat
alkaline (Bryant et al., 2002; Merricks et al., 2007). The pH is generally 7.0 or greater (Bryant
et al., 2002; see Tables 4-3, 4-4, 4-5). The alkaline pH has been attributed to exposure of the
water to carbonate minerals within the valley fill that originate from fragmentation of the
noncoal formations that form the overburden or are added during construction of the valley fill
(Sobek et al, 1978; Banks, 1997; Skousen, 1997). Other methods that may moderate pH include
physically isolating the pyritic materials within the mine or valley fill (Skousen et al., 2000) and
treatment within the sediment retention pond.
Iron forms relatively insoluble compounds, such as Fe(OH)3, under more alkaline
conditions and may not be found in elevated concentrations in the effluent waters below valley
fills (Bryant et al., 2002; see Table 4-3). However in some conditions, such as during higher
flows, Fe can remain elevated (Hartman et al., 2005; see Table 4-5).
Most other metals, such as cadmium (Cd), chromium (Cr), Cu, lead (Pb), and Zn,
coprecipitate with or sorb to the iron compounds (Kimball et al., 1995; Lee, 2002; Larsen and
Mann, 2005) and were not found (in one study) at elevated concentrations in the effluent waters
(Bryant et al., 2002; see Table 4). Exceptions to this are Mn and nickel (Ni), which may be
elevated in the effluent waters below valley fills (Bryant et al., 2002; Hartman et al., 2005; see
Tables 4 and 6). Mn may occur in association with siderite (FeC03) in shales within the
overburden and is more soluble in the more alkaline waters (Larsen and Mann, 2005).
Aluminum (Al) is found primarily associated with clay minerals in soils and is not soluble unless
the pH is less than 4.9 (Nordstrom and Ball, 1986).
Sulfate (S042-), calcium (Ca2+ from calcite-type minerals), magnesium (Mg2+ from
dolomite-type minerals) and bicarbonate (HC03~ from both calcite and dolomite), which are
formed in the above reactions (see Eq. 1-3), are commonly present at elevated concentrations in
the effluent waters (Bryant et al., 2002; Hartman et al., 2005). In addition, other water-soluble
compounds within coal or overburden may be solubilized by the above reactions or just by the
increased exposure to water in the fragmented overburden (Yudovich and Ketris, 2006; Vesper
et al., 2008). These ions, including Se, K+, Na+, and Cf may occur at elevated concentrations in
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 24 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
the effluent waters (see Tables 4 and 6). All these ions are components of the elevated specific
conductivity, a measure of the stream's ability to conduct an electrical current, which reflects the
concentration of dissolved ions in the water (measured in units of microSiemens per cm, (iS/cm),
and TDS observed in these waters (Green et al., 2000; Howard et al., 2001; Bryant et al., 2002;
Bodkin, 2007; Merricks et al., 2007; Pond et al., 2008; see Tables 4-7). Hardness is another
measure of these dissolved ions—particularly the divalent ones like Ca2+ and Mg2+.
Most studies have not assessed the seasonal variability of water chemistry at these sites,
but Green et al. (2000) present seasonal data for five consecutive seasons from 1999 to 2000.
There appears to be little seasonal pattern to pH, but mean conductivities were greatest in all four
watershed types during the summer sampling period, possibly because of seasonally reduced
discharges (see Table 8). In particular, mean conductivity exceeded 1,000 |iS/cm in streams in
filled and filled/residential watersheds during the summer sampling period. In all seasons,
conductivities at reference sites in unmined watersheds were an order of magnitude (10 times)
lower than at sites in filled and filled/residential watersheds (see Table 8). Pond et al. (2008)
observed conductivities up to 2,540 |iS/cm in streams from mined watersheds.
4.3.2. Water Temperature
Valley fills reduce the annual variation in water temperature. Comparing mean daily
water temperatures between an unnamed tributary of Ballard Fork near Mud, West Virginia; a
stream downstream from a valley fill; and a reference site, Spring Branch near Mud, West
Virginia; Wiley et al. (2001) found that mean stream temperatures were warmer downstream of
the valley fill during the autumn, winter and spring, with the greatest difference being in
February. In the summer, the mean stream temperatures downstream from the valley fill were
cooler than those in the reference site. Moreover, the range of variation both annually and within
different seasons was less downstream from the valley fill. The minimum and maximum
temperatures downstream of the valley fill were 3.3°C and 16.5°C, respectively, while those in
the reference stream were below 0°C and 20.0°C.
4.3.3. Nutrients
Bryant et al. (2002) found generally low median concentrations of nitrate (NO3) plus
nitrogen dioxide (NO2) in streams from unmined watersheds and below valley fills, with some
samples having concentrations less than the detection limit of 0.10 mg/L. However, the mean
concentration of N03 plus NO2 was slightly greater in the streams below valley fills (Pond et al.,
2008), and a maximum concentration of 17 mg/L was observed. Bryant et al. (2002) speculated
that this could be caused by use of nitrogen-containing explosives at these sites or by spreading
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 25 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
nitrogen containing fertilizers during reclamation. Phosphorus (P) was not detected in any
samples with a detection limit of 0.10 mg/L (Pond et al., 2008).
4.3.4. Dissolved Oxygen
In the studies that have measured dissolved oxygen (DO), concentrations in unmined
streams and streams in either mined and valley fill streams have been reasonably high and
similar among the different types of watersheds (see Tables 5 and 6; Green et al., 2000; Howard
et al., 2001; Bryant et al., 2002; Hartman et al., 2005). Published concentrations range from 6.5
to 13.0 mg/L. However, no studies have looked at diurnal variation of dissolved oxygen in these
streams.
4.4. CHANGES IN SEDIMENT CHEMISTRY
Data on sediment chemistry downstream of valley fills are limited to a study by Merricks
et al. (2007). They sampled three to six stations at 100- to 150-m intervals in each of three
streams downstream from sedimentation ponds below valley fills in West Virginia and a single
reference site (see Table 9).
Sediment concentrations were generally greater at one stream, Lavender Fork, which was
downstream from a reclaimed, 6-year old valley fill and that also had the greatest measured
stream water conductivities. Sediment concentrations also generally decreased with increasing
distance below the sedimentation ponds.
4.5. CUMULATIVE IMPACTS
In terms of downstream water chemistry, the primary cumulative impact of MTM-VF or
at least the mining of coal by different methods in the region affected by MTM-VF has been
elevated concentrations of S042- and conductivity. In larger streams of the Kanawha basin,
Paybins et al. (2000) found that one-fourth of all water samples exceeded a S042- concentration
of 250 mg/L and 70% of the water samples collected downstream of coal mines exceeded a
regional background concentration of 21 mg/L that was calculated from data for basins with no
history of coal mining. Moreover, the median concentration of S042' had increased by 1.6 times
in these streams between 1980 and 1998, and conductivity had increased by 1.2 times (Paybins
et al., 2000). S042~ and some of the other ions contributing to conductivity are conservative ions
in water, meaning that there are no chemical or biological processes removing ions from or
adding ions to the waters. Therefore, any changes in S042" concentrations are the outcome of
mixing of waters with differing S042- concentrations (Cooper, 2000). Therefore, the increased
S042- and conductivity are associated with increased sources of water with elevated S042~ and
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 26 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
conductivity within the Kanawha basin. Because other land disturbances, such as residential
development, are not origins of elevated S042~ and conductivity, MTM-VF appear to be these
sources.
Conversely, while total Fe, total Mn and total A1 in many larger streams within mined
basins exceeded regional background concentrations of 129 (ig/L, 81 (ig/L, and 23 jag/L,
respectively, the median concentrations of total Fe and total Mn had decreased between 1980 and
1998 by approximately one-third and one-half, respectively, and pH had increased (Paybins
et al., 2000). As discussed previously, these metals are not as soluble under more alkaline
conditions, and their decrease may reflect the increase in pH associated with the increased
sources of alkaline water within the Kanawha basin, the valley fills.
In the absence of other direct evidence on the cumulative effects of the changes in water
chemistry associated with MTM-VF on downstream water quality, it should be noted that
headwater streams, such as those affected by MTM-VF, have a large influence on downstream
water quality. Alexander et al. (2007) found that lst-order, headwater streams contributed 70%
of the mean annual water volume in 2nd-order streams and 55% of the volume in higher-order
rivers. For nitrogen, a nutrient that is not conservative like the ions associated with MTM-VF,
these ls,-order streams contributed 65% of the flux in 2nd-order streams and 40% of the flux in
higher-order rivers (Alexander et al., 2007).
In terms of sediment contaminants, Paybins et al. (2000) found significant concentrations
of polycyclic aromatic hydrocarbons (PAH) at several stations within the Kanawha River basin
(see Table 10). However, most of these PAHs appear to be constituents of particles of coal that
occur in sediments because of the extensive coal mining and transport of coal in the region.
Downing-Kunz et al. (2005) found sediment concentrations of coal ranging from 1 to 53 g/kg in
streams draining more southern parts of the Central Appalachian Coalfields in Kentucky. PAHs
are a natural component of coal (Chapman et al., 1996; Paybins et al., 2000), but these PAHs are
unlikely to be bioavailable to benthic invertebrates or fish (Carlson et al., 1979; Ahrens and
Morrissey, 2005; Yang et al., 2008). Arsenic (As) and metals were also detected in sediments
(see Table 10) of the Kanawha River. However, the source of these sediment contaminants is
less clear.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 27 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 2. Substrate measures
2
Substrate measure:
mean (standard deviation)
Unmined
(n = 9)
Filled
(n = 15)
Filled/
residential
(« = 6)
Other
mined
(« = 4)
Mean substrate size class (unitless)
3.7 (0.3)
3.5 (0.5)
3.6 (0.8)
4.0 (0.3)
Calculated mean substrate size
(diameter in mm)
53
38
42
109
% <2 mm diameter (sand & fines)
16.9(9.9)
20.7(12.9)
29.7(24.1)
8.0(9.2)
4 Source: Green et al. (2000).
5
6
7 Table 3. Proportion of sediments that were sand and fines (mean [standard
8 error]) in paired sites
9
Site names (reference/impaired)
Reference
Filled
W. Br. Atkins Creek/E. Br. Atkins Creek
0.35 (0.00)
0.46 (0.10)
1 Big Buck Fork/Hill Fork
0.78 (0.03)
0.50 (0.06)
| Bend Branch/Rockhouse Creek
0.25 (0.07)
0.23 (0.02)
N. Br. Sugar Tree Creek/S. Br. Sugar Tree Creek
0.27 (0.02)
0.50 (0.04)
11 Source: Hartman et al. (2005).
12
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 28 DRAFT—DO NOT CITE OR QUOTE
-------
to
Table 4. Water quality variables in unmined streams versus streams below valley fills
Unmined
Filled
Detection
Variable
Median
Mean
Range
Median
Mean
Range
limit
S04*
12.6
16.0
11.0-21.6
523
696
155-1,520
5.0
Ca, total*
4.88
7.50
2.70-12.0
104
138
38.0-269
0.10
Mg, total*
4.10
4.30
2.30-7.00
86.7
122
28.0-248
0.50
Hardness*
29.1
42.0
17.0-72.0
617
801
225-1,620
3.3
TDS*
50.5
...
—
847
—
...
5.0
Mn, total
<0.005
0.034
<0.005 0.083
0.044
0.14
0.009-9.0
0.010 1
Conductivity ((iS/cm)*
66.4
62.0
34.0-133
585
1,020
159-2,540
...
HCO3*
—
20.9
6.10-35.0
—
183
10.7-502
NA 1
Se, total*
<0.0015
<0.0015
<0.0015
0.012
0.011
<0.0015-0.037
0.003
Alkalinity
20.0
—
—
150
—
...
5.0
K, total*
1.58
1.60
1.30-2.00
8.07
9.90
3.00-19.0
0.75 |
Na, total*
1.43
2.40
0.70-5.50
4.46
12.6
2.60-39.0
0.50
Mn, dissolved
<0.005
0.021
<0.005-0.055
0.044
0.11
0.0065-0.85
0.01
CI*
<2.5
2.8
<2.5^4.0
4.5
4.6
<2.5-11
5.0
Acidity
2.5
—
—
4.2
...
—
2.0
Ni, total
—
<0.010
<0.010
—
0.014
<0.010-0.059
0.02 |
no3/no2*
0.81
0.40
<0.10-0.90
0 95
3.4
0.80-17
0.10
pH (standard)*
6.8
7.1
6.1-8.3
7.8
7.9
6.3-8.9
...
Acidity, hot
<2.5
—
—
<2.5
—
...
5.0
Al, dissolved
<0.050
0.093
<0.050-0.19
<0.050
0.096
<0.050-0.27
0.10 j
Sb, total
<0.0025
—
—
<0.0025
—
...
0.005
As, total
<0.001
—
—
<0.001
...
—
0.002
Be, total
<0.0005
—
—
<0.0005
—
...
0.001
Cd, total
<0.0005
—
—
<0.0005
—
—
0.001
-------
to
5 a
s*
§•
K
TO
3
*
TO
<
TO"
•a
c
•3
§
UJ TO
o ^
o
s
<5-
S5
rL
§¦
Cc
3
O
Table 4. Water quality variables in unmined streams versus streams below valley fills (continued)
Variable
Cr, total
Unmined
Median
<0.0025
Mean
Range
Filled
Median
<0.0025
Mean
Range
Detection I
limit j
0.005
Co, total
Cu, total
Pb, total
Hg, total
<0.0025
<0.0025
0.005
<0.0025
0.0029
<0.0025-0.005
<0.0025
0.0026
<0.0025-0.0034
0.005
<0.001
0.0012
<0.0010-0.0021
<0.001
0.0012
<0.0010-0.0040
0.002
<0.0001
<0.0001
0.0002
Total organic carbon
P, total
1.4
1.4
1.0
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
0.10
Au, total
<0.005
<0.005
0.01
Th, total
<0.001
<0.001
0.002
V, total
<0.005
<0.005
0.01
Ba, total
Dissolved oxygen
0.029
0.040
0.015-0.072
0.025
0.041
0.022-0.068
0.020
13.6
11.0
Dissolved organic carbon
Total suspended solids
2.45
1.95
1.00
5.75
4.25
5.00
5
H
Fe, total
0.42
0.18
0.065-0.47
0.19
0.28
0.066-0.65
0.10
8 8
2*
£ 5
o s.
H TO
rag
O Sr
H
m
Fe, dissolved
Zn, total
Al, total
0.22
0.074
<0.050-0.19
0.10
0.092
<0.050-0.28
0.10
0.0060
0.010
0.0033-0.023
<0.0025
0.0091
<0.0025-0.027
0.005
0.15
<0.10
0.10
Units are mg/L, unless indicated otherwise. The table shows median, mean and range of concentrations for various water quality variables in unmined streams
versus streams below valley fills from a data set reported on by Bryant et al. (2002) (median, 9 unmined sites and 21 filled sites, each sampled about six times
from August 2000 to February 2001) and Pond et al. (2008) (mean and range for a subset that also had biological data, 7 unmined sites and 13 filled sites, except
for pH and conductivity, which were measured at 10 unmined sites and 27 filled sites). If a concentration was less than the detection limit, the value is shown as
-------
Table 5. Water quality parameters for unmined or reference streams or streams downstream from mined,
filled, or filled and residential watersheds in West Virginia
Variable
Green et al. (2000)
Merricks et al. (2007)
Hartman et al. (2005)
Unmined
58-140
59(38-178)
Filled
643-1,232
850(159-2,500)
Filled/
residential
538-1,124
843 (155-1,532)
Mined
172-385
187 (90-618)
Reference
Filled
Reference
Filled
Conductivity
(fiS/cm)
247 ±87
923 ± 380-
2,720 ± 929
47.6 ± 2.4-
259.7 ±30.6
502.0 ± 98.4-
1,479.0 ± 110.6
pH (standard)
7.1-7.5
7.5 (5.7-9.4)
7.1-7.9
7.7 (5.9-8.5)
7.1-8.3
8.0 (6.4-8.7)
6.7-8.4
7.4 (6.0-8.7)
7.2 ± 0.36
7.93 ± 0.18-
8.37 ±0.47
6.5 ± 0.6-
7.0 ± 0.4
7.2 ± 0.6-
7.5 ± 1.0
Dissolved 02
(mg/L)
6.5-13.3
10.9 (5.6-15.2)
7.5-13.0
10.0 (5.8-14.5)
8.5-14.0
9.4 (7.3-16.1)
8.7-12.7
10.2 (7.4-14.5)
—
—
8.5 ± 0.8-
13.4 ±0.4
9.1 ± 1.0-
13.0 ±0.6
Hardness
(mg/L)
—
—
—
—
86 ±20
544 ± 226-
1,904 ±596
—
—
Sources: Green et al. (2000) (range of means among seasons, overall mean, overall range), Merricks et al. (2007) (range of means and standard deviations) and
Hartman et al. (2005) (range of means and standard deviations).
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Table 6. Alkalinity, pH and metals in control streams and streams
downstream from filled watersheds in West Virginia
Parameter
Reference
Filled
Mean
Range
Mean
Range
Alkalinity*
12.8
0.400-46.8
163
16.2-319
pH (standard)
7.2
6.7-7.7
7.7
6.9-8.2
Na*
2.9
0.80-3.1
10
3.9-22
K*
3.3
1.5-5.1
10
1.8-14
Mg*
23
2.2-52
86
4.9-130
Ca*
37
2.6-67
130
5.9-200
Cu*
0.00080
0.00020-0.0019
0.0012
0.00050-0.0018
Ni*
0.0076
<0.00030-0.018
0.025
<0.00030-0.051
Zn
0.0027
0.0014-0.0047
0.0028
0.00090-0.0086
Mn*
0.019
0.0016-0.046
0.062
0.0020-0.17
Al
0.012
0.0090-0.019
0.019
0.00090-0.064
Fe*
0.016
0.0014-0.030
0.047
<0.00050-0.082
Units are mg/L, unless indicated otherwise. If the concentration was less than the detection limit, the value is shown
as < the detection limit. An asterisk marks those measures where the fill streams were statistically significantly
greater (p < 0.05) than the reference streams.
Source: Hartman et al. (2005).
Table 7. Dissolved oxygen, pH and conductivity for sites in eastern Kentucky
[Parameter
Reference (n = 4)
Filled (n = 8)
[Dissolved oxygen (mg/L)
9.1-9.6
8.4-9.7
pH (standard)
7.1-7.4
7.2-8.2
Conductivity (|4,S/cm)
30-66
420-1,690
Values are the range.
Source: Howard et al. (2001).
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 32 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 8. Seasonal mean (standard deviation) of conductivity (^iS/cm) for the
2 four classes of streams
3
Season
Unmined
Filled
Filled/residential
Mined
Spring 1999
64 (19) h = 9
946(614)« =15
652 (237) n = 6
172 (90) n = 4
Summer 1999
140 (54) n-2
1,232 (643) n= 15
1,124 (282) n = 6
385 (202) n = 3
Autumn 1999
91 (59)« = 2
958 (430) n = 14
984 (221)/i = 6
260 n - 1
Winter 2000
73 (29) n = 9
836 (425) n = 14
844 (173) n = 6
254 (171) n = 3
Spring 2000
58 (28) n= 10
643 (382) n= 15
438 (249) n = 6
192 (155) n = 5
4
5 The number of sites (n) analyzed is also given.
6
7 Source: Green et al. (2000).
8
9
10 Table 9. Range of sediment concentrations of metals and arsenic (mg/kg) in
11 streams downstream from the sedimentation ponds below valley fills in 2002
12 and 2004 and from a reference site in 2002
13
Metal or arsenic
Reference—2002
(n = l)
Downstream from
valley fill—2002
(" = H)
Downstream from
valley fill—2004
0n = 18)
Al
11
9-20
2-28
As
—
—
0.015-0.070
Cd
—
—
0.005-0.015
Cu
0.018
0.012-0.122
—
Fe
51
49-158
10-151
Hg
—
—
0.006-0.015
Mn
1.4
1.6-17
1.0-41
Se
—
—
0.001-0.011
Zn
—
—
0.1-2.5
14
15 The reference site was only sampled in 2002, and the analytes measured differed between the 2 years. The
16 unmeasured analytes are indicated by —.
17 Hg = mercury.
18
19 Source: Merricks et al. (2007).
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 3 3 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 10. Poly cyclic aromatic hydrocarbons, arsenic and metals detected in
2 sediments of larger streams in the Kanawha Basin
3
Chemical and units of
concentration
Number of detects/number
of samples
Range of detections
benz[a]anthracene (fxg/kg)
12/13
5-800
dibenz[a,h]anthracene (jag /kg)
4/13
40-200
2,6-dimethylnaphthalene (ng/kg)
10/13
50-500
fluoranthene (jj.g/kg)
13/13
30-1,100
fluorene (|xg/kg)
7/13
60-300
naphthalene (ng/kg)
9/13
3-700
phenanthrene (jag/kg)
13/13
9-900
As (mg/kg)
13/13
4-20
Cr (mg/kg)
13/13
60-110
Pb (mg/kg)
13/13
20-50
Ni (mg/kg)
13/13
50-100
Zn (mg/kg)
13/13
200-600
Source: Paybins et al. (2000).
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 34 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
5. TOXICITY TESTS
In this section, we report on results of toxicity tests relevant to evaluating water quality
downstream of MTM-VF. Toxicity tests expose organisms under laboratory conditions to
ambient media (i.e., water or sediment samples), whole effluents, reconstituted effluents, or
specific effluent constituents. Toxicity tests are valued because they can reflect the mixture as a
whole, including antagonistic and synergistic effects. They also help distinguish the effects of
water quality from other stressors (e.g., habitat quality, flow regime changes, temperature).
Toxicity tests have been used as the basis for deriving water quality criteria and permitting
industrial and waste water effluents.
The most common standard toxicity tests used to evaluate the effects of effluents measure
the survival of the crustacean Ceriodaphnia dubia after 48 hours of exposure and the survival of
fathead minnows (Pimephales promelas) after 96 hours of exposure. Both of these tests have
limitations for evaluating MTM-VF effects: neither Ceriodaphnia dubia nor Pimephales
promelas are native to the streams of the study area, and the standard test durations are much
shorter than the exposures experienced by organisms downstream of MTM-VF operations.
There are likely more sensitive responses than death. In particular, because ions are so
influential in regulating membrane permeability during fertilization and egg development, effects
on reproduction would be expected (Zotin, 1958; Ketola et al., 1988). Still, the standard survival
tests provide a useful benchmark for understanding toxic potential. Other tests, which are more
difficult and time consuming to run, can be used to extrapolate short-term tests on survival to
longer-term exposures, sublethal responses and other species.
5.1. TOXICITY TESTS ON WATER COLUMN AND SEDIMENTS DOWNSTREAM
OF MTM-VF
Only one study (Merricks et al., 2007) tested media downstream of MTM-VF within the
study area. Water and sediment collected from some, but not all, sites downstream of valley fills
produced significant toxicity in laboratory organisms.
Water was tested using Ceriodaphnia dubia. Results were reported as the percent
dilution that killed one-half of the test organisms over 48 hours (48-hour LC50). Three streams
were tested. The frequency of toxicity was highest in Lavender Fork; undiluted water from three
of the eight sites killed 50% or more of the test organisms. Lavender Fork also had the highest
specific conductivity levels; the undiluted water at the three toxic sites averaged 3,050, 2,497,
and 2,657 (iS/cm. Specific conductivity measurements were available for two of the five sites
from Lavender Fork that did not result in 50% or greater mortality; specific conductivity
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 35 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
measurements (2,720 and 2,667 fiS/cm) were comparable to the toxic sites. Only 1 of 20 sites
from the other two streams was sufficiently toxic to kill 50% or more of the test organisms.
Specific conductivity measurements in these streams ranged from 923 to 1,643 nS/cm. There
was no obvious relationship between toxicity and water column measurements of trace metals
(e.g., Al, Fe, Mn, Zn, and Se).
Merricks et al. (2007) also conducted toxicity tests on sediments with another crustacean
Daphnia magna. The organisms were exposed to sediments for 10 days; results were reported as
percent survival and reproduction. Sediments from two of eight sites on Lavender Fork
significantly reduced survival or reproduction of Daphnia magna. Sediments from 3 of 19 sites
on the other two tested streams produced reduced survival or reproduction. Of the three streams,
Lavender Fork generally had the highest concentrations of trace metals in sediments (i.e., Al, Fe,
Cu, Cd, mercury (Hg), Se, As, Mn, and Zn). Concentrations of major ions or other chemicals
were not measured. Because of the way the sediment chemistry results were grouped for
summary, it is difficult to quantitatively relate them to the toxicity test results.
Asian clams (Corbicula fluminea) were deployed at monitoring stations (Merricks et al.,
2007). Growth was significantly greater below the ponds and decreased downstream, indicating
that the ponds increased the food available to the clams. Significant mortality was observed at 1
of 16 test sites. The authors attributed the mortality to Al and Cu, which had been detected in a
previous, unpublished study at water concentrations of 223 and 7.6 |ag/L, respectively.
5.2. TOXICITY TESTS ON WATER FROM OTHER ALKALINE COAL MINING
EFFLUENTS
In a series of studies, Kennedy et al. tested the toxicity of a mining effluent from Ohio
using Ceriodaphnia dubia and the mayfly lsonychia bicolor (Kennedy et al., 2003, 2004, 2005).
The effluent originated from a surface mine, an underground coal mine and a preparation facility.
Discharges from the underground mine and preparation facility were treated in a settling pond to
neutralize pH and reduce Mn, resulting in an effluent with high S042-, Na+, and CI"
concentrations and a mean hardness of 770 mg/L as CaC03. Toxicity tests using Ceriodaphnia
dubia were conducted following EPA protocols and used moderately hard reconstituted water
(MHRW)8 to dilute the effluent. Survival of Ceriodaphnia dubia in 48-hour tests significantly
decreased relative to controls at a mean specific conductivity of 6,040 (jS/cm (Kennedy et al.,
2003). Decreased survival in 7-day tests was observed at a mean specific conductivity of
4,730 jaS/cm. Decreased reproduction in 7-day tests was observed at a mean conductivity of
'Moderately hard reconstituted water (MHRW) was used as diluent in this study and many of the other studies
discussed in section. MHRW has low chloride concentrations (mean of 1.9) and a Ca:Mg molar ratio of 0.88;
hardness ranges from 80-100 mg/L as CaC03 (Smith et al., 1997).
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 36 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
3,254 (iS/cm, about 1.9 times lower than the 48-hour results for survival (Kennedy et aL 2005).
Tests on simulated effluent made using only the major ions (i.e., no heavy metals) agreed well
with the whole effluent, providing evidence that the toxicity was caused by the ions, rather than
an unmeasured toxicant (Kennedy et al., 2005).
The same field-collected effluent was tested with a nonstandard test species, the mayfly
Isonychia bicolor (Kennedy et al„ 2004) in 7-day tests. In these tests, water from an unpolluted
reference stream was filtered and used as dilution water for the tests. Toxicity was greater at the
warmer temperature tested (20°C vs. 15°C); those results are reported here. Survival of
Isonychia significantly decreased relative to controls at specific conductivities of 1,562, 966. and
987 (aS/cm for three tests. These conductivities are about 3 times lower than those that reduced
Cerioduphnia reproduction in 7-day tests using the same dilution water but a higher temperature
of 25°C.
Chapman et al. (2000) tested a high sulfate alkaline coal mine effluent from Alaska in
10-day tests using the insect Chironomus tentans. No effects on chironomid survival were
found, but dry weight was reduced approximately 45% in synthetic effluent (2,089 TDS/L). The
researchers also tested the effects of synthetic effluent on rainbow trout using two exposures:
eggs were exposed for 4 days starting immediately after fertilization, and swim-up fry were
exposed for 7 days. No adverse effects were seen in embryo viability or fry survival in the
highest synthetic effluent concentrations tested (2,080 TDS/L).
5.3. TOXICITY OF MAJOR IONS: K\ HC03", Mg2\ CP, S042~, Na\ Ca2+
Laboratory studies that vary ion mixtures provide additional insight into which ions may
be driving toxicity and how interactions may be producing observed effects. We report on the
results of three of these study groups. Then, we compare the experimental results to ion
concentrations reported downstream of MTM-VF operations to gauge whether ion
concentrations would be expected to cause toxicity.
5.3.1. Mount et al., 1997
Mount et al. (1997) tested the acute toxicity of over 2,900 ion solutions using two
crustacean species: Ceriodaphnia dubia and Daphnia magna and the fathead minnow
(Pimephales promelas). Ceriodaphnia was the most sensitive of the three organisms. The
toxicity of ion mixtures varied greatly with composition; total ion concentrations corresponding
to acute LC50 for Ceriodaphnia ranged from 390 mg/L to over 5,610 mg/L. For Pimephales
promelas, LC50 values ranged from 680 to 7,960 mg/L. The authors reported relative toxicity as
K+ > HCCV = Mg2+ > Cl > S042-. They also developed regression models that could be used to
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 37 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
predict the 48-hour acute toxicity of field-collected samples. In the models, the effects of the
anions and cations were generally additive with two notable exceptions: Solutions with high
concentrations of multiple cations had lower toxicity than expected based on concentration
addition, and Naf and Ca2+ did not add any explanatory value after the other ions were included
in the model.
The regression models have been used to predict the toxicity of several complex
effluents. Tietge et al. (1997) used them to predict the acute toxicity of the ionic component of
production waters to Ccriodaphnia dubia, Daphnia magna and Pimephales promelas (Tietge
et al., 1997). Toxicity of the Ohio coal mine effluent (described above) to Ceriodaphnia dubia
was less than expected based on the equations, although estimates were within a factor of 2
(Kennedy et al., 2005). Soucek (2007b) found that the model overestimated the toxicity of high
hardness solutions to Ceriodaphnia dubia by a factor of about 5 (10% survival predicted vs. 50%
survival observed).
5.3.2. Soucek (2007a, b); Soucek and Kennedy (2005)
Soucek (2007a, b) and Soucek and Kennedy (2005) conducted a series of 48-hour tests
on S042~ using MHRW dilution water and varying levels of other ions and hardness. At the
highest hardness tested (600 mg/L), the 48-hour LC50 value for Ceriodaphnia dubia was
3,288 mg SO4/L (Soucek and Kennedy, 2005). In all tests, the crustacean Hyalella azteca was
the most sensitive test organism, followed by Ceriodaphnia dubia, the bivalve Sphaertum simili
and the insect Chironomus tentans 9 Hyalella azteca was particularly sensitive to SO4 at low CP
concentrations. At CT concentrations of 1.9 mg/L, Hyalella azteca was four times more
sensitive to S04 than Ceriodaphnia dubia (Soucek, 2007a). Toxicity decreased as Ca increased
relative to Mg concentrations (Soucek and Kennedy, 2005). Toxicity also decreased with
increasing hardness, although the ameliorative effects of hardness appeared to level off above
500 mg/L hardness as CaCC^.
In three-brood, 7-day tests on Ceriodaphia dubia, sublethal effects of S04 occurred at
concentrations 2.5 times lower than those that reduced survival (Soucek, 2007a). The lowest
concentration at which effects were significant compared with controls (the LOEC) was
899 mg SO4/L for a reproductive endpoint (mean number of neonates per female) compared with
2,216 mg/L for percent survival. Other sublethal effects were investigated using 24-hour tests;
significant declines in feeding rates and oxygen consumption were observed in Ceriodaphia
dubia exposed to 1,000-mg SO4/L.
9Chironomus tentans has since been renamed Chironomus dilutus
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 38 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
5.3.3. Meyer et al., 1985
Meyer et al. (1985) tested four salts using 48-hour tests on Daphnia magna and 96-hour
tests on Pimephalespromelas. High hardness dilution water was used (563 mg/L as CaCOi).
Daphnia magna was more sensitive to all of the salts than Pimephales promelas. The relative
toxicity of the salts was MgSC>4 > NaCl > NaNC^ > Na2S04.'° The LC50 values calculated for
MgSC>4 were 4,300 mg/L and 7,900 mg/L for Daphnia magna and Pimephales promelas,
respectively. All of these values are well above concentrations reported downstream of
MTM-VF (see Table 4).
5.4. COMPARING TOXICITY TESTS ON MAJOR IONS TO OBSERVATIONS
DOWNSTREAM OF MTM-VF
Applying the Mount et al. (1997) regression models to ion concentrations reported
downstream of MTM-VF suggests that the ion mixture at some sites may reach acutely lethal
levels. The models predict minimal mortality of Ceriodaphnia duhia (1%) at mean
concentrations of each ion reported in Pond et al. (2008) (mean specific conductance of
1,023 nS/cm).11 However, applying the assumption that ion concentrations are strongly
correlated, we also calculated predictions using the maximum reported concentrations for each
ion (maximum specific conductance of 2,540 (iS/cm). More than 75% mortality is predicted at
these maximum concentrations. The models predict minimal mortality (1%) for Pimephales
promelas even at maximum concentrations.
Model predictions of toxicity are generally consistent with the observed Ceriodaphnia
duhia toxicity test results reported by Merricks et al. (2007). Five sites tested by Merricks et al.
(2007) had specific conductivity measurements comparable or greater than the maximum
specific conductivity reported by Pond et al. (2008) (2,540 (aS/cm). If the relative proportion of
ions was the same in Merricks et al. (2007) as in Pond et al. (2008), we would expect these high
conductivity sites to produce greater than 75% mortality. Three of these five sites exhibited 50%
or greater mortality in 48-hour tests. Of the 11 sites with substantially lower specific
l0MgSO4 = magnesium sulfate; NaCl = sodium chloride; NaN03 = sodium nitrate; Na2S04 = sodium sulfate.
"For Ceriodaphnia dubia, proportion surviving (P) in 48-hour tests was calculated as logit (P) = ln[/V(l - /')] r
8.83 x ((-0.0299 * [Kf]) + (-0.00668 * [Mg2 ]) + (-0.00813 x [CI"]) + (-0.00439 x [S042 ]) +
(-0.00775 x [HC03~]) + (-0.446 x 2) + (0.00870 x 2 x [K*]) + (0.00248 x 2 x [CP]) + (0.00140 x 2 x [S042 ]))
(Mount et al., 1997). Concentrations are as reported in Pond et al. (2008) except for HC03 . HC03" concentrations
were reported as CaC03 (Personal Communication from M.A. Passmore, U.S. EPA Region III, Wheeling, WV,
2009) and were converted to HC03~ concentrations by multiplying by 1.22.
For Pimephales promelas, proportion surviving (P) in 96-hour tests was calculated as logit (P) = ln[/V(l - P)] =
4.70 x ((-0.00987 x [K+]) + (-0.00327 x [Mg2t]) + (-0.00120 x [CI ]) + (-0.000750 x [S042 ]) +
(-0.00443 x [HC03~])) (Mount et al., 1997). Concentrations are as reported in Pond et al. (2008) except for HC03~.
HC03~ concentrations were reported as CaC03 (Personal Communication from M.A. Passmore, U.S. EPA Region
III, Wheeling, WV, 2009) and were converted to HC03~ concentrations by multiplying by 1.22.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 39 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
conductivity readings (all less than 1,643 (iS/cm), only 1 exhibited greater than 50% mortality in
the toxicity tests.
The SO4 48-hour LC50 for Ceriodaphnia dubia survival under high hardness conditions
(3,288 mg SO4/L) is twice the maximum concentration reported downstream of MTM-VF (see
Table 4), suggesting that concentrations of SO4 alone are unlikely to reach levels that are acutely
toxic. However, SO4 concentrations may be sufficiently elevated at some sites to cause sublethal
effects in Ceriodaphnia dubia. The LOEC reported for reproductive effects in the 7-day tests
(899 mg/L SO4) falls between the median and maximum values reported in Table 4, suggesting
that effects would be expected at many but not all sites downstream of MTM-VF. The higher
hardness levels shown in Table 4 would be expected to ameliorate toxicity. However,
bicarbonate and potassium concentrations are higher downstream of MTM-VF relative to the test
conditions. Based on the Mount et al. (1997) tests described in Section 5.3.1, higher
concentrations of bicarbonate and potassium would be expected to increase the toxicity of the
mixture.
The toxicity tests on other alkaline mine effluents discussed in Section 5.3.1 suggest that
effects to other organisms should be expected at concentrations below those that affect
Ceriodaphnia. Tests using the mayfly lsonychia bicolor and the amphipod Hyalella azleca
found effects on survival at concentrations 3-4 times lower than those affecting Ceriodaphnia,
If effects on reproduction in these organisms are similarly more sensitive than survival, effects
would be expected at most sites downstream of MTM-VF.
The relatively high sensitivity of mayflies to ions in alkaline mine effluent is consistent
with relative sensitivity of mayflies to other salts. Mayflies were the most sensitive order of
invertebrates tested in 72-hour laboratory studies of NaCl on South African invertebrate species
(Kefford et al., 2004). In studies on metal salts in experimental streams and toxicity tests from
the United States, the most sensitive invertebrates tend to be mayflies (Warnick and Bell, 1969;
Clark and Clements, 2006). In studies on artificial seawater (dominated by NaCl) from
Australia, the most sensitive species also were mayflies (Kefford et al., 2003).
Finally, there is some evidence that effects on survival of younger organisms may be
observed at concentrations below those that affected the test organisms. The mayfly test
reported by Kennedy employed older, larger instars of lsonychia (Kennedy et al., 2004). In tests
with bicarbonate, 7-day-old Hyalella azleca were two times more sensitive than 14-day old
organisms (Lasier et al., 1997). In studies on metal salts in experimental streams (Cu, Cd, and
Zn), toxicity increased as organism size decreased (Kiffney and Clements, 1996).
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 40 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
5.5. TOXICITY OF TRACE METALS IN WATER
5.5.1. Selenium
Se is a metalloid element that is a micronutrient and, at higher exposures, a toxicant.
Selenium from coal ash and coal mine wastes has resulted in elevated Se concentrations in
surface waters and toxicity to aquatic organisms (Orr et al., 2005). Se is unusual in that its
toxicity results from complex processes of transformation and bioaccumulation, analogous to
mercury toxicity. Environmental exposures of animals are primarily dietary, and effects on
sensitive early life stages are due primarily to maternal transfer. The current chronic AWQC for
Se is 5.0 |ig/L, and the median, mean and range of Se concentrations in streams draining valley
fills arc 12.5, 10.6, and <1.5-36.8 jag/L, respectively (Bryant et al., 2002; Pond et al., 2008). The
chronic criterion is relevant because the discharge from mining operations is a chronic source.
This section discusses effects of Se on aquatic invertebrates, fish and birds, emphasizing studies
of waters receiving coal overburden leachates because the valley fills are filled with coal
overburden.
5.5.1.1. Invertebrates
A review of the literature estimated that the range of thresholds for sublethal toxicity in
aquatic invertebrate genera is 1-30 |ig/L (DeBruyn and Chapman, 2007). A recent study showed
that dietary selenium is bioaccumulated by the mayfly Centroptilum triangulifer and suggested
that reproductive effects occur at aqueous exposures of 13.9 (ag/L dissolved Se (Conley et al.,
2009). These results are consistent with data from streams draining Canadian coal mines that
found a >50% decline in the .abundance of some taxa in the range of 5-100 jj.g/L (DeBruyn and
Chapman, 2007). In outdoor artificial streams dosed with Se, isopods (Caecidolea) and
oliogochaete worms (Tubifex) were severely reduced in abundance at 30 |ig/L and statistically
significantly reduced at 10 |ig/L (Swift, 2002). However, the abundances of baetid mayfly
nymphs (Baetis, Callibaetis), damselfly nymphs (Enallagma) and chironomid larvae were not
statistically significantly reduced—even at 30 jag/L.
5.5.1.2. Fish
Numerous studies have shown severe effects of Se on fish reproduction in the field as
well as in the laboratory, and effects on fish are the basis for the national criterion (U.S. EPA,
2004). Cutthroat trout embryos from a pond at a coal mine in British Columbia with 93 pg/L Se
showed effects ranging from deformities of larvae to mortality (Rudolph et al., 2008). The
probability of mortality was correlated with Se concentrations in the embryos. These trout are
much less sensitive than other species such as bluegill sunfish. In the artificial stream study.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 41 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
bluegill sunfish exhibited mortality and characteristic skeletal deformities at all concentrations,
including 2.5 |ig/L, although the effects were not statistically significant at that lowest level
(Swift, 2002). Creek chubs and blacknose dace from the Coal, Little Coal, Big Coal and Mud
River watersheds in West Virginia contained Se from <0.48 to 6.89 mg/kg dry weight (Paybins
et al., 2000). Fish from 3 of 22 of these streams had concentrations >5 mg/kg, putting them at
"moderate hazard" for toxic effects based on the scale developed by Lemly (1993).
5.5.1.3. Birds
Se has caused reproductive failure and gross deformities in birds that forage in
Se-contaminated waters, but their sensitivity is highly variable (Ohlendorf et al., 2003). Birds
foraging in streams receiving leachate from coal mine overburden in the Elk River, British
Columbia, watershed showed reproductive effects, but they were less severe than expected given
the high Se concentrations (8.1-34.2 jag/L) (Harding et al., 2005). In particular, spotted
sandpipers experienced a reduction in egg hatchability from 92% in reference streams to 78% in
streams receiving overburden leachate. Spotted sandpipers forage in streams in the Appalachian
Range, but the Louisiana waterthrush occurs more commonly in the area of concern and forages
on aquatic invertebrates, so it would be similarly exposed. The authors suggest that the low level
of effects relative to other Se-contaminated waters was due to low bioaccumulation, which was
due to the low rates of biotransformation and uptake in those streams. Piscivorous birds
(primarily Belted Kingfishers and Great Blue Herons) may be at risk from Se-contaminated fish.
Creek chubs and blacknose dace from the Coal, Little Coal, Big Coal, and Mud River watersheds
contained Se from <0.48 to 6.89 mg/kg dry weight (Paybins et al., 2000). The 10th percentile
effective concentration for hatchability in dietary exposures of mallard ducks (a surrogate species
for the piscivorous birds) to Se in dry diet was 4.87 mg/kg (Ohlendorf et al., 2003). Five of the
22 fish samples from 13 streams analyzed by Paybins et al. (2000) for Se from the Coal, Little
Coal, Big Coal and Mud River watersheds exceeded that endpoint.
5.5.2. Manganese and Iron
Concentrations of Mn and Fe were higher at some sites downstream of valley fills
compared with sites in unmined watersheds (see Table 6). In addition, Mn and Fe deposits have
been observed on caddisflies downstream of MTM-VF (see Figure 9), suggesting that exposure
is occurring under some circumstances (Pond, 2004).
Maximum concentrations of Mn reported downstream of MTM-VF are substantially
lower than those associated with effects in the few available toxicity tests. Maximum
concentrations of dissolved Mn reported in Pond et al. (2008) were 0.853 mg Mn/L. Tests using
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 42 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Ceriodaphnia dubia in hard water (hardness = 184 mg/L) yielded a mean 48-hour LC50 of
15.2 mg Mn/L for Ceriodaphnia dubia and a 96-hour LC50 value for Hyalella azteca of
13.7 mg Mn/L (Lasier et al., 2000). In 7-day tests, Ceriodaphnia dubia reproduction (number of
young per female) was inhibited 50% at mean concentrations of 11.5 mg Mn/L.
Pond et al. (2008) reports maximum concentrations of total and dissolved Fe as 650 ng/L
and 281 jj.g/L, respectively. These concentrations are similar to the median tolerance limit
concentration of 320 |ig FeSCU/L (water hardness = 48 mg/L) reported for Ephemerella sp.
survival in a study conducted prior to standardized toxicity test protocols (Warnick and Bell.
1969). The concentrations are also greater than several of the family-level benchmarks for total
Fe derived from field observations of benthic macro invertebrates from West Virginia (see
Table 11). Benchmark values (called field effect concentrations, FEC20s) corresponded to a
20% decline in the organism numbers compared with reference sites and were estimated from the
90th percentile quantile regression relationship between total Fe and numbers of organisms
collected from different families. However, because the benchmark derivation did not control
for stressors that covary with iron, the benchmarks may reflect the effects of other stressors in
addition to iron.
5.6. TOXICITY OF TRACE METALS IN SEDIMENT
Only two studies measured concentrations of trace elements in sediments. Most
concentrations were below available consensus-based screening levels (see Table 12). The
consensus-based screening levels are based on analysis of paired sediment chemistry and toxicity
test results from field studies and should be interpreted as concentrations at which effects in
toxicity tests are frequently observed. Zinc and Ni concentrations in Kanawha Valley sediments
exceed the probable effects levels and warrant further investigation. Toxicity of Zn and Ni is a
function of particle size, organic carbon content, pH and acid volatile sulfides (Di Toro et al.,
2001; Doig and Liber, 2006). It is difficult to interpret the observed concentrations without
measurements of the factors that influence toxicity, or alternatively, pore-water concentrations.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 43 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Table 11.
2007)
Field-based 20% effect concentrations (FEC20) (Linton et al.,
Macroinvertebrate family
FEC20 (jig total iron/L)
Leptophlebiidae
210
Emphemerellidae
430
Philopotamidae
440
Psephenidae
480
Heptageniidae
660
Elmidae
1,130
Baetidae
1,480
Tipulidae
7,050
Table 12. Comparison of measured sediment concentrations with probable
effects levels
Chemical
Concentration downstream
of MTM-VF (mg/kg)
(Merricks et al., 2007)"
Concentration in Kanawha
Valley sediments (mg/kg)
(Paybins et al., 2000)b
Consensus probable effects
level (mg/kg)
(MacDonald et al., 2000)c
Al
3-32
As
0.015-0.070
4-20
33
Cd
0.005-0.045
4.98
Cr
60-110
11 1
Cu
0.019-0.122
149
Fe
<48.5-157.6
Pb
20-50
128
Mn
1-41
Hg
0.006-0.015
1.06
Ni
50-100
48.6
Se
0.1-2.5
Zn
2.0-2.5
200-600
459
"Data from Table III and Figure 3 combined.
bData from figures in appendix.
"We note that the concentrations reported in Merricks et al. (2007) are substantially lower than ranges of values
reported in Paybins et al. (2000) or used to develop the PECs (e.g., see Smith et al., 1996) suggesting that any
comparisons should be made with caution.
Blank cells indicate that the metal was not measured, or there is no probable effects level available.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 44 DRAFT—DO NOT CITE OR QUOTE
-------
1
2 Figure 9. Mn (black) and Fe (orange) deposits on a caddisfly collected
3 downstream of a mountaintop mine and valley fill.
4
5 Source: Pond (2004).
This document is a draft for review purposes only and does not constitute Agency policy.
7/09 45 DRAFT—DO NOT CI IE OR QUO TE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
6. IMPACTS ON AQUATIC ECOSYSTEMS
In this section, we report on changes in stream community composition associated with
MTM-VF. In contrast to the toxicity tests discussed in Section 5, field studies are our primary
resource for this section because they directly consider both the exposures and biota of interest.
Macroinvertebrate and fish assessments indicate degraded biological condition downstream of
MTM-VF.
6.1. EFFECTS ON BIOLOGICAL COMPOSITION
Mountaintop mining and associated valley fill in a watershed is associated with degraded
community composition in downstream habitats. Though there are relatively few studies on the
direct ecological effects of mountaintop mining, the findings are unambiguous (Howard et al.,
2001; Stauffer and Ferreri, 2002; Fulk et al., 2003; Hartman et al., 2005; Merricks et al., 2007;
Pond et al., 2008). Across all the relevant studies reviewed, mayfly, i.e., insect order
Ephemeroptera, populations were consistently lower in streams draining watersheds with
MTM-VF than in streams draining watersheds with intact forest. Associated with the extirpation
of mayfly species, biological assessment metrics indicate degraded conditions immediately
downstream of MTM-VFs.
6.1.1. Benthic Macroinvertebrates
6.1.1.1. Benthic Macroinvertebrate Indices
All surveys that used multimetric and aggregate taxonomic indices observed degraded
biological conditions in streams affected by mining and valley fills (see Table 13). Fulk et al.
(2003) used the West Virginia Stream Condition Index (WV SCI) to analyze benthic
macroinvertebrate data from 34 streams in West Virginia. The index is composed of several
metrics that are responsive to environmental and chemical stress, e.g., EPT (Ephemeroptera,
Plecoptera and Triehoptera) taxa, total taxa and percent EPT were expected to decrease with
increasing stress and percent Chironomidae; Hilsenhoff biotic index (HBI) and percent of the top
two dominant taxa were expected to increase with increasing stress. Four classes of streams
were compared: (1) no mountaintop mining upstream (n = 9), (2) upstream valley fills (n - 15),
(3) mountaintop mining in watershed (n = 4) and (4) upstream valley fill and residential
development in the watershed (n = 6). Fulk et al. (2003) observed that benthic macroinvertebrate
indices were lower in streams with upstream valley fills. With the exception ot the fall of 2000,
the macroinvertebrate index showed significant differences among stream classes, and these
differences were caused by fewer total taxa and fewer EPT taxa in streams with valley fill.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 46 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
While unmined sites were typically classified as "very good" streams with upstream valley fills
had WV SCI scores that ranged from "fair" to "good", indicating that stream sites with valley fill
were degraded compared to unmined sites. Similar results were observed in an assessment of the
biological condition of streams in Kentucky. Howard et al. (2001) calculated the Kentucky
Macroinvertebrate Bioassessment Index (MBI) (Pond and MacMurray, 2002), which includes
four components of macroinvertebrate community condition, and showed that streams with
mining activity in the watershed in Kentucky had lower MBI ranks than streams in watersheds
without mining (mined streams had a rank of poor, and reference streams were good).
Merricks et al. (2007) observed a 1.91 HBI score in a reference stream, indicative of excellent
water quality, and HBI values ranging between 5.64 and 5.7, indicative of fair water quality,
below valley fills and ponds in three streams. Finally, in a comparison of streams with and
without mining in the watershed, Pond et al. (2008) observed that streams below fills had a
significantly lower macroinvertebrate biotic index score than those without fills using both a
genus-level index of most probable steam status (GLIMPSS, 2.4 vs. 4.5, respectively) and a
family-level biotic index (WV SCI, 3.4 vs. 4.3, respectively).
Most field studies reported a reduction in commonly used measures of sensitive
macroinvertebrates, the aggregated EPT metrics, at sites downstream of MTM-VF. EPT
taxonomic richness was lower in one study (EPT generic richness of 17.9 at unmined sites and
8.9 at filled sites. Pond et al. 2008) and showed no significant difference in another
(Merricks et al., 2007). Hartman et al. (2005) observed no difference in EP T richness between
mined and unmined streams. An EPT index was lower in streams in mined watersheds
compared to measures in streams in watersheds without mining activity (an average of 8.9 in
mined sites and 21 in reference sites) (Howard et al., 2001). And the percentage of the
community comprised of EPT taxa was lower at sites downstream of MTM-VF (Merricks et al.,
2007; Pond et al., 2008). The mixed effects of mining on EPT aggregate measures likely reflect
legacy land-use differences, differences in location of sample sites (e.g., sampling close to a
pond) and taxonomic shifts within insect orders.
6.1.1.2. Bent hie Macroin vertebrate Diversity
In most cases, lower taxonomic diversity was observed at sites downstream of MTM-VF.
A pattern of lower macroinvertebrate richness in streams with mining in the watershed was
found in Kentucky (mean of 31 at mined sites and 43 at reference sites, Howard et al., 2001) and
in West Virginia (mean generic richness of 21.7 at mined sites and 31.9 at unmined sites, Pond
et al., 2008). In contrast, Merricks et al. (2007) found no significant difference in taxonomic
richness between filled streams and a stream without fill in the watershed.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 47 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
6.1.1.3. Benthic Macroinvertebrate Density
No difference was found in the total density of macroinvertebrates between streams with
valley fill and reference streams (Hartman et al., 2005).
6.1.1.4. Benthic Macroinvertebrate Functional Groups
MTM-VF were associated with changes in the functional composition of
macroinvertebrate communities. Typically, macroinvertebrate communities in headwater
streams are dominated by shredders, which feed on leaf detritus (e.g., Vannote et al., 1980). In
the case of mining activities, shredder density metrics (Hartman et al., 2005) and proportion of
the community (3% in streams with mining and 50% in a reference stream, Merricks et al., 2007)
were lower in streams below fills. Other changes include lower percentage of the community as
dingers (i.e., organisms that cling to rocks in riffles) in mined watersheds than in watersheds
without mining (Howard et al., 2001). Also, a scraper (i.e., organisms that feed on attached
algae) density metric was lower in filled streams than it was in streams without valley fill in the
watershed (Hartman et al., 2005). The percentage of the community as collector-filtering
macroinvertebrates (i.e., organisms that feed on floating algae) was higher in streams below fills
(Merricks et al., 2007), but a collector density metric showed no difference between streams
below fills and reference streams in another study (Hartman et al., 2005).
6.1.1.5. Benthic Macroinvertebrate Taxa
Specific changes in macroinvertebrate taxonomic composition are described below.
6.1.1.5.1. Coleoptera.
Only one study included coleopteran populations in their assessment found that a density
metric of Coleoptera was lower in streams below valley fill than in streams without valley fills in
the watershed (Hartman et al., 2005).
6.1.1.5.2. Diptera.
The effect of MTM-VF on Diptera population characteristics was mixed. In some cases,
there were no observed effects of fills or mining on the watershed. For example, perhaps owing
to moderate degradation in the reference sites, density metrics for Diptera and Chironomidae, a
family within the insect order Diptera, showed no difference between streams downstream of
valley fills and those without (Hartman et al., 2005). Merricks et al. (2007) had similar findings,
where the percentage of the community comprised of Chironomidae showed no difference
between sites downstream of valley fills and a stream without fills (Merricks et al., 2007). In
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 48 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
another study, the percent Chironomidae was lower in streams with mining in the watershed than
in streams with no mining (27% in mined and 13% in unmined streams. Pond et al.. 2008). A
combined measure of the percent Chironomidae and Oligochaeta was higher in streams in mined
watersheds compared to streams in watersheds without mining (63% in mined and 3% in
reference streams, Howard et al., 2001). The family Chironomidae includes both tolerant and
intolerant taxa, which may account for the equivocal results.
6.1.1.5.3. Ephemeroptera.
Ephemeroptera population characteristics showed the most definitive changes associated
with mining activities, where they were consistently lower in streams affected by MTM-VF.
Ephemeroptera density metrics were lower in sites downstream of valley fills than in streams
without fill (Hartman et al., 2005). The proportion of the community as Ephemeroptera was
lower in impacted streams. Howard et al. (2001) found an average of 1% in streams with
mountaintop mining in the watershed and 55% in reference streams. Two additional studies
report similar observations of proportion (3%) in streams with mountaintop mining in the
watershed and 17% in reference streams [Merricks et al., 2007]; 7% in streams with mountaintop
mining in the watershed and 45% in streams with no mining [Pond et al., 2008]). Likewise,
Ephemeroptera richness was significantly lower in mine-impacted streams (Merricks et al., 2007,
Pond et al., 2008).
6.1.1.5.4. Odonata.
An Odonata density metric was significantly lower at sites downstream of valley fills
than it was in streams without upstream valley fills (Hartman et al., 2005).
6.1.1.5.5. Plecoptera.
MTM-VF had mixed effects on Plecoptera populations. In one case, richness was lower
in streams with mining in the watershed (2.7 genera) than in streams without mining (6 genera)
(Pond et al., 2008). But in another case, there was no significant difference in Plecoptera
richness between sites downstream of valley fills compared to those without upstream fill
(Merricks et al., 2007). A similar discrepancy was found with percent Plecoptera, where
Merricks et al. (2007) found lower percentages in sites downstream of valley fills (4.5% in
mined streams and 52% at a reference site), whereas Pond et al. (2008) did not detect a
difference between streams with mountaintop mining in the watershed and streams with no
mining in the watershed. No difference was observed in a Plecoptera density metric between
streams with and without valley fills in Hartman et al. (2005).
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 49 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
6.1.1.5.6. Trichoptera.
MTM-VF had mixed effects on Trichoptera populations in streams. When the area below
the pond was sampled, the proportion of the macroinvertebrate community that was in the order
Trichoptera was higher in streams with mining in the watershed compared to streams without
mining (20% in mined and 3.7% in reference watersheds, Merricks et al. 2007). Another study
found no difference among streams downstream of fills and those without fills (Hartman et al.,
2005).
6.1.1.5.7. Noninsect Benthic Macroinvertebrates.
A density metric of noninsect macroinvertebrates was significantly lower in at sites
downstream of valley fills than in streams without fills (Hartman et al., 2005).
6.1.2. Fish
Fish community attributes, e.g., species richness, arc widely used to evaluate stream
condition (Karr, 1981; Angermeier et al., 2000). Species richness and the number of benthic fish
species were consistently associated with site quality in Mid-Atlantic Highland streams
(Angermeier et al., 2000), where both attributes declined with increasing degradation
(Barbour et al., 1999). Mountaintop mining for coal and creation of valley fills has had a
harmful effect on the composition of stream fish communities. Comparison of streams without
mining in the watershed and sites downstream of valley fills in Kentucky (five unmined sites and
seven filled sites) and West Virginia (four unmined sites and two mined sites) indicate that
streams affected by mining had significantly fewer total fish species and fewer benthic fish
species than streams without mining in the same areas (Stauffer and Ferreri, 2002). A similar
pattern of fewer taxa in streams affected by mining was observed with species richness (median
of 6 in sites downstream of valley fills and 12 in unmined streams, Stauffer and Ferreri, 2002).
For example, in Kentucky, Stauffer and Ferreri (2002) observed sites downstream of valley fills
had a median richness of 7 fish species, compared to a median richness of 12 fish species in
streams without mining in the watershed. In these streams, the number of benthic fish species
was also lower in the sites downstream of valley fills (median = 1 benthic species) than in the
streams without mining in the watershed (median = 6 benthic species).
Fulk et al. (2003) used the Mid-Atlantic Highlands Index of Biotic Integrity (IBI) to
analyze fish data from 27 streams in West Virginia. The index is composed of several metrics
that are responsive to environmental and chemical stress, e.g., native intolerant taxa, native
Cyprinidae taxa, native benthic invertivores, percent Cottidae, percent gravel spawners, and
percent piscivore/invertivores were expected to decrease with increasing stress, and percent
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 50 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
macro-omnivore, percent tolerant fish and percent exotic fish were expected to increase with
increasing stress. In their study, Fulk et al. (2003) classified streams (e.g., no mining in the
watershed, sites downstream of valley fills, mountaintop mining in the watershed, sites
downstream of valley fills and with residential development in the watershed) and compared fish
metrics and the composite IBI among stream classes. 1BI scores from the sites downstream of
valley fills were significantly lower than scores from sites without mining in the watershed by an
average of 10 points, indicating that fish communities were degraded in sites downstream of
valley fills. In their analysis, Fulk et al. (2003) suggest that the reduced index score was caused
by fewer minnow species and benthic insectivores in sites downstream of valley fills compared
to streams without mining in the watershed. Index scores were also lower at sites with mining in
the watershed compared to scores from streams without mining in the watershed. Watershed
size was also an important factor in this analysis. Sites with mining and valley fills in small
watersheds (<10 km2) showed more degradation than sites with mining and valley fills in large
watersheds (>10 km2) (Stauffer and Ferreri, 2002; Fulk et al., 2003).12
6.2. EFFECTS ON ECOLOGICAL FUNCTION
No studies were found that assessed the impact of MTM-VF on ecological function, e.g.,
biogeochemical cycling, in downstream habitats. Additional research is needed to better
understand the effects of MTM-VF on ecological function in downstream sites.
6.3. BIOLOGICAL CONDITION
The biological effects downstream of MTM-VF are consistent with generic narrative
descriptions of moderately to severely degraded condition (Davies and Jackson, 2006). The
biological condition gradient (BCG) is a framework that identifies 10 attributes of stream
ecosystems indicative of biological status ranging from pristine, natural condition (Tier 1) to
severely degraded condition (Tier 6) (Davies and Jackson, 2006) (see Figure 10). Evidence was
available to evaluate 3 of the 10 ecological attributes described in the BCG. Sensitive taxa,
specifically insect Order Ephemeroptera, are markedly diminished downstream of MTM-VF
(Tier 5). The spatial and temporal extent of detrimental effects is between the reach- and
catchment-scale (Tiers 4 to 6). Finally, the burial of the headwaters and the construction of
channels correspond with a loss of ecosystem connectance between 'some' and 'complete' (Tiers
4 to 6). The attributes identified in the BCG highlight many data gaps—including documenting
the extent of regionally endemic taxa, reporting the relative tolerance of taxa to the stressors
12Because larger watersheds typically have greater fish diversity than smaller watersheds, both studies adjusted their
analyses to account for the potential effect of watershed size.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 51 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
specific to the region, identifying the presence of nonnative organisms, reporting the condition of
organisms and measuring ecosystem functions in both reference and MTM-VF streams. The
BCG provides a general framework and is intended to be locally calibrated by state and regional
scientists and resource managers. Local calibration would provide a useful framework for
describing the effects of MTM-VF and restoration efforts on stream condition.
6.4. RELATIONSHIP BETWEEN BIOLOGICAL METRICS AND ENVIRONMENTAL
FACTORS
Five environmental variables associated with mining and valley Fills are commonly
considered to potentially affect the ecological condition of downstream habitats: ion
concentration, heavy metal concentration, organic enrichment, changes to instream habitat and
changes to upstream land use/land cover. This section describes associations between these
variables and biological characteristics.
6.4.1. Ion Concentration
All studies report elevated ion concentration in MTM-VF (see Table 14), and many show
strong negative correlative relationships between biological metrics and specific conductance
and/or the concentrations of individual ions (Floward et al., 2001; Stauffer and Ferreri, 2002;
Fulk et al., 2003; Hartman et al., 2005; Merricks et al., 2007; Pond et al., 2008). Though not
focused on mountaintop mining effects, elevated conductivity (i.e., 500-8,000 (iS/cm, has been
shown to have a significant negative correlation with the number of pollution sensitive taxa in
benthic macroinvertebrate communities [Soucek et al., 2000; Kennedy et al., 2003]).
Ephemeroptera richness was negatively correlated with specific conductivity (Hartman et al.,
2005). Though Merricks et al. (2007) did not assess conductivity-macroinvertebrate
relationships among sites, they note that only sites with the highest levels of conductivity,
ranging between 2,657 to 3,050 ^iS/cm, lacked Ephemeroptera. Pond et al. (2008) performed the
most complete analysis of ions and observed strong negative relationships between specific
conductance and their biological assessment measures, GLIMPSS (r = -0.90) and WV SCI
(r = -0.80), total generic richness (r = -0.74), EPT generic richness (r = -0.88), number of
Ephemeroptera genera (r = -0.90) and the number of Plecoptera genera (r = -0.75). Pond et al.
(2008) further demonstrated a decline in number of Ephemeroptera taxa and community
proportion when conductivity levels exceeded around 500 (iS/cm. HCO3 , Ca, hardness, Se
S042~, Mg), K\ and Na' were also found to have strong negative correlations with biological
metrics (Stauffer and Ferreri, 2002; Pond et al., 2008). While these studies do not provide
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 52 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
enough detail to elucidate the mechanistic relationship of biological degradation to ion
concentration, the pattern clearly suggests a strong association between the two.
6.4.2. Specific Metals
Though contributing to overall ion concentration, the concentrations of individual metals
were negatively correlated with many of the biological metrics in streams. Hartman et al. (2005)
found strong negative relationships (Pearson's correlation coefficients ranged from -0.70 to
-0.98) between macroinvertebrate metrics and metals. For example, Ephemeroptera richness
was negatively related to Cu, Fe, Mn, and Ni; EPT taxa richness was negatively related to Mn
and Ni; Plecoptera richness was negatively related to Cd. That study as well as Merricks et al.
(2007) reported that metal concentrations were higher in mining streams with biological
degradation than in reference streams. These results suggest that elevated metal concentrations
associated with mine-impacted streams may contribute to differences in stream biota.
6.4.3. Organic and Nutrient Enrichment
Two studies suggest a possible association between differences in biota and organic
enrichment in streams affected by MTM-VF. Merricks et al. (2007) evaluated changes in the
composition of functional feeding groups below settling ponds to assess potential organic
enrichment. Stations closest to the ponds had significantly higher collector-filtercr populations
compared to stations further downstream. Merricks et al. (2007) also noted that the HBI was
elevated at all fill-influenced sites compared to a reference site. The HBI was developed to
respond to a nutrient and organic pollution gradient, but it is also responsive to other stressor
gradients, including increased fine sediments and specific conductivity (Paybins et al., 2000).
Pond et al. (2008) found moderate to strong correlations between NO3 and biological metrics
(Spearman correlation coefficients ranged between 0.39 and 0.90), but total phosphorus levels
were below detection levels at all sites.
6.4.4. Instream Habitat
There was little evidence of an association between changes in macroinvertebrate
community metrics and characteristics of instream habitat at sites downstream of MTM-VF. In
general, characteristics of macroinvertebrate community composition, such as percent dingers,
are predictably affected by stream habitat characteristics (e.g., Pollard and Yuan, 2009). In the
case of MTM-VF, most field studies found no systematic relationship between macroinvertebrate
metrics and habitat assessment measures (Howard et al., 2001; Hartman et al., 2005; Merricks
et al., 2007), which may suggest that habitat characteristics were not all that different between
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 53 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
reference and mined stream sites. Similarly, Hartman et al. (2005) did not observe a relationship
between macroinvertebrate metrics and fine sediment or turbidity. Individual physical habitat
variables and total rapid bioassessment procedure habitat scores were correlated with
macroinvertebrate indices in the study by Pond et al. (2008). Because the multimetric indices
used by Pond et al. (2008) are sensitive to many factors, the relationship between habitat and
macroinvertebrate metrics would be expected. However, they did not observe excessive
sedimentation in sites downstream of valley fills.
6.4.5. Disturbance and Loss of Upland Habitat
In addition to the effects described above, changes in upland and headwater areas of the
watershed may alter macroinvertebrate composition in downstream habitats. Headwater streams
are critical to downstream ecological condition and their alteration, as in mountaintop mining
and valley fill activities, may impact the integrity and the sustainability of downstream habitats.
Headwater streams provide downstream habitats with water, nutrients, food and woody debris
(Gomi et al., 2002; Wipfli et al., 2007). Moreover, the physical structure of headwater streams in
the landscape may affect populations and communities of stream organisms by influencing the
movement of sediment, chemicals and individuals to downstream reaches within the network
(Lowe et al., 2006).
The loss of trophic subsidies from headwater streams may lead to lower secondary
productivity in downstream habitats. Food resources, specifically organic matter and
macroinvertebrate prey, are transported from headwater streams to downstream habitats.
Organic material enters headwater streams through litter fall from riparian vegetation, surface
runoff of particulate and dissolved material and subsurface movement (Cummins et al., 1989;
Wallace et al., 1999). Once introduced, organic material can be retained in the headwater
stream, transformed through feeding of organisms in the headwater stream, or transported
downstream (Webster et al., 1999; Wipfli et al., 2007). For example, in a study of 52 streams in
Alaska, Wipfli and Gregovich (2002) found that headwater streams were a source of
macroinvertebrates and detritus to downstream habitats.
The loss of headwater stream habitat might cause less viable populations of taxa in
downstream reaches. Headwater streams sometimes serve as refugia and source areas for
downstream biological diversity (Meyer et al., 2007). These areas may facilitate a 'rescue effect'
where there is the potential for recolonization from undisturbed sites, and the presence of this
source of colonists can be a strong determinant of population resilience (Brown and
Kodricbrown, 1977).
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 54 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Finally, watershed characteristics and activities greatly affect the structure and the
function of streams. Houser et al. (2005) showed that intact riparian zones were not sufficient to
protect streams from the effect of upland disturbance. They examined the effects of upland soil
and vegetation disturbance on ecosystem respiration and found lower ecosystem respiration rates
in streams with higher levels of upland disturbance. This is relevant because mountaintop
removal represents a significant disturbance to the vegetation and soil characteristics in upland
areas. As a result, even when downstream reaches and associated riparian areas of a stream
appear intact, as in the case of MTM-VF, they may incur significant impacts from mountaintop
removal occurring upstream.
6.5. CUMULATIVE EFFECTS
There is little evidence in the peer-reviewed literature of cumulative impacts of mining
on downstream ecology. Fulk et al. (2003) found no evidence of additive effects of multiple
mines on the fish IBI. In another MTM-VF study, Pond et al. (2008) reported no evidence of a
significant relationship between the number of upstream valley fills and macroinvertebrate
indices.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 55 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 13. Summary of research examining the relationship between
2 mountaintop mining and ecological characteristics in downstream habitats
3
Reference
Experimental design
Ecological response
Observed effect"
Fulk et al.,
Fish and benthic macroinvertebrate survey
Fish IBI
Lower
2003
comparing MTM-VF streams {n = 21) to
regional reference streams (n = 5)
Invertebrate IBI
Lower
Hartman et al.,
Benthic macroinvertebrate survey
Coleoptera density
Lower j
2005
comparing filled streams (n = 4) to
reference streams without valley fill (n = 4)
Diptera density
No difference
Ephemeroptera density
Lower
Odonata density
Lower
Plecoptera density
No difference
Trichoptera density
No difference
Total density
No difference
EPT density
No difference
Chironomidae density
No difference
Noninsect density
Lower
Collector density
No difference
Scraper density
Lower
Shredder density
Lower
Howard et al.,
Benthic macroinvertebrate survey
Taxa richness
Lower
2001"
comparing streams in mined watersheds
{n = 8) to streams in watersheds without
mining activity (n = 4)
EPT index
Lower
Biotic index
Higher |
% dinger
Lower
% Ephemeroptera
Lower
% chironomids + oligochaetes
Higher
KY MB1
Lower
Merricks et al.,
Benthic macroinvertebrate survey
Total richness
No difference
2007
comparing filled streams (n = 4) to a
reference stream without valley fill (« = 1)
EPT richness
No difference
Ephemeroptera richness
Lower
Plecoptera richness
No difference
Trichoptera richness
No difference
Hilsenhoff Biotic Index
Higher
% Chironomidae
No difference
% EPT
Lower
% Ephemeroptera
Lower
% Plecoptera
Lower
% Trichoptera
Higher
% collector-filterer
Higher
% shredder
Lower
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 56 DRAFT—DO NOT CITE OR QUOTE
-------
1
Table 13. Summary of research examining the relationship between
mountaintop mining and ecological characteristics in downstream habitats
(continued)
Reference
Experimental design
Ecological response
Observed effect"
Pond et al..
Benthic macroinvertebrate survey
comparing MTM-VF streams (n = 27) to
untnined reference streams (n = 10)
Total richness
Lower
2008
EPT richness
Lower
F.phemeroptera richness
Lower
Plecoptera richness
Lower
WV genus biotic index
Lower
WV family biotic index
Lower
Shannon FT
Lower
% Orthocladiinae
Lower
% Chironomidae
Lower
% Ephemeroptera
Lower
% Plecoptera
No difference
% EPT
Lower
Stauffer and
Ferreri, 2002
Fish communities were compared in filled
streams {n = 9) to stream without mining
activity {n = 9)
Fish species richness
Lower
Benthic fish richness
Lower
2
3 "Comparing the mean values from the reference and downstream and/or mined sites, where lower indicates that the
4 mined/valley fill site has a significantly lower metric values than the reference site (significance as determined by
5 statistical analyses in original study).
6 ''The original study did not present statistical analyses on these comparisons.
This document is a draft for review purposes only and does not constitute Agency policy,
12/17/09 57 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 14. Average ion concentration (reported as specific conductance) in
2 MTM-VF and reference streams reported in conjunction with biological
3 data. Range values are included when reported by the source literature, Standard
4 error values were not reported in the source literature.
5
Source
Units of
measure
Filled
Reference
n
Mean (range)
n
Mean (range)
Hartman et al., 2005a
muhm/s [.s/c]
4
1051
4
150
Howard et al., 2000
|imhos/cm
8
994 (420-1,690)
4
47 (30-66)
Merricks et al., 2007ab
faS/cm
3
1,653 (991-2,720)
1
247
Pond et al., 2008
(aS/cm
27
1,023 (159-2,540)
10
62 (34-133)
Stauffer and Ferreri, 2002c
|amhos/cm
8
1,716(513-2,330)
9
164 (125-210)
6
7 "Averages calculated from reported values.
8 bValues taken from Site 1, which is the first site below valley fill and pond.
9 'Values reported were limited to the Mud River watershed.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 58 DRAFT—DO NOT CITE OR QUOTE
-------
1
Natural
c
o
c
o
o
"5
o
en
o
o
00
Degraded
Native or natural condition
Minimal loss of species; some
density changes may occur
Some replacement of
sensitive-rare species;
functions fully
maintained
Tolerant species show increasing
dominance; sensitive species are
rare; functions altered
Some sensitive species
maintained but notable
replacement by more-tolerant
taxa; altered distributions;
functions largely maintained
Severe alteration of
structure and function
Low Stressor gradient High
3
4 Figure 10. Conceptual model depicting stages of change in biological
5 conditions in response to an increasing stressor gradient.
6
7 Source: Davies and Jackson (2006) permission pending.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 59 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
7. RECLAMATION, MITIGATION AND RECOVERY
Since 1975, surface mining and reclamation have produced significant changes in the
topography, hydrology, vegetation, groundwater and wildlife of the central Appalachian Plateau
(Loveland et al., 2003; U.S. EPA, 2003, 2005) and have been the dominant driver of land
cover/land use change in the Central Appalachian Coalfields (Townsend et al., 2009). Hooke
(1999) identified MTM-VF in eastern coalfields as the greatest contributor to earth-moving
activity in the United States (see Figure 11).
Under the SMCRA (SMCRA, 30 U.S.C. § 1231, et seq.), reclamation is not considered
complete until water quality leaving the site complies with CWA standards without additional
treatment. Among other things, SMCRA and OSM regulations require that mine operators:
• minimize disturbances and adverse impacts on fish, wildlife, habitat and hydrologic
balance,
• recover the approximate original contour (AOC) and vegetation in mined areas and
• restore or approximate the original stream channels and riparian vegetation in permanent
constructed stream diversions (30 U.S.C. 1260 and 1265; 30 C.F.R. 816.43, cited in the
PE1S, Chapter II [U.S. EPA, 2003, 2005]).
In practice, the ecological conditions at reclaimed/mitigated mined sites vary with the
topography and geology of the location, the extent and scale of mining, excavation and
reclamation, the techniques used, state and federal regulations in place, goals for the postmining
land use, surface property ownership, time elapsed since reclamation and differences in local
reclamation or mitigation standards and requirements.
7.1. OVERVIEW OF RECLAMATION AND MITIGATION PRACTICES
7.1.1. Reclamation
Under the SMCRA, reclamation is required to restore lands disturbed by mining to their
premining conditions and land uses. Pre-SMCRA (pre-1977) mining practices left large areas of
unstable land and eroding hill slopes that impaired streams and created human health risks from
mudslides and pollution. That history, plus new concerns about the much larger volumes of
blasted rock and debris being produced by then-new mountaintop removal technology, led to
early SMCRA enforcement priority on stability and flood control. Reclamation techniques
developed prior to 2000 focused on regrading, soil compaction, fast-growing herbaceous
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 60 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
vegetation and stabilization, rather than reforestation or stream restoration, for protection of
water quality.
Since 2000, reclamation techniques have been developed that seek to restore at least
some of the productivity and ecological function of native forests, based in part on research and
extension efforts of the Appalachian Regional Reforestation Initiative (ARRI), a coalition of
groups formed to promote reforestation of eastern coal mine sites (http://arri.osmre.gov).
7.1.2. Reclamation Bonds
Activation of a mining permit requires posting of reclamation bonds to insure that the
coal operator will implement the approved reclamation plan, or provide funds for the government
to complete this work should a coal operator forfeit its responsibilities: (SMCRA, 30 U.S.C. §
1259).
Reclamation bonds are released upon inspection in three phases:
• Phase 1: released after backfilling, placement of homogenized or crush-rock topsoils
and contour regrading have been completed on a given area.
• Phase 2: released upon completion of revegetation activities.
• Phase 3 (final): released after the mine site has been inspected and accepted as being
satisfactorily reclaimed to the approved postmining land use (i.e., meets all
performance standards and the approved permit plan) (U.S. EPA, 2003, 2005).
Historically, release of bonds at a given site typically occurred within 5 years after
completion of reclamation and was based on percentage of land covered by fast-growing grasses
or legumes (Holl and Cairns, 1994; U.S. EPA, 2003, 2005). Reclamation to forested land uses,
which are preferred over pasture or hay land uses, may take longer but are more consistent with
the SMCRA and CWA (U.S. EPA, 2003, 2005; Skousen et al„ 2006).
7.1.3. Mitigation
USACE requirements on Section 404 CWA permits "strive for no net loss of aquatic
functions." This requirement to avoid, minimize, or compensate for unavoidable impacts to
waters of the United States has become an important economic factor in mining decisions (U.S.
EPA, 2003, 2005).
The USACE uses CWA Section 404(b)(1) to evaluate proposals to convert waters of the
United States to dry land (U.S. EPA, 2003, 2005). The preferred alternative is to avoid placing
fill in streams; if avoidance is not possible, fills must be minimized "to the extent practicable."
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 61 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
In either case, the proposal must not result in significant degradation to waters of the United
States. Unavoidable impacts to waters are allowed if they can be offset by mitigation that
compensates for the aquatic ecosystem functions lost in conversion (U.S. EPA, 2003, 2005).
Mitigation plans are developed prior to the start of mining and involve the use of stream
assessment methods to evaluate stream quality before and after impact. Compensatory
mitigation requirements are intended to minimize individual and cumulative mining impacts by
incentivizing applicants to avoid the delay and increased need for financial assurances or
bonding to guarantee that mitigation will be completed (U.S. EPA, 2003, 2005). Fills resulting
in "minimal impact, individually and cumulatively, including compensatory mitigation" as
determined by local regulatory agencies and the USACE, can be authorized by a Nationwide
Permit to expedite the review process. Fills causing more than minimal impact undergo a more
detailed individual review. Further, the cost of compensatory mitigation is higher, and permits
are at greater risk of being delayed or denied when "high value" aquatic resources are at stake.
Mitigation can be on-site or off-site of the permit area. On-site mitigation is preferred,
but off-site mitigation may be the only option for lost mountaintop ecosystems and is common in
MTM-VF permits. Compensatory mitigation plans include a variety of actions, including
• Riparian restoration (e.g., revegetation, wetland creation, floodplain connectivity).
• Creating channels to replace channels that have been filled using natural stream design
techniques.
• Enhancing or improving existing stream channels (e.g., riffles/pools, dredging, sinuosity,
bank stabilization).
• Sediment and pollution control (e.g., reclamation of abandoned mine lands and
remediation of other adverse environmental conditions within the watershed, anoxic
limestone drains, drums, flumes and other passive treatment systems).
• Reforestation of areas adjacent to mining sites.
• Improving fisheries habitat (e.g., shading, increasing habitat heterogeneity, aeration
through riffles or other natural means).
• Removing stream encroachments (e.g., roads, crossings, ponds, or other fills).
• Removing large woody debris.
We discuss only the practices and effectiveness of on-site mitigation for MTM-VF
disturbances. A discussion of off-site mitigation, including wetland creation, may be found in
Chapter III and Appendix D of the PEIS (U.S. EPA, 2003, 2005).
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 62 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
7.2. EVIDENCE OF RECOVERY: VEGETATIVE COVER
Vegetative cover at most reclaimed mine sites consists of rapidly growing grasses and
legumes, which for decades have served as the low cost, low-risk option for reclamation bond
release (Loveland et al., 2003). Even when permitted reclamation plans call for the planting of
trees, in response to SMCRA, reclamation activities typically result in excessive compaction of
the rooting medium, which reduces erosion but severely reduces tree growth (Holl, 2002;
U.S. EPA, 2003, 2005; Skousen et al., 2006).
Non-native grasses and legumes are used for reclamation at most sites (U.S. EPA, 2003,
2005, Table 3.J-1). Over time, native plant species do move onto reclaimed mine sites. In a
study of 18 reclaimed mine sites in Virginia, Holl and Cairns (1994) found that the total number
of plant species was increasing through time. However, succession rates were slower and plant
biodiversity greatly reduced relative to forests at nearby unmined sites. Of the 80 plant species
occurring in unmined forests, after 2-30 years, only 38 (47%) were found at one or more
mined/reclaimed sites, possibly due to the loss of seed banks in stripped topsoils.
Soil compaction and the initial planting of nonnative grasses interfere with the long-term
ecological recovery of mined sites. Holl (2002) resurveyed 15 mined sites and 5 logged sites
from the earlier study and found that the variety of plant species at the mined sites was more
similar to the unmined sites than during the first study. However, native plant biodiversity was
still substantially lower than in natural forests. Trees and woody shrub diversity was reduced by
45%, and herbaceous plant diversity was reduced by 56%, compared with forests at unmined
sites.
Pasture, grasses and hay lands planted to meet the legal requirements of reclamation may
be viable while maintained but may collapse when agronomic practices are neglected after
release of bond (Barnes, 2003). Reclaimed areas seeded with grasses and legumes plants will
eventually be recolonized by native plants as the nonnative grasses and legumes gradually
decline, but the time frame for recovery is not clear. Establishment of mid- to late-successional
trees may take decades (Skousen et al., 2009).
The combined effect of forest fragmentation and herbaceous reclamation planting may
undermine the ecological condition of unmined forest sections adjacent to mining sites as well.
Handel (2001) surveyed 55 mountaintop mining sites in southern West Virginia that were
reclaimed with herbaceous vegetation 6-24 years earlier and determined that both abundance
and species richness of trees and shrubs were extremely low on mine sites compared to
surrounding forests. Further, nonwoody plant species richness was reduced on plots adjacent to
mountaintop mining sites compared with interior forests. This edge effect extended from the
reclamation area 50 m into the adjacent forest.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 63 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
As a result of these studies and reports, reforestation of mined land is emerging as the "best
practices" postmining land use option, especially in mined areas that previously supported
high-quality natural forests (U.S. EPA, 2003, 2005). Although forestry and soil research is being
done at government, mining and university extension and experiment stations, little has been
published on the subject in the peer-reviewed literature. One exception is a study comparing
different conditions for the reforestation of reclaimed surface mined lands in the Appalachian
coalfields by Casselman et al. (2006), who found that tree survival and growth rate were affected
by differences in the composition of minesoils, the tree species planted and silvicultural methods
selected.
7.3. EVIDENCE OF RECOVERY: MINESOILS
Arguably the most limiting factors in the successful reclamation of vegetative cover at a
site are the density and composition of minesoils, Minesoils are the mixtures of soils, debris and
fractured rock overburden that are spread on reclaimed surfaces to support plant growth
(U.S. EPA, 2003, 2005). Natural soils develop over thousands of years as a result of physical
and chemical weathering and biological activity on parent rock substrates (Sencindiver and
Ammons, 2000). Minesoils are synthetic topsoils created by mixing natural topsoils and crushed
rock with soil amendments (fertilizers, biosolids, manure, coal combustion by-products, or
mulches) to enhance organic matter. Challenges to the recovery of native vegetative include loss
of natural soil structure and the loss of native seed pools and microbial communities in the
mining process.
Compaction of minesoils with the use of heavy equipment during valley fill and
reclamation is identified as one of the chief factors reducing the value of reclaimed forest lands.
Compacted soils limit root growth and greatly reduce the growth and survival of trees (e.g.,
Skousen et al., 2009). Soil research suggests that compaction below the plow layer may persist a
century or more (U.S. EPA, 2003, 2005). Because so much mine area has been reclaimed with
compacted soils and heavy herbaceous ground cover, converting such areas to a commercially
valuable forest requires expensive tilling techniques such as bulldozer "ripping" (for loosening
compacted minesoils), application of herbicides (to reduce the ground cover competition) and
use of weathered "brown" sandstone in substitute topsoils (Skousen et al., 2009).
Minesoils reflect the geochemical environment of the blasted rock used to make them and
the amendments added to them. Almost all synthetic or substitute soils at MTM-VF sites are
characterized by low soil organic carbon (SOC) content and low fertility. Most minesoils
develop weak to moderate structure in their surface horizons in as little as 2 years (Sencindiver
and Ammons, 2000). Rapid replenishment of soil carbon levels was documented by Simmons
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 64 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
et al. (2008), who estimate that normal levels of SOC could be achieved as quickly as 20-50
years after reclamation, although estimates from other studies are considerably longer (discussed
below).
Low SOC levels are correlated with high carbon sequestration rates in minesoils. Ussiri
and Lai (2005) reported that SOC sequestration rates in the top 15 cm of minesoil are high in the
first 20 to 30 years after reclamation, and that higher rates of SOC sequestration were observed
in soils (excluding aboveground and leaf-litter storage, where most carbon storage in forests
exists) at pasture and grassland mine sites than forested sites. They propose that minesoils have
potential for resequestering some of the carbon (C) lost through the soil disturbance associated
with deforestation.
Hardwood forests sequester large amounts of carbon for long (>100-year) periods of time
(Ussiri and Lai, 2005). Unlike pasture and croplands, forests store large amounts of carbon in
litter layers and aboveground biomass, as well as soils. Sperow (2006) found that potential C
storage rates are highest in mined lands that are reclaimed directly to forests, as opposed to
pasture, crops, or natural forest succession. This study estimated potential C sequestration for
2 • •
11 km of mine permit land in seven states, including the Central Appalachian Coalfields, which
showed mining activity since 1992. Reclaiming all land directly to forests provided an average
total sequestration potential of 29.1 Tg C over a 20-year period, with the largest sequestration
potential in West Virginia (10.1 Tg C). West Virginia has high sequestration potential because
of the dense forest cover and large area of mining permit land in that state. By comparison,
average potentials for pasture and croplands over the 11-km2 area were 16.3 and 10.9 Tg C over
20 years. Although reclaimed forests do not mitigate for the CO2 emissions of deforestation,
especially in mature hardwood forests, differences in reclamation practices can significantly
affect projected greenhouse gas emissions from mining regions (Sperow, 2006).
Amichev et al. (2008) measured carbon sequestration in mined and unmined forests in the
Midwestern and Appalachian coalfields. Unmined hardwood forests sequestered 62% more
carbon than mined forests of any type. Most of this difference (-60%) was in the soil carbon
content. Although carbon-poor minesoils have high potential for carbon sequestration, the
realization of this potential may have a long lag time (>60 years, with estimates ranging up to
760 years, reviewed in Amichev et al., 2008).
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 65 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
7.4. EVIDENCE OF RECOVERY: WATER QUALITY, STREAM ECOLOGICAL
FUNCTION, AQUATIC BIOTA
Sections 3, 4, 5, and 6 have reviewed the impacts of MTM-VF on water quality and
aquatic ecosystems. In this section, we provide an overview and discuss the effectiveness of
on-site reclamation activities.
7.4.1. Long-Term Effects on Downstream Water Quality
The field studies of downstream conditions reviewed in Section 4, Impacts on water
quality, were conducted 3 to 24 years after permitting, and field studies reviewed in Section 6,
Impacts on aquatic ecosystems were done 3 to 15 years after reclamation. The results indicate
that reclamation and mitigation, while providing stabilization, flood control and partial recovery
of terrestrial soils and vegetation, do not eliminate the deleterious effects of dissolved chemical
ions associated with effluent from MTM-VF.
7.4.2. Created Headwater Streams
Thousands of kilometers of headwater streams have been buried or removed by
mountaintop removal, and there is no substantive evidence in the literature or PEIS that onsite
mitigation by constructed channels or wetlands has replaced or will replace the lost functions and
biodiversity. Appalachian headwater streams are integrally associated with the geologic strata,
aquifers, subsurface flows, soil layers, surface topography and vegetation of the parent mountain
(see Figure 3). When a mountain is leveled, the geologic structures, groundwater exchanges,
hillslopes and subsurface flows that supported its small streams are permanently dismantled.
Stream restoration techniques were developed to restore one or more features of an existing
stream that has the basic structure of terrestrial and subsurface ecosystems intact, not to create
streams starting from scratch (Palmer, 2009).
The combination of ditches, groin ditches and sediment retention ponds are designed to
convey runoff for large (e.g., 100-year) storm events. No credible evidence was found to support
claims that the constructed channels and diverted surface flows on valley fills restore the
diversity and functions of natural headwaters lost to mining. Evidence that the constructed
channels and ponds are supporting any aquatic communities is scarce (Green et al., 2000), but
the small data sets available show that assemblages colonizing reclaimed surface waters differ
from natural headwater assemblages in predictable ways: loss of headwater-specific taxa;
increased ion- and sediment-tolerant taxa; and presence of taxa adapted to ponds or turbid,
slow-moving water (Kirk, 1999). Local habitat and biodiversity of stream ecosystems are
strongly influenced by the land use and land forms within the surrounding valley at multiple
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 66 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
scales (Allan, 2004), so constructed channels from which all terrestrial and subsurface
ecosystems have been removed are likely to differ from the natural systems they mimic.
With some exceptions, restoring ecosystems at mined sites to their premining condition is
not a realistic goal when mining methods destroy the underlying structure of those ecosystems.
To recover deep coal layers, mountaintop mining operations remove up to 1000 vertical feet of
unweathered rock by blasting. Much of this crushed rock is moved to the valleys to construct the
fills (see Figure 3). When mountaintop and coal removal are complete, the new landscape has
dramatically different elevations, contours and geology, all of which profoundly influence
ecological conditions at a site. Created channels on regraded contours provide channelized
flows, and some even provide limited habitat for aquatic invertebrates. However, surface and
subsurface conditions are so different that even with stream restoration techniques far better than
those available today, flow channels constructed at these sites cannot reasonably be expected to
restore or mitigate for the premining condition of aquatic life or water quality.
Flood control is an important function in the modified postmining landscape, as increased
stormflow volumes have been observed (U.S. EPA, 2003, 2005). Ferrari et al. (2009) modeled
the flood response in the 187-km2 mined watershed of Georges Creek in Western Maryland and
found parallels to what would be expected for urban areas with large areas of impervious surface.
Infiltration rates in reclaimed sites are typically 1 to 2 orders of magnitude smaller than for
undisturbed forest (Negley and Eshleman, 2006). Ferrari states that, "As a consequence, the act
of mine reclamation should not be interpreted as meaning the land is returned to a state that is the
hydrological equivalent of the premining landscape."
In a 2008 Federal Register notice, the USACE clarified the need to identify streams as
"difficult-to-replace" aquatic resources and acknowledged the need to avoid and minimize
impacts to them:
We recognize that the scientific literature regarding the issue of stream
establishment and re-establishment is limited and that some past projects have had
limited success (Bernhardt et al., 2007). Accordingly, we have added a new
paragraph at 33 C.F.R. 332.3(e)(3) [40 C.F.R. 230.93(e)(3)] that specifically notes
that there are some aquatic resources types that are difficult to replace and streams
are included among these. It emphasizes the need to avoid and minimize impacts
to these 'difficult-to-replace' resources and requires that any compensation be
provided by in-kind preservation, rehabilitation, or enhancement to the extent
practicable. This language is intended to discourage stream establishment and
re-establishment projects while still requiring compensation for unavoidable
stream impacts in the form of stream corridor restoration (via rehabilitation),
enhancement and preservation projects, where practicable [italics added for
emphasis].
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 ' 67 DRAFT—DO NOT CITE OR QUOTE
-------
1 Differences in the actual size of Appalachian headwaters (see Table 15) and the impact
2 threshold drainage area set for mitigation indicate that difficult-to-replace resources may be
3 overlooked in the planning stages of mine development. The mitigation threshold is determined
4 by local authorities and typically set to a drainage area ranging from 0.8-1.9 km2
5 (200-480 acres). Drainage areas for headwaters in the PEIS study range from 0.03-0.61 km2
6 (6-150 acres) (see Table 15).
7
8 Table 15. Drainage areas for headwaters in the PEIS study area (KY, TN,
9 WV, VA)
10
Median drainage area
Range of drainage areas
Headwater drainage type
km2 [acres]
km2 [acres] J
Ephemeral-intermittent
0.06 [14.5]
0.03 to 0.18 [6.3 to 45.3]
Intermittent-perennial
0.17 [40.8]
0.04 to 0.61 [10.4 to 150.1]
11
12 Source: Paybins (2003).
13
14
15
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 68 DRAFT—DO NOT CITE OR QUOTE
-------
1 Figure 11. Earth movement by humans and streams. Maps of the United
2 States showing, by variations in peak height, the rates at which earth is moved in
3 gigatonnes per annum in a grid cell measuring 1° (latitude and longitude) on a
4 side, by (A) humans and (B) rivers.
5 Source: Hooke (1999), used with permission from the publisher.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 69 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
8. SUMMARY, INFORMATION GAPS AND RESEARCH OPPORTUNITIES
8.1. A CONCEPTUAL MODEL OF THE IMPACTS OF MTM-VF
Mountaintop mining is a type of surface mining that is currently used, particularly in the
Central Appalachian Coalfields, to mine relatively low-sulfur coal. The subject of this review,
mountaintop removal mining, differs from other types of coal surface mining (e.g., contour strip
mining) in its scale. Whereas contour mining removes overburden along a bench cut into the
side of a mountain to expose a coal seam and then removes the coal, mountaintop removal
operations use explosives to remove all or portions of the entire mountain or ridge top, as much
as 300 vertical meters (1,000 vertical feet) of overburden, to expose and mine one or more coal
seams.
Aquatic ecosystems downstream of MTM-VF differ in significant ways from streams that
receive little human influence. Observations of which organisms are lost and how the streams
they live in are altered can improve our understanding of how MTM-VF result in these impacts.
Our review of the peer-reviewed, published literature and the PEIS is summarized here in a
conceptual model that traces the impacts of MTM-VF on aquatic ecosystems (see Figure 12).
Impacts begin with the preparation of the mountaintop or ridge top site. Access roads are
built, all trees are cleared and some topsoil is stockpiled for future use in reclamation. Then,
explosives are used to blast the top of the mountain or ridge to expose and mine one or more coal
seams. The coal that is removed is processed and transported to market; we did not encompass
the impacts of these processes in our review. Instead, we follow the fate and impacts of the
excess overburden and the mined site that remains.
When the mountaintop is removed, so are the intermittent streams, springs and small
perennial streams that comprise the headwaters of rivers. The excess overburden is disposed of
in constructed fills in small valleys or hollows adjacent to the mountaintop site. When trees are
removed from the valley fill footprint prior to construction of the fill, it also removes habitat for
amphibians that move between forest and stream during their life cycle.
When the valley fill is constructed, the headwaters beneath the footprint are buried, and
most organisms that lived there are killed. These headwaters support diverse biological
communities of aquatic invertebrates, such as insects, and vertebrates, including fish and
salamanders, that are often distinct from the species found in further downstream in the stream
system. Permanent headwater reaches can be spawning and nursery areas for native brook trout,
Salvelinus fontinalis. Intermittent and permanent headwater reaches, particularly those too small
to support fish, support numerous amphibian species. This particularly includes salamanders, of
which nearly 10% of the global diversity is found in streams of the southern Appalachians.
This document is a drajifor review purposes only and does not constitute Agency policy.
12/17/09 70 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
These streams are also habitat for diverse assemblages of Ephemeroptera, Plecoptera,
Trichoptera and other aquatic insects, many of which are unique to these headwater reaches.
Again, this biodiversity is lost when these headwater streams are buried under valley fills. The
hyporheic habitat is also buried, eliminating the interface of groundwater and surface water that
harbors the microbial community responsible for much of the nutrient processing and increasing
the export of nitrogen downstream.
As multiple streams in a mountain range are buried, the distance between the headwaters
that remain becomes greater. This hinders the movement of biota that is required to sustain
populations.
Both the water and sediment discharged into downstream ecosystems are altered by
MTM-VF. Water runoff is increased when the forest is cleared for the mine and valley fills. The
compacted, bare soils, which result from the removal of the overburden and coal, form a large
impervious surface that increases surface runoff. Depending on the degree to which they have
been compacted, the valley fills can ameliorate the effects of moderate precipitation events on
high flows, because they temporarily store water from surface flows and direct precipitation.
However, precipitation from more intense storms may produce greater storm flows, because
compaction of the fill surface and mined area by heavy equipment may reduce infiltration of
precipitation and increases overland runoff.
Surface runoff is diverted into ditches and sedimentation ponds, replacing natural
subsurface flow paths. Under most circumstances, the sedimentation ponds are effective at
settling out fine sediments. However, organisms that cling to rocks in riffles were observed to
decrease downstream of the ponds, a finding associated with increased fine sediments in other
ecosystems. The ponds themselves change the predominant source of energy in the downstream
systems from tree leaves to algae. Organisms that feed on leaves (shredders) decline; organisms
that feed on algae (filterers) increase.
The overburden in backstacks and valley fills has increased surface area available for
water contact with rock particles, and the water that emerges has higher concentrations of major
ions and some trace metals, including selenium. Native mayflies are consistently among the first
to disappear as concentrations of ions and trace metals increase. Most studies have found strong
negative correlations between the biotic metrics for fish or macroinvertebrates and specific
conductance, total dissolved solids, the concentration of individual ions, like SO42 , and other
measures of the elevated concentrations of various ions observed in streams below valley fills.
These studies have also generally found negative correlations between the various biotic metrics
and some measures of metals, but trace metals are generally less elevated in streams below
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 71 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
valley fills than in other mining situations and are usually correlated with greater increases in
other dissolved ions.
Discharges with high concentrations of ions reduce reproduction and survival as shown in
standard toxicity tests using organisms such as the crustacean Ceriodaphnia dubia.
Concentrations of selenium in some streams have been measured at levels that have been shown
to cause fish deformities and reduced fish reproduction in other streams. Fe and Mn deposits
have been observed on invertebrates, suggesting that concentrations are elevated under some
circumstances. Ni concentrations in sediments downstream of MTM-VF exceed empirical
screening values.
The loss of the headwaters has ramifications for the ecosystems downstream. The loss of
headwater invertebrate taxa removes a source of food. Fish that specialize in feeding on
invertebrates (the benthic invertivores) decrease downstream. Headwaters are active sites of
organic matter processing and nutrient uptake. The loss of headwater invertebrates and
microbial communities reduces dissolved carbon exports, an important food resource for
downstream biota, and increases nutrient loads in downstream waterbodies.
After the coal is removed, the extraction area and valley fill are graded and planted to
control sediment runoff. There is evidence that erosion is partially controlled and this mitigates
but does not completely eliminate the amount of fine sediments deposited downstream.
8.2. CONCLUSIONS
This section summarizes our major conclusions of potential consequences of MTM-VF in
six categories:
(1) Loss of headwater streams and forests
(2) Impacts on water quality
(3) Impacts on aquatic toxicity
(4) Impacts on aquatic ecosystems
(5) Cumulative impacts of multiple mining operations
(6) Effectiveness of mining reclamation and mitigation
We formed our conclusions by reviewing evidence from two sources of information:
(1) the peer-reviewed, published literature and (2) the PEIS and its associated appendices
(U.S. EPA, 2003, 2005). Only a few peer-reviewed papers were found that studied water quality
or stream ecosystems in headwaters or downstream of MTM-VF in the Central Appalachian
Coalfields. Our conclusions are based on evidence from these papers and relevant research
findings from laboratory studies and observational studies from other locations and mining
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 72 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
activities. We also discuss the findings published in the PE1S, which was published as two
separate documents; the Draft, published in 2003, and the Final, published in 2005.
8.2.1. Loss of Headwater and Forest Resources
Based on permits approved from 1992 to 2002, more than 1,900 km of headwater streams
are scheduled to be permanently lost or buried because of MTM-VF. These streams represent
more than 2% of the stream miles in the study area (KY, TN, WV, and VA), and their total
length is more than triple the length of the Potomac River. The area of valley fills based on
permit approved from 1985 through 2001 was 1774.4 km2. Additionally. 1,540 km2 of forests
were lost to MTM-VF from 1992-2002. More current statistics were unavailable at the time this
report was written. However, by 2012, the area that will have been deforested by MTM-VF in
the 4-state study area has been projected to be 5,700 km2, about 1.4 times the size of the state of
Rhode Island.
8.2.2. Impacts on Water Quality
Changes in water quality observed in streams downstream of MTM-VF include alteration
of flow and temperature regimes, increased fine sediments and increases in ions, some metals
and nitrogen.
Flows in streams below valley fills were generally more constant in both discharge and
temperature than in unimpacted streams. Valley fills influence downstream water quality by
acting like aquifers that store at least some of the water that enters from groundwater, surface
drainage, or direct precipitation. The removal of vegetation, particularly plants that have deep
roots, from the mined area and the area covered by the fills increases flow by decreasing
evapotranspiration.
Valley fills ameliorated the effects of moderate precipitation events on high flows, likely
because they temporarily store water from surface flows and direct precipitation. However, there
is some evidence that precipitation from more intense storms results in greater storm flows,
because of compaction of the fill surface and mined area by heavy equipment that reduces
infiltration of precipitation and increases overland runoff.
Effluent waters below valley fills were often alkaline. Most valley fills contain sufficient
carbonate minerals to neutralize the acid produced by pyrite oxidation that has historically
caused water quality problems from coal mining. In addition, the sediment retention ponds can
be used as treatment basins to neutralize pH. As a result, the metals that are not soluble under
higher pH conditions, such as Fe, Cd, Cr, Cu, Pb, Zn, and A1 were generally not elevated in
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 73 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
effluent waters below valley fills. Under some conditions, such as during higher flows,
particulate forms of less soluble metals, such as Fe, may be washed out of the valley fills.
Other ions were consistently observed at greatly elevated concentrations in the discharges
from valley fills. S042-, HCO3-, Ca2+, and Mg2+ are the dominant ions, but others include K+,
Na+, and CP. These ions all contribute to the elevated levels of total dissolved solids, typically
measured as specific conductivity observed in the effluent waters below valley fills. Selenium
concentrations are also elevated. Selenium can bioaccumulate through aquatic food webs, and
elevated levels have been found in fish tissues of the mining region. More than half of the sites
surveyed exceeded the chronic AWQC for selenium. Selenium reaches concentrations that have
been associated with effects in fish and birds in studies of mining effluents from other regions.
Despite the construction of sediment retention ponds below valley fills, several studies
found increased fine sediments in stream reaches downstream. Concentrations of NO3" plus
NO2'were also slightly elevated downstream.
8.2.3. Toxicity Impacts on Aquatic Organisms
Results of laboratory toxicity tests using the crustacean Ceriodaphnia dubia predict that
acute lethality will occur at the high end of specific conductivity observed downstream of
MTM-VF operations. This expectation was confirmed by reduced survival of Ceriodaphnia
observed in short-term tests using water from sites with high specific conductivity. Laboratory
tests of major ions reported effects on reproduction at concentrations 2—3 times lower than
effects on survival. Evidence from other alkaline mine effluents suggests that effects to
organisms should be expected at concentrations below those that affected Ceriodaphnia. Tests
using the mayfly Isonychia bicolor and the amphipod Hyalella azteca reported effects on
survival at concentrations 3-4 times lower than those affecting Ceriodaphnia. If effects on
reproduction in these organisms are similarly more sensitive than survival, effects from ions
would be expected at most sites downstream of MTM-VF.
Results from laboratory studies that varied the mixture of ions indicate that the interplay
among ions is complex. The ion mixture reported downstream of MTM-VF sites is dominated
by S042", Ca\ HC03", and Mg+. The low concentrations of CI" are likely making effects more
severe. The high overall hardness may be a mitigating factor. If the relative proportions of ions
are generally consistent downstream of MTM-VF in the Central Appalachian Coalfields, specific
conductivity (|iS/cm) may be the best surrogate indicator to use to predict when adverse effects
would occur.
Se concentrations reported from waters in the study area were high enough to warrant
concern. In some streams, they exceeded the chronic AWQC and fall in the range of
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 74 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
concentrations that have caused toxic effects on aquatic invertebrates, fish and birds in the field,
including in waters receiving valley fill and overburden dump leachates at coal mines in Canada.
Although the bioavailability of selenium is difficult to predict, measurements of selenium in fish
tissue indicated that the selenium is elevated in a form that is bioavailable.
Other toxicants were also high enough to warrant further investigation. Fe and Mn
deposits have been observed on macroinvertebrates. Ni and Zn concentrations in sediments are
higher than empirical screening level values.
8.2.4. Impacts on Aquatic Ecosystems
All surveys that used multimetric and aggregate taxonomic indices reported degraded
biological conditions in streams downstream of MTM-VF. Both fish and macroinvertebrate
communities were affected. Within the communities, changes were observed in organisms
grouped by food source: benthic macroinvertebrates that feed on leaf detritus declined, benthic
macroinvertebrates that feed on algae increased and specialist Fish that eat benthic
macroinvertebrates declined. Changes were also observed in organisms grouped by habit;
macroinvertebrates that cling to rocks in riffles declined. All studies showed a reduction in
mayflies. Trends observed for other taxonomic groups were less striking. Declines of the
aggregate indices and mayflies were most strongly correlated with increased concentrations of
ions and selenium.
8.2.5. Cumulative Impacts of Multiple Mining Operations
Few studies were found that investigated the cumulative impacts of multiple mining
operations. Specific conductivity and S04 levels were elevated in larger streams of the
Kanawha basin, downstream of multiple mining operations. Concentrations increased between
1980 and 1988 as more areas were mined. However, based on the results of one analysis, fish or
macroinvertebrate multimetric indices did not decline further as the number of upstream mines
or valley fills increased.
8.2.6. Effectiveness of Mining Reclamation and Mitigation
Nonnative grasses and legumes were used for reclamation at most of the mines studied.
Over time, the total number of native plant species increased but remained less than half the
diversity found at nearby unmined sites. Plant species richness in forests adjacent to mining sites
also remained lower than interior forests. By current reclamation standards, return of mature
forests to mined areas may require millennia due to soil compaction and loss of organic matter.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 75 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
The results of the water quality studies indicate that reclamation efforts partially
controlled the amount of soil erosion and fine sediments transported downstream. However,
there is no evidence that reclamation efforts altered or reduced the ions or toxic chemicals
downstream of valley fills. Ion concentrations have either remained constant or increased over
time. Given that the alterations of the stream ecosystems reported above were observed after
sites were reclaimed for 3 to 15 years, the observed effects would be expected to persist.
No published studies were found that investigated the replacement of stream function or
biodiversity in stream channels constructed on top of valley fills. However, constructed channels
at locations experiencing far less disturbance than mountaintop mining are consistently found to
differ significantly from natural systems in biota and functions.
8.3. INFORMATION GAPS AND RESEARCH OPPORTUNITIES
The evidence in our review is consistent enough that we have a high degree of confidence
in our conclusions. Still, our review uncovered a number of information gaps that could be filled
by research. Filling these can improve our quantitative understanding of how MTM-VF impacts
aquatic ecosystems, potentially leading to more effective regulatory and mitigation approaches.
Future assessments should consider the comparative risks of MTM-VF. However, the
comparisons should be to real alternatives that might be implemented by real decisions. For
example, risks from MTM-VF might be compared with those associated with other coal sources,
or risks associated with electricity generated by burning coal from MTM-VF might be compared
to those associated with sources of electricity other than coal combustion. Alternatively, if a
land use decision is being made, MTM-VF risks might be compared with other uses of mountain
tops and headwater streams such as logging or tourism.
8.3.1. Update the MTM-VF Inventory and Surveys of Impact Extent
The most recent data available in the published literature on the extent and potential
additional development of MTM-VF mines in the Central Appalachian Coalfields are those
compiled for the PEIS (U.S. EPA, 2003, 2005). These data were only for MTM-VF mines
developed between 1985 and 2002, when at least some mines had been developed as early as
1967 (U.S. EPA, 1979), and permitting and construction of MTM-VF have continued since then.
Therefore, the inventory of filled valleys and of stream miles buried by those valley fills should
be updated. Moreover, the inventory should be adjusted to reflect the actual extent of these
valley fills versus their permitted extent. Remote sensing and GIS, combined with field
sampling, would make this possible (Townsend et al., 2009). The updated inventory of
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 76 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
MTM-VF can be used to design a statistically robust estimate of the extent of impacts within the
region, based on a probabilistic sampling design.
8.3.2. Quantify the Contributions of Headwater Streams
It would be desirable to more fully understand the role of the headwater streams buried
by valley fills in the retention and cycling of nutrients, such as nitrogen and phosphorus, and the
downstream transport of trophic resources, such as allochthonous organic matter (i.e., leaf litter,
small particulate organic matter produced from the leaf litter, and dissolved organic carbon),
algae and animal prey. This understanding would allow assessors to better understand and
model the cumulative effects of burying these headwater streams on stream function (i.e.,
nutrient transport and cycling, processing and transport of organic matter) and other ecosystem
services supplied by these stream systems.
It would be desirable to more fully understand the metapopulation and metacommunity
linkages among different headwater streams and between these headwaters and downstream
reaches. This information would increase understanding of the effects of burying these
headwater streams on regional biodiversity, including the cumulative effects of this practice of
burying headwater drainages in this region of the Appalachians.
8.3.3. Improve Understanding of Causal Linkages
Our understanding of the causal linkages between MTM-VF and stream ecosystems
could be improved by bringing together additional data. Sources of data include reports that we
were unable to include in this report because we could not confirm that they had been peer-
reviewed and additional monitoring data that may also be available from various states,
particularly West Virginia and Kentucky. Questions that might be answered include
(1) At what concentrations of major ions and trace metals do different taxa disappear?
(2) Which downstream organisms in addition to Ephemeroptera are most affected by valley
fills?
(3) How do these effects differ among different insect orders and between insects and
noninsect aquatic taxa?
(4) How do the species within these large orders change? Some evidence indicates that
headwater species are replaced by more downstream species below valley fills.
(5) Are there observable effects on individuals of sensitive taxa?
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 77 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
When selecting such monitoring data, care should be taken that the sampling was timed
so that the common species can be sampled if they are present. This is not a concern for fish, but
many macroinvertebrates are present for part of the year as eggs or larval instars that are too
small to be sampled by the standard net mesh sizes used to sample benthic macroinvertebrates.
8.3.4. Develop Tests Using Sensitive Taxa
Although the field observations of taxa declines agree qualitatively with the toxicity test
results of MTM-VF effluent using standard laboratory organisms, effects to some native
organisms appear to be occurring at lower concentrations. Quantitative estimates of the
concentrations at which effects occur could be improved by testing effluents using a life-cycle
test, especially with vertebrate and invertebrate species found in these headwater streams.
For invertebrates, we would recommend an Ephemeroptera species or a physiologically similar
aquatic insect. An example of a full life cycle with a species of Ephemeroptera is described by
Sweeney et al. (1993) and Conley et al. (2009). Tests using these insects would help verify that
the differences in sensitivity between laboratory tests using Ceriodaphnia dubia and field
observations of Ephemeroptera declines are due to differences in sensitivity to the ions, rather
than a confounding factor. For fish and amphibians, it would be desirable to perform
reproductive toxicity tests with waters like those found below valley fills using headwater taxa,
such as dace, brook trout, or sculpins.
The vertebrate and invertebrate fauna found in headwater streams of the southern
Appalachian Mountains are adapted to waters characterized by low hardness, total dissolved
solids, ionic strength, conductivity and alkalinity and by neutral to slightly acidic pH. The
streams below valley fills are altered such that the waters are characterized by high hardness,
total dissolved solids, ionic strength, conductivity and alkalinity and slightly alkaline pH. These
waters also have relatively high concentrations of individual ions, such as SC>42~, HCC>3~, and K+.
These multiple changes in the dissolved constituents in these waters are likely to have interactive
effects on aquatic fauna and are not duplicated well by any laboratory test data found in the
published literature. Moreover, the species and life stages used in the laboratory tests found in
the published literature differ from the native fauna of these streams.
Most of the invertebrates that have been used in laboratory toxicity tests of the effects of
t f t t ... 2 • •
conductivity, total dissolved solids, or the individual effects of SO4 and other dissolved ions are
Crustacea. This includes the cladocerans, Ceriodaphnia dubia and Daphnia magna, and the
amphipod, Hyalella azteca. These Crustacea have very different evolutionary histories
compared to the aquatic insects that dominate the headwater streams. In the case of Crustacea
and other wholly aquatic groups like Mollusca, their evolutionary ancestors moved directly into
freshwater environments from marine or estuarine environments (Thorp and Covich, 1991).
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 78 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Some species, such as Daphnia magna (Martinez-Jeronimo and Martinez-Jeronimo, 2007), have
populations found in brackish waters. Aladin and Potts (1995) describe Cladocera as strong
osmoregulators. Hence, in addition to not being found in these headwater streams, the standard
invertebrate test species do not appear to be sensitive to the sorts of major ions leaching from
valley fills.
The evolutionary ancestors of insects moved from marine or estuarine environments into
terrestrial environments. Then, in turn, the evolutionary ancestors of aquatic insects, such as
Ephemeroptera, Plecoptera, Trichoptera, and Odonata moved from terrestrial environments to
freshwater environments (Merritt and Cummins, 1996). As a result, aquatic insects possess very
different mechanisms for osmotic regulation compared to the wholly aquatic groups. In the
aquatic insects found in these streams, osmotic regulation is accomplished in part by tissues
called chloride cells or chloride epithelia, which are involved in ion absorption, an important
adaptation in the low ionic strength, freshwater habitats where aquatic insects are found
(Kominick, 1977). In addition, the insects differ from the test species in that their eggs develop
externally, so they are directly exposed to contaminated waters. This suggests that Crustacea are
not appropriate surrogates for these aquatic insects in laboratory toxicity tests, particularly those
that test the effects of the alterations in water chemistry associated with valley fills.
Even insect species, like Chironomus teutons and lsonychia bicolor, may not be good
surrogates. Aquatic Diptera possess anal papillae, which though different in structure are
functionally equivalent to chloride epithelia. In the case of lsonychia bicolor, its natural
distribution is in larger streams with higher alkalinities and conductivities than those found in the
streams affected by MTM-VF (Kondratieff and Voshell, 1984). Also, the bioassays testing
lsonychia bicolor used relatively large (~ 9 to 14 mm in length), late instar nymphs in 7-day
tests, where molting and survival were the only measurement endpoints. Other life stages and
measurement endpoints appropriate to the survival of these mayfly populations could be more
sensitive to the chronic stresses imposed by the observed changes in water chemistry.
8.3.5. Further Investigate Selenium and Sediments
Aqueous selenium concentrations and concentrations in fish fall within a range that can
cause effects on fish and fish-eating birds. Additional analyses, including possibly a study of
stream-based food webs, could better define the extent of this problem, and reproductive tests of
fish collected in high-selenium streams could better define the nature of the problem. To
confirm effects of Se on fish reproduction, fish would be collected from high Se streams and
spawned in the laboratory. This would be required because Se acts by bioaccumulation in the
females and transfer to the eggs. Little is known about the effects of selenium on stream
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 79 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
invertebrates. Analyses of invertebrates from high-selenium streams and reproductive tests
could determine whether selenium is contributing to observed effects.
Few data are available concerning the effects of MTM-VF on the chemical quality of
sediments in streams below valley fills. While dissolved trace metals in effluent waters below
valley fills appear to be low, there is evidence along with geochemical theory that particulate
metals should be produced within valley fills and may under some conditions be flushed
downstream. Also, there are some metals (i.e., Mn, Ni) whose solubilities are not affected by pH
and whose dissolved concentrations maybe somewhat elevated in effluent waters. Therefore,
data on sediment concentrations of metals could be used to assess whether sediment
contamination may be a concern associated with MTM-VF. Observations could also determine
if effects associated with the deposition of particulate metals occur. These effects could be
similar to those observed with iron hydroxides in more acidic situations. To completely assess
this exposure pathway, such sampling could include measurement of pore water concentrations
of the dissolved metals and ammonia or use techniques such as simultaneously extracted metals
and acid volatile sulfide.
8.3.6. Quantify Cumulative Effects
Additional studies explicitly designed to quantify the cumulative effects of MTM-VF
would help differentiate those effects from the other land uses in the Central Appalachian
Coalfields, such as abandoned mines, oil and gas development and residential development.
Additional water chemistry sampling, combined with spatial analyses of the number and volume
of valley fills, could reveal how specific conductivity and other measures of the dissolved ions
increase as the percentage of the watershed in valley fills increases and how export of dissolved
ions changes with time after the creation of a valley fill.
Currently, little is known about the cumulative effects of incremental loss of headwater
streams and naturally occurring mountain aquifers on the region's hydrology and water supplies.
Given the huge scales at which landscape disturbances above and below ground occur in coal
mining areas, groundwater sampling, tracer studies and surface flow-groundwater interaction
monitoring, in addition to water quality sampling, may be needed.
8.3.7. Quantify Functionality of Constructed Streams
Finally, although there is a large body of literature on stream restoration ecology in urban
and agricultural streams, we found there is a lack of evidence on the biota and ecosystem
functioning associated with the constructed sediment and flow control channels on valley fills. If
these streams are argued to mitigate the effects of stream burial, the type and degree of
mitigation should be quantified.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 80 DRAFT—DO NOT CITE OR QUOTE
-------
forest clearing
overburden handling
backstacks
d soils
t stream
burial
j water contact
with overburden
| riparian cover
overburden
composition
f erosion of
fill material
| fertilizer
application
| groundwe
water con
I headwater habitat
i hyporheic
habitat
ApH
(water)
t total dissolved solids
(water)
t metals
(water)
t metals
(sediment)
sediment retention
& treatment ponds
1 headwater
_ taxa _
distance from
orest to stream
ay input
leaves
T energy input
from algae
from
LJ
t distance to other
headwater populations
f hardness
J fish
formitii
I amphibians &
macroinvertebrates
t armoring of
aquatic organisms
t temperature
1 survival, growth &
reproduction in toxicity tests
metal-sensitive fish
macroinvertebrates
1 ion-sensitive fish
& macroinvertebrates
f filterers
nutrie
I bioticindexquaiity^)
tl model of the impacts of mountaintop mines and valley fills on aquatic ecosystems (narrative description in Section 8.1).
-------
1 LITERATURE CITED
2
3
4 Ahrens, MJ; Morrissey, DJ. (2005) Biological effects of unburnt coal in the marine environment. Oceanogr Marine
5 Biol Ann Rev 43:69-122.
6 Aladin, NV; Potts, WTW. (1995) Osmoregulatory capacity ofthe Cladocera. J Comp Physiol B Biochem Syst
7 Environ Physiol 164(8):671-683.
8 Alexander, RB; Smith, RA; Schwarz, GE. (2000) Effect of stream channel size on the delivery of nitrogen to the
9 Gulf of Mexico. Nature 403(6771) :758-761.
10 Alexander, RB: Boyer, EW; Smith, RA; et al. (2007) The role of headwater streams in downstream water quality. J
11 Am Water Res Assoc 43( 1 ):41 —59.
12 Allan, J.D. (2004) Landscapes and riverscapes: the influence of land use on stream ecosystems. Annu Rev Ecol
13 Evol Syst 35:257-284.
14 Altschuler, ZS; Schnepfe, MM; Silber, CC; et al. (1983) Sulfur diagenesis in Everglades peat and origin of pyrite in
15 coal. Science 221(4607):221-227.
16 Amichev, BY: Burger, JA; Rodrigue, JA. (2008) Carbon sequestration by forests and soils on mined land in the
17 Midwestern and Appalachian coalfields of the US. Forest Ecol Manage 256(11): 1949-1959.
18 Angermeier, PL: Smogor, RA; Stauffer, JR. (2000) Regional frameworks and candidate metrics for assessing biotic
19 integrity in mid-Atlantic highland streams. Trans Am Fish Soc 129:962-981.
20 Angradi, TR. (1997) Hydrologic context and macroinvertebrate community response to floods in an Appalachian
21 headwater stream. Am Midi Nat 138:371-386.
22 Angradi, T; Hood, R; Tarter, D. (2001) Vertical, longitudinal and temporal variation in the macrobenthos of an
23 Appalachian headwater stream system. Am Midi Nat 146(2):223—242.
24 Banks, D; Younger, PL; Arnesen, RT; et al. (1997) Mine-water chemistry: the good, the bad and the ugly. Environ
25 Geol 32(3): 157— 174.
26 Barbour, MT; Gerritsen, J; Snyder, BD; et al. (1999) Rapid bioassessment protocols for use in streams and wadeable
27 rivers: periphyton, benthic macroinvertebrates, and fish. Second Edition., U.S. Environmental Protection Agency ,
28 Office of Water, Washington, DC; EPA/84 l/B-99/002.
29 Barnes, RF; Nelson, CJ; Collins, M; et al.; eds. (2003) Forages: an introduction to grassland agriculture. 6th edition,
30 Vol 1. Ames, I A: Iowa State Press.
31 Barros, AP; Lettenmaier, DP. (1994) Dynamic modeling of orographically induced precipitation. Rev Geophys
32 32:221-227.
33 Benda, L; Hassan, MA; Church, M: et al. (2005) Geomorphology of steepland headwaters: the transition from
34 hillslopes to channels. J Am Water Resour Assoc 41 (4):835—851.
35 Benson, MA; Dalrymple, T. (1967) General field and office procedures for indirect discharge measurements.
36 Chapter Al. In: Techniques of water resources investigations, book 3: applications of hydraulics. Department of
37 Interior, U.S. Geological Survey, Alexandria, VA. Available online at http://pubs.usgs.gov/twri/.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 82 DRAFT—DO NOT CITE OR QUOTE
-------
1 Bernhardt, ES; Likens, GE; Hall, RO; et al. (2005) Can't see the forest for the stream? In-stream processing and
2 terrestrial nitrogen exports. Bioscience. 55(3):219—230.
3 Bernhardt, ES; Sudduth, EB; Palmer, MA; et al. (2007) Restoring rivers one reach at a time: Results from a survey
4 of U.S. river restoration practitioners. Restor Ecol 15(3):482-493.
5 Bodkin, R; Kern, J; McClellan, P; et al. (2007) Limiting total dissolved solids to protect aquatic life. J Soil Water
6 Conserv 62(3):57A—61 A.
7 Boulton, AJ; Findlay, S; Marmonier, P; et al. (1998) The functional significance of the hyporheic zone in streams
8 and rivers. Annu Rev Ecol Syst 29:59-81.
9 Brown, JH; Kodric-Brown, A. (1977) Turnover rates in insular biogeography - effect of immigration on extinction.
10 Ecology 58(2):445-449.
1 1 Bryant, G; McPhilliamy, S; Childers, H. (2002) A survey of the water quality of streams in the primary region of
12 niountaintop / valley fill coal mining, October 1999 to January 2001. In: Draft programmatic environmental impact
13 statement on mountaintop mining / valley fills in Appalachia - 2003. Appendix D. U.S. Environmental Protection
14 Agency, Region 3, Philadelphia, PA. Available online at
15 http://www.epa.gOv/Region3/mtntop/pdf/appendices/d/stream-chemistry/MTMVFChemistryPart 1 .pdf.
16 Carlson, RM; Oyler, AR; Gerhart, EH; et al. (1979) Implications to the aquatic environment of polynuclear aromatic
1 7 hydrocarbons liberated from northern Great Plains coal. U.S. Environmental Protection Agency, Office of Research
18 and Development, Environmental Research Laboratory, Duluth, MN: EPA 600/3-79/093. Available online from the
19 National Technical Information Service, Springfield, VA, PB 80-115967.
20 Caruccio, FT; Ferm, JC; Home, J; et al. (1977) Paleoenvironment of coal and its relation to drainage quality. U.S.
21 Environmental Protection Agency, Office of Research and Development, Industrial Environmental Laboratory,
22 Cincinnati, OH. EPA 600/7-77/067. Available online from the National Technical Information Service, Springfield,
23 Va, PB-270 080/5.
24 Casagrande, DJ. (1987) Sulphur in peat and coal. Geol Soc SP 32:87-105.
25 Casselman, CN; Fox, TR; Burger, JA; et al. (2006) Effects of silvicultural treatments on survival and growth of trees
26 planted on reclaimed mine lands in the Appalachians. For Ecol Manage 223(1-3):403-414.
27 Chapman, PM; Downie, J; Maynard, A; et al. (1996) Coal and deodorizer residues in marine sediments:
28 contaminants or pollutants? Environ Toxicol Chem 15(5):638-642.
29 Chapman, PM; Bailey, H; Canaria, E. (2000) Toxicity of total dissolved solids associated with two mine effluents to
30 chironomid larvae and early life stages of rainbow trout. Environ Toxicol Chem 19:210-214.
31 Clark, J; Clements, WH. (2006) The use of in situ and stream microcosm experiments to assess population- and
32 community-level responses to metals. Environ Toxicol Chem 25(9):2306-2312.
33 Clarke, A; Mac Nally, R; Bond, N; et al. (2008) Macroinvertebrate diversity in headwater streams: a review.
34 Freshw Biol 53:1707-1721.
35 Collins, BM; Sobczak, WV; Colburn, EA. (2007) Subsurface flowpaths in a forested headwater stream harbor a
36 diverse macroinvertebrate community. Wetlands 27(2):319-325.
37 Conley, JM; Funk, DH; Buchwalter, DB. (2009) Selenium bioaccumulation and maternal transfer in the mayfly
38 Centrnptilum triangulifer in a life-cycle, periphyton-biofilm trophic assay. Environ Sci Technol 43:7952-7957.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 83 DRAFT—DO NOT CITE OR QUOTE
-------
1 Cooper, DM; Jenkins, A; Keffington, R; et al. (2000) Catchment-scale simulation of stream water chemistry by
2 spatial mixing: Theory and application. J Hydrol 233:121-137.
3 Cummins, KW; Wilzbach, MA; Gates, DM; et al. (1989) Shredders and riparian vegetation - leaf litter that falls into
4 streams influences communities of stream invertebrates. Bioscience 39:24-30.
5 Davie, R; Welsh, HH. (2004) On the ecological roles of salamanders. Annu Rev Ecol Evol Syst 35:405-434.
6 Davies, SP; Jackson, SK. (2006) The biological condition gradient: A descriptive model for interpreting change in
7 aquatic ecosystems. Ecol Appl 16(4): 1251 —1266.
8 DeBruyn, AMU; Chapman, PM. (2007) Selenium toxicity to invertebrates: will proposed thresholds for toxicity to
9 fish and birds also protect their prey? Environ Sci Technol 41(5): 1766-1770.
10 Di Toro, DM; Allen, HE; Bergman, HL; et al. (2001) Biotic ligand model of the acute toxicity of metals. 1.
11 Technical basis. Environ Toxicol Chem 20(10):2383-2396.
12 Dickens, PS; Minear, RA; Tschantz, BA. (1989) Hydrologic alteration of mountain watersheds from surface mining.
13 J Water PollutCont Fed 61:1249-1260.
14 Doig, LE; Eiber, K. (2006) Nickel partitioning in formulated and natural freshwater sediments. Chemosphere
15 62:968-979.
16 Downing-Kunz, MA; Unthank, MD; Crain, AS. (2005) Compilation of concentrations of total selenium in water,
17 coal in bottom material, and field measurement data for selected Streams in eastern Kentucky, July 1980 Open-File
18 Report 2005-1354. Department of Interior, U.S. Geological Survey, Reston, VA. Available online at
19 http://www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA442142&Location=U2&doc=GetTRDoc.pdf.
20 Ewers, RM; Didham, RK. (2006) Confounding factors in the detection of species responses to habitat fragmentation.
21 Biol Rev 81:1 17-142.
22 Feminella, JW. (1996) Comparison of benthic macroinvertebrate assemblages in small streams along a gradient of
23 How permanence. J N Am Benthol Soc 15(4):651 —669.
24 Ferrari, JR; Lookingbill, TR; McCormick, et al. (2009) Surface mining and reclamation effects on flood response of
25 watersheds in the central Appalachian Plateau region. Water Resour Res 45(4):W04407.
26 Freeman, MC; Pringle, CM; Jackson, CR. (2007) Hydrologic connectivity and the contribution of stream headwaters
27 to ecological integrity at regional scales. J Am Water Resour Assoc 43:5-14.
28 Fulk, F; Autrey, B; Hutchens, J; et al. (2003) Ecological assessment of streams in the coal mining region of West
29 Virginia using data collected by the U.S. EPA and environmental consulting firms. In: Mountaintop mining/valley
30 fills in Appalachia. Final programmatic environmental impact statement. U.S. Environmental Protection Agency,
31 Region 3, Philadelphia, PA. Appendix D. Available online at http://www.epa.gov/Region3/mtntop/pdf/mtm-
32 vf fpeis_full-document.pdf.
33 Gomi, T; Sidle, RC; Richardson, JS. (2002) Understanding processes and downstream linkages of headwater
34 systems. Bioscience 52(10):905-916.
35 Google maps. (2009) Satellite images of Hobet 21 mine complex and Washington DC. Available online at
36 http://maps.google.com (accessed 11/19/2009).
37 Grant, EHC; Green, LE; Lowe, WH. (2009) Salamander occupancy in headwater stream networks. Freshw Biol
38 54(6): 1370-1378.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 84 DRAFT—DO NOT CITE OR QUOTE
-------
1 Green, J; Passmore, M; Childers, H. (2000) A survey of the conditions of stream in the primary region of
2 mountaintop mining/valley fill coal mining. Mountaintop mining/valley fills in Appalachia. Final programmatic
3 environmental impact statement. U.S. Environmental Protection Agency, Region 3, Philadelphia, PA. Appendix D.
4 Available online at http://www.epa.gOv/Region3/mtntop/pdf/appendices/d/streams-invertebrate-study/FINAL.pdf.
5 Green, NB; Pauley, TK. (1987) Amphibians and reptiles in West Virginia. Pittsburgh, PA: University of Pittsburg
6 Press.
7 Greenwood, JL; Rosemond. AD. (2005) Periphyton response to long-term nutrient enrichment in a shaded
8 headwater stream. Can J Fish Aqua Sci 62(9):2033-2045.
9 Gulis, V; Kuehn, KA Suberkropp, K. (2006) The role of fungi in carbon and nitrogen cycles in freshwater
10 ecosystems. In: Gadd; GM; ed. Fungi in biogeochemical cycles. Cambridge University Press: Cambridge, UK. pp
1 ] 404-435.
12 Gulis, V; Suberkropp, K. (2004) Effects of whole-stream nutrient enrichment on the concentration and abundance of
1 3 aquatic Hypohomycete conidia in transport. Mycologia 96:57-65.
14 Hakala, JP; Hartman, KJ. (2004) Drought effect on stream morphology and brook trout (Salvelinus fontinalis)
15 populations in forested headwater streams. Hydrobiologia 515:203-213.
16 Handel, SN. (2001) Terrestrial plant (spring herbs, woody plants) populations of forested and reclaimed sites. In:
17 Draft programmatic environmental impact statement on mountaintop mining / valley fills in Appalachia - 2003.
18 Appendix E. U.S. Environmental Protection Agency, Region 3, Philadelphia, PA. Available online at
19 http://www.epa.gov/Region3/mtntop/eis2003appendices.htni.
20 Harding, LW; Graham, M; Paton, D. (2005) Accumulation of selenium and lack of severe effects on productivity of
21 American dippers (Cinclus mexicanus ) and spotted sandpipers (Actitis macularia). Arch Environ Contain Toxicol
22 48:414-423.
23 Harper, KA; Macdonald, SE; Burton, PJ; et al. (2005) Edge influence on forest structure and composition in
24 fragmented landscapes. Conserv Biol 19(3):768—782.
25 Hartman, KJ; Kaller, MD; Howell, JW; et al. (2005) How much do valley fills influence headwater streams?
26 Hydrobiologia 532:91-102.
27 Hewlett, JD. (1982) Principles of forest hydrology. Athens, GA: The University of Georgia Press.
28 Holl, KD. (2002) Long-term vegetation recovery on reclaimed coal surface mines in the eastern USA. J Appl. Ecol
29 39(6):960-970.
30 Holl, KD; Cairns. J. (1994) Vegetational community-development on reclaimed coal surface mines in Virginia. Bull
31 Torrey Bot Club 121:327-337.
32 Hooke, RL. (1999) Spatial distribution of human geomorphic activity in the United States: Comparison with rivers.
33 Earth Surf Proc Land 24(8):687-692.
34 Houser, JN; Mulholland, PJ; Maloney, KO. (2005) Catchment disturbance and stream metabolism: patterns in
35 ecosystem respiration and gross primary production along a gradient of upland soil and vegetation disturbance. J N
36 Am Benthol Soc 24(8):538-552.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 85 DRAFT—DO NOT CITE OR QUOTE
-------
1 Howard, HS; Berrang, B; Flexner, M; et al. (2001) Kentucky mountaintop mining benthic macro invertebrate survey:
2 Central Appalachian Ecoregion, Kentucky. In: Draft programmatic environmental impact statement on mountaintop
3 mining / valley fills in Appalachia - 2003. Appendix D. U.S. Environmental Protection Agency, Region 3,
4 Philadelphia, PA. Available online at http://www.epa.gov/Region3/mtntop/pdf/appendices/d/kentucky-
5 macroinvertebrate-study/report.pdf.
6 Jennings, G. and W. Harman. (2002) Natural stream processes. River Course, Fact Sheet No. 1. North Carolina
7 Cooperative Extension Service, North Carolina State University, Raleigh, NC. Available online at
8 http://www.bae.ncsu.edu/programs/extension/wqg/sri/rv-crs-l ,pdf (accessed 11/16/2009).
9 Johnson, BR; Fritz, KM; Blocksom, KA; et al. (2009) Larval salamanders and channel geomorphology are
10 indicators of hydrologic permanence in forested headwater streams. Ecol Indica 9:150-159.
11 Karr, JR. (1981) Assessment ofbiotic integrity using fish communities. Fisheries 6:21-27.
12 Kaufmann, PR; Robinson, EG. (1998) Physical habitat characterization. In: Surface waters: field operations and
13 methods for measuring the ecological condition of wadeable streams. U.S. Environmental Protection Agency,
14 Environmental Monitoring and Assessment Program, Washington, DC. EPA/620/R-94/004.
15 Kefford, BJ; Papas, PJ; Nugegoda, D. (2003) Relative salinity tolerance of macroinvertebrates from the Barwon
16 River, Victoria, Australia. Mar Freshw Res 54(6):755~765.
17 Kefford, BJ; Papas, PJ; Metzeling, L; et al. (2004) Do laboratory salinity tolerances of freshwater animals
18 correspond with their field salinity? Environ Pollut 129(3):355-362.
19 Kennedy, AJ; Cherry, DS; Currie, RJ. (2003) Field and laboratory assessment of a coal processing effluent in the
20 Leading Creek Watershed, Meigs County, Ohio. Arch Environ Contam Toxicol 44(3):324-331.
21 Kennedy, AJ; Cherry, DS; Currie, RJ. (2004) Evaluation of ecologically relevant bioassays for a lotic system
22 impacted by a coal-mine effluent, using Isonychia. Environ Monit Assess 95:37—55.
23 Kennedy, AJ; Cherry, DS; Zipper, CE. (2005) Evaluation of ionic contribution to the toxicity of a coal-mine effluent
24 using Ceriodaphnia dubia. Arch Environ Contam Toxicol 49:155-162.
25 Ketola, HG; Longacre, D; Greulich, A; et al. (1988) High calcium-concentration in water increases mortality of
26 salmon and trout eggs. Prog Fish-Cult 50:129-135.
27 Kiffney, PM; Clements, WH. (1996) Size-dependent response of macroinvertebrates to metals in experimental
28 streams. Environ Toxicol Chem 15(8): 1352-1356.
29 Kimball, BA; Callender, E; Axtmann, EV. (1995) Effects of colloids on metal transport in a river receiving acid
30 mine drainage, upper Arkansas River, Colorado, U.S.A. Appl Geochem 10(3):285—306.
31 Kirk, E. (1999) An evaluation of the aquatic habitats provided by sediment control ponds and other aquatic
32 enhancement structures located on mine permitted areas in southern West Virginia. In: Draft programmatic
33 environmental impact statement on mountaintop mining/valley fills in Appalachia - 2003. Appendix D. U.S.
34 Environmental Protection Agency, Region 3, Philadelphia, PA. Available online at
35 http://www.epa.gOv/region03/mtntop/pdf/appendices/d/aquatic-ecosystem-enhanC"
36 symp/Proceedings/support/Maggard/PondStudy2.pdf.
37 Kominick, H. (1977) Chloride cells and chloride epithelia of aquatic insects. Int J Cytol 49:285-328.
38 Kondratieff, BC; Voshell, JR. (1984) The North and Central American species of Isonychia (Ephemeroptera:
39 Oligoneuriidae). Trans Am Entomol Soc 110:129-144.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 86 DRAFT—DO NOT CITE OR QUOTE
-------
1 Larsen, D; R. Mann, R. (2005) Origin of high manganese concentrations in coal mine drainage, eastern Tennessee.
2 J Geochem Explor 86:143-163.
3 Lasier, PJ; Winger, PV; Reinert, RE. (1997) Toxicity of alkalinity to Hyalella azteca. Bull Environ Contain Toxicol
4 59(5):807—814.
5 Lasier, PJ; Winger, PV; Bogenrieder, KJ. (2000) Toxicity of manganese to Ceriodaphnia dubia and Hyalella azteca.
6 Arch Environ Contain Toxicol 38(3):298-304.
7 Laudon, H, Sjoblom, V; Buffam, I; et al. (2007) The role of catchment scale and landscape characteristics for runoff
8 generation of boreal streams. J Hydrol 344(1-3): 198-209.
9 Lee, G; Bigham, JM; Faure, G. (2002) Removal of trace metals by coprecipitation with Fe, Al, and Mn from natural
10 waters contaminated with acid mine drainage in the Ducktown Mining District, Tennessee Appl Geochem
1 1 17(5):569—581.
12 Lemly, AD. (1993) Guidelines for evaluating selenium data from aquatic monitoring and assessment. Environ
13 Monit Assess 28:83-100.
14 Leopold, LB. (1964) Fluvial processes in geomorphology. San Francisco, CA: W.H. Freeman and Company.
15 Linton, TK; Pacheco, MAW; Mclntyre, DO; et al. (2007) Development of bioassessment-based benchmarks for
16 iron. Environ Toxicol Chem 26(6): 1291 — 1298.
17 Loveland, T; Gutman, G; Buford, M; et al. (2003) Land-use/land cover change. In: Mahoney, JR; ed. Strategic plan
18 for the U.S. climate change science program. U.S. Climate Change Science Program, Washington, DC; pp 63-70.
19 Lowe, WH; Likens, GE; Power, ME. (2006) Linking scales in stream ecology. Bioscience 56(7):591-597.
20 MacDonald, DD; ingersoll, CG; Berger, TA. (2000) Development and evaluation of consensus-based sediment
21 quality guidelines for freshwater ecosystems. Arch Environ Contam Toxicol 39( 1 ):20—31.
22 Maggard, R; Kirk, E. (1998) Downstream impacts of surface mining and valley fill construction. Report for Pen
23 Coal Corporation. In: Draft programmatic environmental impact statement on mountaintop mining / valley fills in
24 Appalachia - 2003. Appendix D. U.S. Environmental Protection Agency, Region 3, Philadelphia, PA. Available
25 online at http://www.epa.gOv/Region3/mtntop/pdf/appendices/d/aquatic-ecosystem-enhanC"
26 symp/Proceedings/support/Maggard/downstream-impacts.pdf.
27 Martinez-Jeronimo, F; Martinez-Jeronimo, L. (2007) Chronic effect of NaCl salinity on a freshwater strain of
28 Daphnia magna Straus (Crustacea : Cladocera): A demographic study. Ecotoxicol Environ Saf 67(3):411-416.
29 Mayer, PM; Reynolds, SK; McCutchen, MD. (2007) Meta-analysis of nitrogen removal in riparian buffers. J
30 Environ Qual 36(4): 1172-1180.
31 McGlynn, B; McDonnell, J; Stewart, M, et al. (2003) On the relationships between catchment scale and streamwater
32 mean residence time. Hydrol Proc 17:175-181.
33 McGuire, KJ; McDonnell, JJ; Weiler, M; et al. (2005) The role of topography on catchment-scale water residence
34 time. Water Resour Res 41(5):W05002.
35 McMahon, TE; Zale, AV; Orth, DJ. (1996) Aquatic habitat measurements. In: Murphy, BR; Willis, DW; eds.
36 Fisheries techniques, 2nd edition. Chapter 4. Bethesda, MD: American Fisheries Society.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 87 DRAFT—DO NOT CITE OR QUOTE
-------
1 Merricks, TC; Cherry, DS; Zipper, CE; et al. (2007) Coal-mine hollow fill and settling pond influences on
2 headwater streams in southern West Virginia, USA. Environ Monit Assess 129(l-3):359-378.
3 Merritt, RW; K. W. Cummins, KW. eds. (1996) An introduction to the aquatic insects of North America. 3rd
4 Edition. Dubuque, 1A: Kendall/Hunt Publishing Company.
5 Messinger, T. (2003) Comparison of storm response of streams in small, unmined and valley-filled watersheds,
6 1999-2001, Ballard Fork, West Virginia. Water-Resources Investigations Report 02-4303, Department of the
7 Interior, U.S. Geological Survey, Charleston, WV.
8 Messinger, T; Paybins, KS. (2003) Relations between precipitation and daily and monthly mean flows in gaged,
9 unmined and valley-filled watersheds, Ballard Fork, West Virginia, 1999-2001. Water-Resources Investigations
10 Report 03-4113, Department of the Interior, U.S. Geological Survey, Charleston, WV.
11 Meyer, JS; Sanchez, DA; Brookman, JA; et al. (1985) Chemistry and aquatic toxicity of raw oil shale leachates from
12 Piceance Basin, Colorado. Environ Toxicol Chem 4(4):559-572.
13 Meyer, JL; Wallace, JB. (2001) Lost linkages and lotic ecology: rediscovering small streams. In: Press, MC;
14 Huntly, NJ; Levin, S: eds. Ecology: achievement and challenge. Maiden, MA: Blackwell Science; pp 295-317.
15 Meyer, JL; Strayer, DL; Wallace, JB; et al. (2007) The contribution of headwater streams to biodiversity in river
16 networks. J Am Water Resour Assoc 43:86-103.
17 Morse, JC; Stark, BP; McCafferty, WP. (1993) Southern Appalachian streams at risk - implications for mayflies,
18 stoneflies, caddisflies, and other aquatic biota. Aqua Conserv Marine Freshw Ecosys 3:293-303.
19 Morse, JC; Stark, BP; McCafferty, WP; et al. (1997) Southern Appalachian and other southeastern streams at risk:
20 Implications for mayflies, dragonflies, stoneflies, and caddisflies. In: Benz, GW; Collins, DE; eds. Aquatic fauna in
21 peril: The southeastern perspective. Special publication 1. Decatur, GA: Southeast Aquatic Research Institute. Lenz
22 Design and Communications, pp 17-42.
23 Mount, DR; Gulley, DD; Hockett, R; et al. (1997) Statistical models to predict the toxicity of major ions to
24 Ceriodaphnia dubia, Daphnia magna, and Pimephales promelas (fathead minnows). Environmental Toxicology and
25 Chemistry 16(10):2009-2019.
26 Nadeau, TL; Rains, MC. (2007) Hydrological connectivity between headwater streams and downstream waters:
27 How science can inform policy. J Am Water Resour Assoc 43:! 18-133.
28 National Geographic Maps and NatureServe. (2008) Hot spots of rarity-weighted richness for critically imperiled
29 and imperiled species. National Geographic and NatureServe, Washington, DC. Available online at
30 http://www.landscope.org/search/, keywords: rarity-weighted species richness map (accessed 12/16/2009).
31 Negley, TL; Eshleman, KN. (2006) Comparison of stormflow responses of surface-mined and forested watersheds
32 in the Appalachian Mountains, USA. Hydrol Proc 20( 16):3467—3483.
33 Nordstrom, DK; Ball, JW. (1986) The geochemical behavior of aluminum in acidified surface waters. Science
34 232:54-56.
35 Ohlendorf, HA. (2003) Ecotoxicology of selenium. In: Hoffman, DJ; Rattner, BA; Burton, GA, Jr; eds. Handbook
36 of Ecotoxicology, Second Edition. Boca Raton, FL: Lewis Publishers; pp 465—500.
37 Orr, PL; Guiguer, KR; Russel, CK. (2005) Food chain transfer of selenium in lentic and lotic habitats of a western
3 8 Canadian watershed. Ecotoxicol Environ Saf 63:175-188.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 88 DRAFT—DO NOT CITE OR QUOTE
-------
1 Palmer, MA. (2009) Reforming watershed restoration: science in need of application and applications in need of
2 science. Estuar Coast 32:1-17.
3 Paybins, KS. (2003) Flow origin, drainage area, and hydrologic characteristics for headwater streams in the
4 moLintaintop coal-mining region of southern West Virginia, 2000-01. Water-Resources Investigations Report 02-
5 4300, Department of the Interior, U.S. Geological Survey, Charleston, WV. Available online at
6 http://pubs.usgs.gov/wri/wri02-4300/pdf/wri02-4300.book.pdf.
7 Paybins, KS; Messinger, T; Eychaner, JH; et al. (2000) Water quality in the Kanawha-New River Basin: West
8 Virginia, Virginia, and North Carolina, 1996-98. Circular 1204. Department of the Interior, U.S. Geological Survey,
9 Charleston, WV. Available online at http://pubs.usgs.gOv/circ/circl204/#pdf.
10 Payne, WJ. (1981) Denitrification. New York: Wiley.
11 Peterson, BJ; Wollheim, WM; Mulholland, PJ; et al. (2001) Control of nitrogen export from watersheds by
12 headwater streams. Science 292:86-90.
13 Petranka, JW. (1998) Salamanders of the United States and Canada. Washington, DC: Smithsonian Institution
14 Press.
1 5 Petty, JT; Lamothe, PJ; Mazik, PM. (2005) Spatial and seasonal dynamics of brook trout populations inhabiting a
16 central Appalachian watershed. Trans Am Fish Soc 134:572-587.
1 7 Pollard, Al; Yuan, L. (2009) Assessing the consistency of response metrics of the invertebrate benthos: a
18 comparison oftrait- and identity-based measures. Freshw Biol doi: 10.1111/j. 1365-2427.2009.02235.x.
19 Pond, GJ. (2004) Effects of surface mining and residential land use on headwater stream biotic integrity in the
20 eastern Kentucky coalfield region. Kentucky Department of Environmental Protection, Division of Water,
21 Frankfort, KY. Available online at http://www.water.ky.gov/NR/rdonlyres/ED76CE4E-F46A-4509-8937-
22 1 A5DA40F3838/0/coal_miningl .pdf.
23 Pond, GJ; McMurray, SE. (2002) A macroinvertebrate bioassessment index for headwater streams of the eastern
24 coalfield region, Kentucky. Kentucky Department of Environmental Protection, Division of Water, Frankfort, KY.
25 http://www.water.ky.gOv/NR/rdonlyres/4CA8D7C4-309B-4175-ACC4-ICBDDDF73798/0/EKyMBI.pdf.
26 Pond, GJ; Passmore, ME; Borsuk, FA; et al. (2008) Downstream effects of mountaintop coal mining: comparing
27 biological conditions using family- and genus-level macroinvertebrate bioassessment tools. J N Am Benthol Soc
28 27:717-737.
29 Pusch, M; Fiebig, D; Brettar, 1; et al. (1998) The role of micro-organisms in the ecological connectivity of running
30 waters. Freshw Biol 40(3):453—495.
31 Roe, GH. (2005) Orographic precipitation. Annu Rev Earth Plan Sci 33:645-671.
32 Rose, AW; Cravotta, CA, III. (1998) Geochemistry of coal mine drainage. In: Brady, KBC; Smith, MW; Schueck, J;
33 eds. Coal mine drainage prediction and pollution prevention in Pennsylvania. Harrisburg, PA, Pennsylvania
34 Department of Environmental Protection: 1.1-1.22. Available online at
35 http://www.techtransfer.osmre.gov/nttmainsite/Library/pub/cmdpppp/chapterl.pdf.
36 Rudolph, BL; Andreller, I; Kennedy, CJ. (2008) Reproductive success, early life stage development, and survival of
37 westslope cutthroat trout (Oncorhynchus clarki lewisi) exposed to elevated selenium in an area of active coal
38 mining. Environ Sci Technol 42(8):3109-3114.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 89 DRAFT—DO NOT CITE OR QUOTE
-------
1 Schorer, M; Symader, W. (1998) Biofilms as dynamic components for the sorption of inorganic and organic
2 pollutants in fluvial systems. In: Haigh, MJ; Krecek, J; Rajwar, GS; Kilmartin, MP; eds. Headwaters: water
3 resources and soil conservation. Rotterdam, Netherlands: A. A. Balkema; pp 187-196.
4 Sencindiver, JC; Ammons, JT. (2000) Minesoil genesis and classification. In: Barnhisel, RI; Darmody, RG;
5 Daniels, WL; eds. Reclamation of drastically disturbed lands. Agronomy Series #41. Madison, WI: American
6 Society of Agronomy; pp 595-613.
7 Shank, M. (2004) Development of a mining fill inventory from multi-date elevation data. West Virginia
8 Department of Environmental Protection, Charleston, WV. Available online at
9 http://gis.wvdep.org/tagis/projects/valley_fill_paper.pdf(accessed 11/5/2009).
10 Simmons, JA; Currie, WS; Eshleman, K.N; et al.. (2008) Forest to reclaimed mine land use change leads to altered
11 ecosystem structure and function. Ecol Appl 18:104-118.
12 Skousen, J; Renton, J; Brown, H; et al. (1997) Neutralization potential of overburden samples containing siderite. J
13 Environ Qual 26:673-681.
14 Skouson, J.G., Sexstone, A; Ziemkiewicz, PF. (2000) Acid mine drainage control and treatment. In: Barnhisel, RI;
15 Darmody, RG; Daniels, WL; eds. Reclamation of drastically disturbed lands. Agron. Monogr. 41. Madison, WI:
16 American Society of Agronomy; pp 131-168.
17 Skousen, J; Ziemkiewicz, P; Venable, C. (2006) Tree recruitment and growth on 20-year-old, unreclaimed surface
18 mined lands in West Virginia. Int J Min Reclam Environ 20(2): 142-154.
19 Skousen, J; Gorman, J; Pena-Yewtukhiw, E; et al. (2009) Hardwood tree survival in heavy ground cover on
20 reclaimed land in West Virginia: Mowing and ripping effects. J Environ Qual 38(4): 1400— 1409.
21 Smith, ME; Lazorchak, JM; Herrin, LE; et al. (1997) A reformulated, reconstituted water for testing the freshwater
22 amphipod, Hyulella azleca. Environ Toxicol Chem 16(6): 1229— 1233.
23 Smith, SL; MacDonald, DD; Keenleyside, KA; et al. (1996) A preliminary evaluation of sediment quality
24 assessment values for freshwater ecosystems. J Great Lakes Res 22(3):624-638.
25 Sobek, A; Schuller, W; Freeman, J; et al. (1978) Field and laboratory methods applicable to overburdens and
26 minesoils. U.S. Environmental Protection Agency, Office of Research and Development, Industrial Environmental
27 Research Laboratory, Cincinnati, OH. EPA-600/2-78-054. Available online at
28 http://www.techtransfer.osmre.gov/nttmainsite/Library/hbmanual/fieldlab/front.pdf.
29 Soucek, DJ. (2007a) Bioenergetic effects of sodium sulfate on the freshwater crustacean, Ceriodaphnia dubia.
30 Ecotoxicology 16(3):317—325.
31 Soucek, DJ. (2007b) Comparison of hardness- and chloride-regulated acute effects of sodium sulfate on two
32 freshwater crustaceans. Environ Toxicol Chem 26(4):773-779.
33 Soucek, DJ; Kennedy, AJ. (2005) Effects of hardness, chloride, and acclimation on the acute toxicity of sulfate to
34 freshwater invertebrates. Environ Toxicol Chem 24(5): 1204—1210.
35 Soucek, DJ; Cherry, DS; Currie, RJ; et al. (2000) Laboratory to field validation in an integrative assessment of an
36 acid mine drainage-impacted watershed. Environ Toxicol Chemistry 19(4): 1036— L043.
37 Sperow, M. (2006) Carbon sequestration potential in reclaimed mine sites in seven east-central states. J Environ
38 Qual 35(4): 1428-1438.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 90 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
Stauffer, JR; Ferreri, CP. 2002 Characterization of stream fish assemblages in selected regions of mountain top
removal/valley fill coal mining. In: Draft programmatic environmental impact statement on mountaintop mining /
valley fills in Appalachia - 2003. Appendix D. U.S. Environmental Protection Agency, Region 3, Philadelphia, PA.
Available online at http://www.epa.gOv/Region3/mtntop/pdf/appendices/d/fisheries-study/staufferferreri-
oct2002.pdf.
Stout, B; Wallace, JB. (2003) A survey of eight major aquatic insect orders associated with small headwater streams
subject to valley fills from mountaintop mining. In: Draft programmatic environmental impact statement on
mountaintop mining/ valley fills in Appalachia - 2003. Appendix D. U.S. Environmental Protection Agency,
Region 3, Philadelphia, PA. Available online at
http://www.epa.goV/region3/mtntop/pdf/appendices/d/StoutWallaceMacroinvertebrate.pdf.
Stumm, W; Morgan, JJ. (1996) Aquatic chemistry: chemical equilibria and rates in natural waters. New York, NY:
John Wiley & Sons, Inc.
Sweeney, BW; Funk, DH; Standley, LJ. (1993) Use of the stream mayfly, Cloean triangulifer, as a bioassay
organism: Life history response and body burden following exposure to technical Chlordane. Environ Toxicol
Chem 12:115-125.
Swift, MC. (2002) Stream ecosystem response to, and recovery from, experimental exposure to selenium. J Aqua
Ecosyst Stress Rec 9:159-184.
Thorp, JH; Covich, AP; eds. (1991) Ecology and classification of North American freshwater invertebrates. San
Diego: Academic Press, Inc.
Tietge, JE; Hockett, JR; Evans, JM. (1997) Major ion toxicity of six produced waters to three freshwater species:
Application of ion toxicity models and tie procedures. Environ Toxicol Chem 16( 10):2002—2008.
Timpe, EK; Graham, SP; Bonett, RM. (2009) Phylogeography of the Brownback Salamander reveals patterns of
local endemism in Southern Appalachian springs. Mol Phylogen Evol 52(2):368-376.
Townsend, PA; Helmers, DP; Kingdon, CC; et al. (2009) Changes in the extent of surface mining and reclamation in
the Central Appalachians detected using a 1976-2006 Landsat time series. Rem Sens Environ 113:62-72.
U.S. (Environmental Protection Agency). (1979) Environmental assessment of surface mining methods: head-of-
hollow fill and mountaintop removal., Industrial Environmental Research Laboratory, Cincinnati, OH- EPA-600/7-
79-062.
U. S (Environmental Protection Agency) (2002) Landscape Scale Cumulative Impact Study. In: Draft programmatic
environmental impact statement on mountaintop mining / valley fills in Appalachia - 2003. Appendix I. U.S.
Environmental Protection Agency, Region 3, Philadelphia, PA. Available online at
http://www.epa.gOv/Region3/mtntop/eis2003appendices.htm#appd.
U.S. (Environmental Protection Agency) (2003) Draft programmatic environmental impact statement on
mountaintop mining / valley fills in Appalachia. U.S. Environmental Protection Agency, Region 3, Philadelphia,
PA. Available online at http://www.epa.gov/Region3/mtntop/eis2003.htm.
U.S. (Environmental Protection Agency) (2004) Draft ambient aquatic life criteria for selenium - 2004., Office of
Water, Washington, DC; EPA 822-R-O4-00I. Available online at
http://www.epa.gov/waterscience/criteria/selenium/pdfs/complete.pdf.
U.S. EPA (Environmental Protection Agency) (2005) Mountaintop mining/valley fills in Appalachia. Final
programmatic environmental impact statement. U.S. Environmental Protection Agency, Region 3, Philadelphia, PA.
Available online at http://www.epa.gov/region3/mtntop/pdf/mtm-vf_fpeis_full-document.pdf.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 91 DRAFT—DO NOT CITE OR QUOTE
-------
1 Ussiri, DAN; Lai, R. (2005) Carbon sequestration in reclaimed minesoils. Crit Rev Plant Sci 24(3): 151-165.
2 Vannote, RL; Minshall, GE; Cummins, KW; et al. (1980) River continuum concept. Can J Fish Aqua Sci
3 37:130-137.
4 Vesper, DJ; Roy, M; Rhoads, CJ. (2008) Selenium distribution and mode of occurrence in the Kanawha Formation,
5 southern West Virginia, U.S.A. Int J Coal Geol 73(3-4):237-249.
6 Wallace, JB; Eggert, SL; Meyer, JL; et al. (1999) Effects of resource limitation on a detrital-based ecosystem. Ecol
7 Monogr 69:409-442.
8 Warnick, SL; Bell, HL. (1969) The acute toxicity of some heavy metals to different species of aquatic insects. J
9 Water Pollut Con Fed 41:280-284.
10 Webster, JR; Benfield, EF; Ehrman, TP; et al. (1999) What happens to allochthonous material that falls into
11 streams? A synthesis of new and published information from Coweeta. Freshw Biol 41 (4):687—705.
12 Wickham, JD; Riitters, K.H; Wade, TG; et al. (2007) The effect of Appalachian mountaintop mining on interior
13 forest. Landsc Ecol 22:179-187.
14 Wiley, JB; Brogan, FD. (2003) Comparison of peak discharges among sites with and without valley fills for the July
15 8-9, 2001, Hood in the headwaters of Clear Fork, Coal River basin, mountaintop coal-mining region, southern West
16 Virginia. Report 03-133, Department of the Interior, U.S. Geological Survey, Charleston, WV. Available online at
17 http://pubs.usgs.gov/of/2003/ofr03-133/pdf/ofr03133.pdf.
18 Wiley, JB; Evaldi, RD; Eychaner, JH; et al. (2001) Reconnaissance of stream geomorphology, low streamflow, and
19 stream temperature in the mountaintop coal-mining region, southern West Virginia, 1999-2000. U.S. Department of
20 the Interior, U.S. Geological Survey, Charleston, WV. Available online at
21 http://pubs.usgs.gov/wri/wri014092/pdf/wri01-4092.book_new.pdf.
22 Williams, DD. (1996) Environmental constraints in temporary fresh waters and their consequences for the insect
23 fauna. J N Am Benthol Soc 15(4):634—650.
24 Wipfli, MS; Gregovich, DP. (2002) Export of invertebrates and detritus from fishless headwater streams in
25 southeastern Alaska: implications for downstream salmonid production. Freshw Biol 47(5):957-969.
26 Wipfli, MS; Richardson, JS; Naiman, RJ. (2007) Ecological linkages between headwaters and downstream
27 ecosystems: Transport of organic matter, invertebrates, and wood down headwater channels. J Am Water Resour
28 Assoc 43:72-85.
29 Wolman, MG. (1954) A method of sampling coarse river-bed material. Trans Am Geophys Union 35(6):951-956.
30 Wunsch, DR; Dinger, JS; Graham, CDR. (1999) Predicting ground-water movement in large mine spoil areas in the
31 Appalachian Plateau. Int J Coal Geol 41:73-106.
32 WV DEP (West Virginia Department of Environmental Protection). 2009, Environmap Explorer. West Virginia
33 Department of Environmental Protection, Charleston, WV. Available online at
34 http://gis.wvdep.org/imap/index.html (accessed 10/27/2009).
35 Yang, Y; Hofmann, T; Pies, C;. (2008) Sorption of polycyclic aromatic hydrocarbons (PAHs) to carbonaceous
36 materials in a river floodplain soil. Environ Pollut 156(3): 1357— 1363.
37 Younger, PL. (2004) Environmental impacts of coal mining and associated wastes: A geochemical perspective. A
38 Geochemical Perspective. Geol Soc 236:169-209.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 92 DRAFT—DO NOT CITE OR QUOTE
-------
Yudovich, YE; Ketris, MP. (2006) Chlorine in coal: A review. Int J Coal Geol 67:127-144.
Zotin, AI. (1958) The mechanism of hardening of the salmonid egg membrane after fertilization or spontaneous
activation. J Embryol Exp Morphol 6(4):546—568.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 93 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
APPENDIX A
LITERATURE SEARCHES
The peer-reviewed journal articles and reports reviewed in this paper were identified
using a variety of search methods. The report authors identified papers using ISI Web of
K.nowledgeSM and Google™ Scholar and references that either cited, or were cited by the
Programmatic Environmental Impact Statement or other relevant papers. This search was
supplemented by two more systematic searches described below. Because of the short project
timeline, author and supplemental searches were conducted in parallel.
A.l. KEYWORD SEARCH OF ISI WEB OF KNOWLEDGE™ AND GOOGLE™
SCHOLAR
Publications were identified using ISI Web of KnowledgeSM and Google™ Scholar based
on keywords (see Table A-l). The search covered publication dates up to August 2009. ISI
Web of KnowledgeSM searches journal articles dating from 1970. Google™ Scholar does not
specify a date range but generally sorts the search returns so that more recent references listed
first.
The keyword searches of ISI Web of KnowledgeSM and Google™ Scholar produced a
total of 277 citations including books, conference proceedings, journal articles, reports,
theses/dissertations and other sources (see Table A-2). Note that only the peer-reviewed journal
articles and reports were included in our review. Google™ Scholar generally returned more
results than ISI Web of KnowledgeSM. Google™ Scholar searches the Web across multiple
disciplines for journal articles, web documents, government reports, other papers,
theses/dissertations, books and abstracts. Searches are performed in such a manner that the most
relevant documents appear on the first page. Relevancy is determined by "weighing the full text
of each article, the author, the publication in which the article appears and how often the piece
has been cited in other scholarly literature." When searching Google™ Scholar, at minimum, the
first five pages were checked for relevant papers. Search terms were then refined if necessary.
ISI Web of Knowledge'™ returned journal articles that were very specific to the keywords that
were entered, which often resulted in fewer or no returns.
Journal articles were classified by region and relevance (see Table A-3). The region of
interest was defined as the Central Appalachian Coalfields (see Figure 1). Most of the journal
articles judged to be "not relevant" focused on acid mine drainage, rather than the alkaline
discharges that are typical of mountaintop mines and valley fills (MTM-VF). Laboratory studies
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 94 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
were classified as "stressors in streams from other regions," and a comment was added to
indicate that it was a laboratory investigation.
A.2. ECOTOXICOLOGICAL SEARCHES
Searches for ecotoxicological studies on the major ions, and iron, aluminum and
manganese were supplemented by keyword and Chemical Abstracts Service (CAS) number
searches using BIOSIS, CAS, TOXLINE, Cambridge Scientific Abstracts and U.S.
Environmental Protection Agency's (EPA)COTOX reference files.
Of the ecotoxicological searches, the one conducted for sulfate compounds calcium
sulfate (CaSC>4), magnesium sulfate (MgS04), potassium persulphate (KSO4), sodium sulfate
(NaS04), and ferrous sulfate (FexS04) was completed in time for inclusion in this appendix.
Citations were reviewed for applicability based on criteria such as the subject of the paper,
species group studied and analytical methods. Of 1,825 citations identified, 193 were considered
to be applicable and relevant to organism groups of interest (see Table A-4). Most of the
citations judged to be nonapplicable studied fate and transport rather than effects. The relevant
citations were further reviewed for relevance to the ion mixture typically observed in discharges
from MTM-VF.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 95 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table A-l. Keywords used for ISI Web of Knowledge81*1 and Google™
2 Scholar searches
3
Keywords
algae
dissolved oxygen
Mg
sediment transport
alkaline
DO
mine reclamation
sediments
amphibian
electrical
conductivity
minnow
selenium
Anuran
Ephemeroptera
mollusc
snail
Appalachian streams
fertilizer
Mollusca
sodium
aquatic biota
fish
mollusk
sodium chloride
aquatic insects
frog
mountain top mining
specific conductance
aquatic toxicity
herpetofauna
mountaintop mining
stoneflies
arsenic
hollow fill
mountaintop removal mining
stonefly
bank stability
hydrologic
alteration
mussels
stream temperature
bivalve
leachate
nickel
streams
caddisflies
macroinvertebrate
nutrients
sulfate
caddisfly
macroinvertebrates
overburden
TDS
calcium
macroinvertebrates
periphyton
temperature
coal mine
macrophyte
pH
thermal
coal mine
overburden
magnesium
Plecoptera
thermal regime
coal mine spoil
manganese
potassium
toad
conductivity
mayflies
riparian
total dissolved solids
diatom
mayfly
salamander
Trichoptera
discharge
metals
salinity
valley fill
4
5 Mg = magnesium; TDS = total dissolved solids.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 ' 96 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table A-2. Breakdown of the literary source of citations from keyword search
2
Literary source
Number of
citations
Books
6
Conference proceedings
22
Journal articles
154
Reports
38
Theses/dissertations
18
Other
39
3
4
5 Table A-3. Categorization of journal articles by region and relevance
6
Description
Number of
citations
MTM-VF in region of interest
5
MTM-VF in other region
0
Stressors in streams of interest
7
Stressors in streams from other regions
59
Review article of stressor of interest
25
Not relevant
30
Defer judgment
46
7
8
9 Table A-4. Breakdown of sulfate ecotoxicological search results by organism group
10
Organism Group
Number of citations
J Fish
62
Herpetofauna
3
Insects
5
Invertebrates
73
Plants
50
11
12
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 97 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
APPENDIX B
REGULATORY ISSUES RELATED TO MTM-VF OPERATIONS
Mountaintop mines and valley fills (MTM-VF) operations are permitted by state and
federal surface mining and environmental protection authorities. While regulations for
individual mines exist under the Surface Mining Control and Reclamation Act, which is
implemented by the Office of Surface Mining (OSM) and delegated States with OSM oversight,
there are several sections of the Clean Water Act (CWA) that apply. These are implemented by
the U.S. Environmental Protection Agency (EPA) and the U.S. Army Corps of Engineers
(USACE) and individual states authorized to implement portions of the CWA. A complete
listing and interpretation of regulations that affect MTM-VF operations is beyond the scope of
this report. The following is a general discussion of how the CWA, and particularly water
13
quality standards, are implemented in the context of MTM-VF.
B.l. IN GENERAL
Section 301 of the CWA prohibits the discharge of any pollutant by any person except in
compliance with, inter alia, a permit: 33 U.S.C. § 1311(a) "Except in compliance with this
section and sections ... 1342 and 1344 of this title, the discharge of any pollutant by any person
shall be unlawful." For purposes of MTM-VF, there are two relevant CWA permits. The
USACE issues a permit pursuant to Section 404 of the CWA (33 U.S.C. § 1344) for the
discharge of dredged and/or fill material. This permit includes construction of the valley fill
itself and the fill necessary to create an impounded sediment pond downstream of the toe of the
valley fill. The second permit is issued by either the EPA or an authorized State pursuant to
Section 402 of the CWA (33 U.S.C. § 1342). The Section 402 program is also known as the
"National Pollutant Discharge Elimination System" or "NPDES" program. The NPDES permit
includes the discharge from the sediment pond and any stormwater associated with the mining
activity.
NPDES permits must include technology-based effluent limitations. For purposes of
MTM-VF, the applicable technology-based effluent limitations are set forth at 40 C.F.R.
Part 434. In addition to industry sector-specific technology-based effluent limitations,
1-'While beyond the scope of this paper, it is worth noting that the Surface Mining Control and Reclamation Act
(SMCRA) and its implementing regulations also state that water quality must be maintained and that water quality
standards should not be violated. See, e.g., 30 U.S.C. § 1258(a)(9); 30 U.S.C. § 1265(b)(8)(C); 30 U.S.C. §
1265(b)( 10); 30 C.F.R. § 810.2(g); 30 C.F.R. § 816.42; and 30 C.F.R. § 816.57(a)(2). SMCRA also specifically
states that it does not supersede the Clean Water Act and other laws related to preserving water quality. See
30 U.S.C. § 1292(a)(3).
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 98 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
Section 301(b)(1)(C) of the CWA requires permits to include limits necessary to achieve water
quality standards 33 U.S.C. § 1311(b)(1)(C).
B.2. WATER QUALITY STANDARDS
Water quality standards are the foundation of the water quality-based control program
mandated by the CWA. Water quality standards define the goals for a waterbody by designating
its uses, setting narrative and numeric criteria to protect those uses and establishing provisions to
protect water quality from pollutants. See 40 C.F.R. § 130.3. A water quality standard consists
of four basic elements:
(1) Designated uses of the water body (e.g., recreation, water supply, aquatic life,
agriculture)
(2) Water quality criteria to protect designated uses (numeric pollutant concentrations and
narrative requirements)
(3) An antidegradation policy to maintain and protect existing uses and high quality waters
and
(4) General policies addressing implementation issues (e.g., low flows, variances, mixing
zones).
B.2.1. Designated Uses
The water quality standards regulation requires that States and authorized Indian Tribes
specify appropriate water uses to be achieved and protected. Appropriate uses are identified by
taking into consideration the use and value of the water body for public water supply, for
protection of fish, shellfish and wildlife, and for recreational, agricultural, industrial, and
navigational purposes. In designating uses for a water body, States and Tribes examine the
suitability of a water body for the uses based on the physical, chemical, and biological
characteristics of the water body, its geographical setting and scenic qualities and economic
considerations. Each water body does not necessarily require a unique set of uses. Instead, the
characteristics necessary to support a use can be identified so that water bodies having those
characteristics can be grouped together as supporting particular uses.
West Virginia has designated all waters of the state with an aquatic life use (ALU):
§47-2-6. Water Use Categories.
6.1. These rules establish general Water Use Categories and Water Quality
Standards for the waters of the State. Unless otherwise designated by these rules,
at a minimum, all waters of the State are designated for the Propagation and
Maintenance of Fish and Other Aquatic Life (Category B) and for Water Contact
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 99 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Recreation (Category C) consistent with Federal Act goals. Incidental utilization
for whatever purpose may or may not constitute a justification for assignment of a
water use category to a particular stream segment.
In addition, West Virginia Department of Environmental Protection (WVDEP) water
quality standards specify that waste assimilation and transport are not recognized as a designated
use:
§47-2-6. Water Use Categories.
6.1 .a. Waste assimilation and transport are not recognized as designated uses.
The classification of the waters must take into consideration the use and value of
water for public water supplies, protection and propagation of fish, shellfish and
wildlife, recreation in and on the water, agricultural, industrial and other purposes
including navigation.
B.2.2. Water Quality Criteria
States establish criteria necessary to protect the designated use. Water quality criteria
may take the form of either specific numeric criteria, such as concentrations of a particular
pollutant, or narrative description of water quality conditions.
B.2.2.1. Numeric Criteria
Section 304(a)(1) of the Clean Water Act requires us to develop numeric criteria for
water quality that accurately reflect the latest scientific knowledge. These criteria are based
solely on data and scientific judgments on pollutant concentrations and ecological or human
health effects. Section 304(a) also provides guidance to States and Tribes in adopting water
quality standards. Numeric criteria are developed for the protection of aquatic life as well as for
human health.
Numeric aquatic life criteria are generally pollutant-specific and reflect numeric limits on
the amount of a pollutant that can be present in a water body without harm to indigenous aquatic
life. Aquatic life criteria are designed to provide protection for aquatic organisms from the
effects of acute (short-term) and chronic (long-term) exposure to potentially harmful chemicals.
Human health criteria set allowable concentrations based on human exposure to water
pollutants when humans drink untreated surface water or eat fish, shellfish, or wildlife that have
been contaminated by pollutants in surface waters. To reduce the risk to humans from these
sources, EPA scientists research information to determine the levels at which specific chemicals
are not likely to adversely affect human health.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 100 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
In making water quality management decisions, a State or Tribe should independently
apply each criterion that has been adopted into its water quality standards. If a water body has
multiple designated uses with different criteria for the same pollutant, States/Tribes should use
the criterion protective of the most sensitive use.
B.2.2.2. Narrative Criteria
While numeric criteria help protect a water body from the effects of specific chemicals,
narrative criteria protect a water body from the effects of pollutants that are not easily measured,
or for pollutants that do not yet have numeric criteria, such as chemical mixtures, suspended and
bedded sediments and floatable debris.
West Virginia's narrative water quality criteria are set forth in a portion of the West
Virginia regulations known as "Conditions Not Allowed:"
WV §47-2-3. Conditions Not Allowable In State Waters.
3.2.i. Any other condition, including radiological exposure, which adversely alters
the integrity of the waters of the State including wetlands; no significant adverse
impact to the chemical, physical, hydrologic, or biological components of aquatic
ecosystems shall be allowed.
Other examples presented here include excerpts from Kentucky surface water standards
(Chapter 10); the narrative standards in 401 KAR 10:026-031, which state in part
001 Definitions 401 KAR Chapter 10
(5) "Adversely affect" or "adversely change" means to alter or change the
community structure or function, to reduce the number or proportion of sensitive
species, or to increase the number or proportion of pollution tolerant aquatic
species so that aquatic life use support or aquatic habitat is impaired.
(38) "Impact" means a change in the chemical, physical, or biological quality or
condition of surface water.
(39) "Impairment" means, a detrimental impact to surface water that prevents
attainment of a designated use.
401 KAR 10:031, Section 2: Minimum Criteria Applicable to All Surface
Waters.
(1) The following minimum water quality criteria shall be applicable to all surface
waters including mixing zones, with the exception that toxicity to aquatic life in
mixing zones shall be subject to the provisions of 401 KAR 10:029, Section 4.
Surface waters shall not be aesthetically or otherwise degraded by substances that
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 ' 101 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
(a) Settle to form objectionable deposits;
(b) Float as debris, scum, oil, or other matter to form a nuisance;
(c) Produce objectionable color, odor, taste, or turbidity;
(d) Injure, are chronically or acutely toxic to or produce adverse physiological
or behavioral responses in humans, animals, fish and other aquatic life;
(e) Produce undesirable aquatic life or result in the dominance of nuisance
species;
(0 Cause fish flesh tainting.
A narrative chemical criterion for total dissolved solids and specific conductance reads
401 KAR 10:031, Section 4: Aquatic Life.
(f) Total dissolved solids or specific conductance. Total dissolved solids or
specific conductance shall not be changed to the extent that the indigenous aquatic
community is adversely affected.
B.2.2.2. Establishing Impairment
Section 303(d) of the CWA requires States to periodically identify those waters that are
not expected to achieve water quality standards even after application of technology-based
effluent limitations to NPDES-permitted point sources (33 U.S.C. § 1313(d)). This identification
is commonly referred to as the State's "Section 303(d) list." By regulation, States must submit
their Section 303(d) lists to EPA for approval every even-numbered year (40 C.F.R. § 130.7(d)).
In establishing its Section 303(d) list, States must consider all existing and readily available
information, including predictive models (40 C.F.R. § 130.7(b)(5)).
In July 1991, EPA transmitted final national policy on the integration of biological,
chemical and toxicological data in water quality assessments. According to this policy, referred
to as "Independent Application," indication of impairment of water quality standards by any one
of the three types of monitoring data (biological, chemical, or toxicological) should be taken as
evidence of impairment regardless of the findings of the other types of data. This policy
continues to the present. See, e.g., Guidance for 2006 Assessment, Listing and Reporting
Requirements Pursuant to Sections 303(d), 305(b) and 314 of the Clean Water Act.
EPA supports use of biological assessments as a direct measure of whether the water
body is achieving the designated use and relevant narrative criteria. A water body in its natural
condition is free from the harmful effects of pollution, habitat loss and other negative stressors.
It is characterized by a particular biological diversity and abundance of organisms. This
biological integrity—or natural structure and function of aquatic life—can be dramatically
different in various types of water bodies in different parts of the country. EPA recognizes that
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 102 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
biological assessments are a direct measure of the aquatic life use. Because of the natural
variability in ecosystems and aquatic life around the country, EPA could not develop national
biocriteria. Instead, EPA developed methodologies that States can use to assess the biological
integrity of their waters and, in so doing, set protective water quality standards. These
methodologies describe scientific methods for determining a particular aquatic community's
health and for maintaining optimal conditions in various bodies of water. States use these
standard methods to develop their own bioassessment methods and tools. Bioassessment results
are used to support many programs under the CWA (see Figure B-l).
The States have increasingly relied upon biological monitoring in lieu of ambient water
chemistry monitoring because biological monitoring allows the States to maximize monitoring
resources and to assess a larger percentage of their waters. Since 2004, West Virginia has
utilized standard field collection, laboratory and data analysis methods to use its biological
assessment data. This has resulted in West Virginia's use of a family-level benthic metric
developed jointly by EPA and West Virginia Department of Environmental Protection called the
West Virginia Stream Condition Index (WV SCI) to identify impairment of the aquatic life use.
See http://www.wvdep.org/Docs/536_WV-Index.pdf. West Virginia also developed an
assessment methodology for using the WV SCI to interpret its narrative criterion and to make
aquatic life use-attainment decisions. For an example, see WVDEP's 2008 Integrated Water
Quality Monitoring and Assessment Report available at
http://www.wvdep.org/Docs/16495_WV_2008_IR_Supplements_Complete_Version_EPA_Appr
oved.pdf.
In Kentucky, the Kentucky Division of Water assessment methodologies for ALU
attainment are similar, where the State uses biological monitoring data and statistical-based
multimetric index analyses to assess waterbody attainment. For macroinvertebrates, the KY
Macroinvertebrate Bioassessment Index is used to evaluate ALU:
http://www.water.ky.gov/NR/rdonlyres/7F189804-4322-4C3E-B267-
5A58E48AAD3F/0/Statewide_MBI.pdf.
In nonheadwater streams, KY uses fish communities as other indicators of ALU with the
KY Index of Biotic integrity, a similarly constructed multimetric index:
http://www.water.ky.gov/NR/rdonlyres/04C65101-AFlC-4751-809B-
4F5D09B7269A/0/KIBI_paper.pdf.
Section 303(d) also requires the States to establish total maximum daily loads (TMDLs)
for their impaired waters. Essentially, a TMDL is a measure of the assimilative capacity of a
waterbody considering seasonal variability and critical conditions, allocated among point sources
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 103 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
and nonpoint sources and incorporating a margin of safety. See 33 U.S.C. § 1313(d);
40 C.F.R. § 130.2(i); andl30.7(c).
B.3. IMPLEMENTATION OF WATER QUALITY STANDARDS THROUGH NPDES
PERMITS
As set forth above, Section 301 of the CWA requires NPDES permits to contain both
technology-based effluent limitations and water quality-based effluent limitations. For the
industry sector, that includes surface coal mining with valley fills, the applicable technology-
based effluent limitations are set forth at 40 C.F.R. Part 434. These effluent limitations include
limitations on discharges from coal preparation plants, acid and alkaline mine drainage,
postmining areas, remining and western alkaline mining. For example, effluent limitations on
discharges from a new source of alkaline mine drainage include limits on iron, total suspended
solids and pH. See 40 C.F.R. § 434.45.
The NPDES regulations implement the water quality-based effluent limitations
requirement as set forth in CWA Section 301(b)(1)(C) through the following regulatory
requirements:
No permit may be issued ... (d) When the imposition of conditions cannot ensure
compliance with the applicable water quality requirements of all affected States...
(40 C.F.R. § 122.4(d)).
[E]ach NPDES permit shall include conditions meeting the following
requirements when applicable .... [A]ny requirements in addition to or more
stringent than promulgated effluent limitations guidelines ... necessary to:
achieve water quality standards under Section 303 of the CWA, including State
narrative criteria for water quality... (40 C.F.R. § 122.44(d)(1)).
No permit may be issued ... (i) To a new source or a new discharger, if the
discharge from its construction or operation will cause or contribute to the
violation of water quality standards (40 C.F.R. § 122.4(i)).
Most States, including West Virginia and Kentucky, have been authorized to issue
NPDES permits for discharges to waters within their borders. EPA retains the ability to review,
object to and if necessary, take over issuance of a particular NPDES permit. See 33 U.S.C. §
1342(d); 40 C.F.R. § 123.44. The scope of EPA's NPDES permit review in a particular state is
generally spelled out in a memorandum of agreement with that state (40 C.F.R. § 123.44). EPA
also retains the ability to enforce discharges without or in violation of an NPDES permit
(33 U.S.C. § 1319).
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 104 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
B.4. IMPLEMENTATION OF WATER QUALITY STANDARDS THROUGH SECTION
404 PERMITS
Section 404(b)(1) directs the EPA in conjunction with the Secretary of the Army to
establish guidelines to be applied by the USACE when considering an application for a permit to
discharge dredged and/or fill material pursuant to Section 404 of the CWA.. This instruction has
resulted in the Section 404(b)(1) Guidelines (40 C.F.R. Part 230), which provide the substantive
environmental criteria that must be applied by the USACE when considering a Section 404
permit application. Among other things, the USACE may issue a permit only if it determines
that the project represents the least damaging practicable alternative:
[N]o discharge of dredged or fill material shall be permitted if there is a
practicable alternative to the proposed discharge which would have less adverse
impact on the aquatic ecosystem, so long as the alternative does not have other
significant adverse environmental consequences (40 C.F.R. § 230.10(a)).
The USACE also must ensure that the project proponent has taken "all appropriate and
practicable steps to avoid and minimize adverse impacts to waters of the United States"
(33 C.F.R. § 332.1(c)); see also 40 C.F.R. § 230.10(a)( 1 )(i); 40 C.F.R. § 230.10(d); and 40
C.F.R. §§ 230.70-.77.
In addition, the Section 404(b)(1) Guidelines prohibit the issuance of a permit "if it:
(1) Causes or contributes, after consideration of disposal site dilution and dispersion, to
violations of any applicable State water quality standard," (40 C.F.R. § 230.10(b)(1)), or if it
"will cause or contribute to significant degradation of the waters of the United States, ...
[including] (1) Significantly adverse effects of the discharge of pollutants on human health or
welfare, including but not limited to effects on municipal water supplies, plankton, fish, shellfish,
wildlife and special aquatic sites. (2) Significantly adverse effects of the discharge of pollutants
on life stages of aquatic life and other wildlife dependent on aquatic ecosystems, including the
transfer, concentration and spread of pollutants or their byproducts outside of the disposal site
through biological, physical and chemical processes; (3) Significantly adverse effects of the
discharge of pollutants on aquatic ecosystem diversity, productivity and stability...." (40 C.F.R. §
230.10(c)). The USACE also must consider the effect of the discharge on fish, crustaceans,
mollusks and other aquatic organisms in the food web (40 C.F.R. § 230.31), the effect on
benthos (40 C.F.R. § 230.61(b)(3)) and the suitability of water bodies for populations of aquatic
organisms (40 C.F.R. § 230.22).
Before issuing a federal permit or license, federal agencies, including the USACE, must
obtain a certification from the State in which the discharge will originate that the discharge will
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 105 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
comply with applicable provisions of 33 U.S.C. § 1311, 1312, 1313, 1316 and 1317. Among
other things, therefore, the USACE must obtain a certification that the discharge will comply
with applicable water quality standards, which are established pursuant to 33 U.S.C. § 1313. In
considering the potential of a discharge to cause or contribute to an excursion from water quality
standards, the USACE generally will consider conclusive the State's CWA Section 401 water
quality certification, unless EPA advises of other water quality aspects to be taken into
consideration (33 C.F.R.. § 320.4(d)).
While the USACE is the permit-issuing authority for Section 404, EPA retains significant
authorities, including the authority to prohibit, deny or restrict the use of any defined area for
specification as a disposal site pursuant to Section 404(c) (33 U.S.C. § 1344( c)), the ability to
request consideration of particular permits by the USACE at the Headquarters level pursuant to
the Memorandum of Agreement described in Section 404(q) (33 U.S.C. § 1344(q)), the ability to
identify waters that are within the scope of the CWA and to determine the applicability of
exemptions pursuant to a Memorandum of Agreement with the USACE under Section 404(f)
(33 U.S.C. § 1344(0), ^d the ability to enforce discharges without a permit (33 U.S.C. § 1319).
Figure B-2 depicts the sequence of actions necessary to address impaired streams under
the CWA.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 106 DRAFT—DO NOT CITE OR QUOTE
-------
Wet Weather
Discharge (CSOs,
Stormwater)
Listing of Impaired
Waters
(CWA §303d)
Nonpoint Source
Assessment
(CWA §319)
Point Source
Discharge
Water Quality
Standards and
Aquatic Life Use
Assessments
(CWA §305b)
Bioassessment
Data
Marine Point
Source
Discharge
Permitting
(CWA §403c)
Sewage
Treatment
Plant
Discharges in
Marine
Waters
(CWA §301 h)
Marine Protection
Research and
Sanctuaries Act—
Ocean Dumping
(MPRSA)
Comprehensive
Comprehensive
Watershed
Assessments
J
Hazardous Waste
Site Assessments
(CWA §104e)
Evaluation and Permitting
of Habitat Modifications
(CWA §404)
Figure B-l. Simple representation of CWA programs that rely on biological
assessment data for program implementation. Coal mining activities sections
highlighted in red.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 107 DRAFT—DO NOT CITE OR QUOTE
-------
1
Determine
Protection Level
(EPA Criteria/State WQS)
Measure Progress
t
Conduct WQ
Assessment
(Identify Impaired Waters)
Monitor and Enforce
Compliance
(Self Monitoring, Agency Monitoring)
V
\
Set Priorities
(Rank/Target Waterbodies)
Establish Source
Controls
(Point Source, NPS)
/
Evaluate Appropriateness
of WQS for Specific Waters
(Reaffirm WQS)
2
3
4
Define and Allocate
Control Responsibilities
(TMDL/W LA/LA)
Figure B-2. Water quality-based approach to pollution control.
This document is a draft for review purposes only and does not constitute Agency policy.
12/17/09 108 DRAFT—DO NOT CITE OR QUOTE
------- |