Content uploaded by Fabian Nippgen
Author content
All content in this area was uploaded by Fabian Nippgen on Sep 26, 2017
Content may be subject to copyright.
Creating a More Perennial Problem? Mountaintop Removal Coal
Mining Enhances and Sustains Saline Baseflows of Appalachian
Watersheds
Fabian Nippgen,*
,†
Matthew R. V. Ross,
‡
Emily S. Bernhardt,
‡
and Brian L. McGlynn
§
†
Department of Ecosystem Science and Management, University of Wyoming, Laramie, Wyoming 82071, United States
‡
Department of Biology and
§
Division of Earth and Ocean Sciences, Nicholas School of the Environment, Duke University, Durham,
North Carolina 27708, United States
ABSTRACT: Mountaintop removal coal mining (MTM) is a
form of surface mining where ridges and mountain tops are
removed with explosives to access underlying coal seams. The
crushed rock material is subsequently deposited in headwater
valley fills (VF). We examined how this added water storage
potential affects streamflow using a paired watershed approach
consisting of two sets of mined and unmined watersheds in
West Virginia. The mined watersheds exported 7−11% more
water than the reference watersheds, primarily due to higher
and more sustained baseflows. The mined watersheds exported
only ~1/3 of their streamflow during storms, while the
reference watersheds exported ~2/3 of their annual water yield
during runoffevents. Mined watersheds with valley fills appear
to store precipitation for considerable periods of time and steadily export this alkaline and saline water even during the dry
periods of the year. As a result, MTMVFs in a mixed mined/unmined watershed contributed disproportionately to streamflow
during baseflow periods (up to >90% of flow). Because MTMVFs have both elevated summer baseflows and continuously high
concentrations of total dissolved solids, their regional impact on water quantity and quality will be most extreme and most
widespread during low flow periods.
■INTRODUCTION
Humans have manipulated their environment on a detectable
scale for thousands of years since at least the onset of
agriculture.
1,2
Today, the anthropogenic impacts on the
landscape include, among others, large-scale deforestation,
3
agriculture,
4
under- and above-ground mining for coal and
other natural resources,
5,6
tar sand mining,
7
damming of major
rivers,
8,9
urbanization,
10
and wars.
11,12
The earth layer affected
by those disturbances has recently been referred to as the
critical zone
13
and includes vegetation, soils, and groundwater-
bearing bedrock. Disturbances of the critical zone occur across
the planet,
14
so it is important to understand how physical and
biological parameters are altered in order to evaluate the
ramifications for encompassing ecosystems. Assessing the
quantitative and qualitative change in hydrologic fluxes during
and after these landscape alterations is often a crucial first step
for understanding ecosystem wide transformations, as hydrol-
ogy has been recognized as a driver for multiple ecosystem
processes throughout the critical zone, such as nutrient or
contaminant export,
15,16
aquatic biodiversity,
17,18
and human
well-being.
19
Here, we present an example of hydrologic change from a
large-scale mining disturbance, which is common in the USA
but is also practiced in other parts of the world, e.g., Canada
20
and China.
21
Mountaintop removal coal mining with valley fills
(MTMVF) is a surface-mining procedure during which the tops
of mountains and ridges are removed to access underlying coal
seams. The resulting rock material is subsequently deposited
into adjacent valleys.
22
These valley fills (VF), designed as
permanent storage for excess spoil and to reduce landslides on
reclaimed mine areas,
23
are estimated to have buried up to 4000
km of headwater streams.
24
MTMVF is endemic to the
Appalachian coal region of Kentucky, Tennessee, Virginia, and
West Virginia (Figure 1b), where it became more prevalent in
the 1990s. The US Environmental Protection Agency estimated
that as of 2012 surface mines would cover approximately 7% of
the region.
24
In contrast to many other disturbances that either
do not extend into the bedrock at all or only to a limited degree
(e.g., deforestation, urbanization, agriculture) and mainly affect
vegetation or infiltration capacities,
25
MTMVF can disturb the
critical zone hundreds of meters deep.
24,26
This disturbance
happens both in former mountaintop and ridge areas where
bedrock is removed up to hundreds of meters deep through
explosion and in the valleys, where the crushed rock is
deposited on the ground surface, essentially adding a highly
Received: May 3, 2017
Revised: June 21, 2017
Accepted: June 23, 2017
Published: July 13, 2017
Article
pubs.acs.org/est
© 2017 American Chemical Society 8324 DOI: 10.1021/acs.est.7b02288
Environ. Sci. Technol. 2017, 51, 8324−8334
This is an open access article published under an ACS AuthorChoice License, which permits
copying and redistribution of the article or any adaptations for non-commercial purposes.
disturbed layer to the critical zone.
26
Despite the scale and
nature of the disturbance, MTMVF has only recently received
more focused attention from the hydrologic community,
27−32
but basic knowledge gaps remain as to how the dramatic
changes in topography and critical zone associated with
MTMVF affect hydrologic response and long-term hydrologic
regimes of watersheds.
While some forms of surface mining (e.g., strip and contour
mining) lead to increased peak flows and overall water export
due to the compaction of soils and spoil during reclama-
tion,
33−36
MTMVF areas feature large volumes of crushed rock,
which could increase watershed storage. Ross et al.
26
estimated
volumes of ∼1500 Appalachian valley fills and conservatively
concluded that mining could increase the water storage capacity
of mined watersheds by a factor of 10 but that individual valley
fills were highly variable in size.
Empirical evidence for this enhanced-storage effect is limited,
relatively recent, and in parts confounded by other disturbances
present in the study watersheds. Messinger and Paybins
30
reported increased runoffvolumes in a small first-order mined
watershed in West Virginia relative to an unmined watershed
and attributed the increase in baseflow to the greater storage
potential of the mined watershed. Somewhat surprisingly, they
also found that during large events, the mined watershed would
export more water than the unmined watershed. On a larger
scale, Zegre et al.
28
did not detect significant changes in annual
streamflow in a 1000 km2watershed in West Virginia over a 16-
year period despite increasing mining activities. It was noted by
the authors that only a limited area was affected by mining (9%
of the surface area). When they extended the time series to ∼40
years, Zegre et al.
31
were able to detect decreases in streamflow
maxima and small but statistically significant increases in
Figure 1. a) Location of the four study watersheds relative to other. Red outline denotes mined watersheds, black outline reference watersheds; b)
Appalachian coal region, highlighted in red are mining impacted areas; c) LB topography pre-mining; d) LB topography post-mining; e) elevation
changes in LB from pre- to post-mining; f) LB with post-mining delineated watershed boundaries. Maps of the mined watersheds were generated
using LiDAR data made available by the West Virginia Department of Environmental Protection (http://tagis.dep.wv.gov/home). Maps of the
reference sites were generated using elevation data from the National Elevation Dataset (https://lta.cr.usgs.gov/NED).
Environmental Science & Technology Article
DOI: 10.1021/acs.est.7b02288
Environ. Sci. Technol. 2017, 51, 8324−8334
8325
baseflow contributions for the same watershed that were
consistent with Messinger and Paybins.
30
However, in addition
to mountaintop mining, their research watershed was also
affected by extensive subsurface mining, making direct
inferences to mountaintop mining difficult.
31,37
In addition to changes in water yield, the disturbance of the
critical zone caused by mountaintop mining also leads to
degraded streamwater quality. Precipitation that enters
MTMVF watersheds flows through a reactive matrix of pyrite
and calcareous bedrock that, via strong acid weathering, releases
large amounts of various ions, such as SO42−,Ca
2+,Mg
2+ (e.g.,
ref 38), as well as the toxic pollutant selenium.
20,39
Together,
these constituents increase the salinity and pH of streams
draining mines in a well-documented phenomenon of alkaline
mine drainage.
38,40,41
Alkaline mine drainage has been shown to
negatively impact stream biota
42−45
in as much as 22% of
streams in central Appalachia.
40
However, the mechanisms and
timing of stream impairment caused by mountaintop mining
are tightly coupled to hydrologic processes and are hence not
well understood.
We quantified water yield and water quality changes in
stormflow and baseflow behavior for two sets of mined and
unmined watersheds in West Virginia. MTMVF was the
dominant disturbance present in our research watersheds,
allowing for direct inferences of observed changes to the critical
zone disturbance. We used high-resolution streamflow gauging,
high-resolution specific conductance monitoring (a proxy for
salinity), precipitation monitoring, landscape analysis, and
empirical baseflow separation methods for 12 months of
rainfall, runoffdata, to address the following questions:
(1) To what degree does mountaintop mining alter baseflow
and stormflow contributions to total runoffand does this
effect change with increasing watershed scale?
(2) How does MTMVF affect the export of total dissolved
solids?
(3) How do mined and unmined portions of partially mined
watershed contribute to runoffacross hydrologic
seasons?
(4) How do hydrologic changes associated with MTMVF
compare to other disturbances?
■METHODS
Site Description. The Mud River watershed is located in
southwestern West Virginia, approximately 40 km southwest of
Charleston. The four study watersheds were paired (first and
fourth order) based on size and the presence or absence of
mining (Figure 1a). The 68 ha Laurel Branch (LB) watershed, a
tributary to the Mud River, is approximately 95% mined, while
46% of the 3672 ha Mud River (MR) watershed into which LB
flows is in active or reclaimed mines. The majority of mining in
MR (>90%) happened between 1985 and 2005. The youngest
mines are located in the northern part of MRwhich contains
the LB subwatershedwhere mining began after 2005.
39
The
unmined reference sites include the 3463 ha Left Fork (LF) of
the Mud River and the 118 ha Rich’s Branch (RB), a tributary
to the Left Fork River. There are no known deep mines in the
area
29
that could confound the analyses and interpretations
through legacy effects.
37,46
We were not able to collect data in
the mined watersheds before the mining activity started.
However, since the watersheds were similar in size and
topography and are close to one another, we will refer to the
unmined watersheds as the reference watersheds.
Soils in the unmined areas of the four watersheds are
generally shallow (<2 m), well-drained silty loams or sandy
loams with moderate to rapid permeability ratings.
47,48
The
underlying geology consists of alternating layers of siltstone,
sandstone, and shale.
49
Vegetation in the unmined areas is
mixed mesophytic forest,
50
and the mined portions are either
barren or with herbaceous and shrub cover. Median vegetation
height derived from Lidar data (first returns minus last returns)
was 0.3 m in LB and 0.4 m in MR. The Lidar data did not cover
the reference watersheds. However, median vegetation height
in the unmined parts of MR, which are representative for the
vegetation in the reference sites, was 24 m. Average annual
precipitation in the base period 1981−2010 was 1183 mm.
51
Precipitation is relatively homogeneously distributed over the
year with slightly wetter months during the summer. Average
annual air temperature for the 1981−2010 base period was
12.7°.
51
The growing season in this area extends from May
through October (data from the Fernow Experimental Forest,
about 230 km northeast of our study sites
52
).
Spatial Analysis. We used pre- and postmining digital
elevation models and methods developed by Ross et al.
26
to
quantify the geomorphic changes associated with MTMVF in
LB and MR (slope, change of watershed area pre- to
postmining, estimate of VF volumes).
Hydrologic Measurements. The study period encom-
passed the 2015 water year (10/01/2014−10/01/2015).
Precipitation was measured at three different locations (Figure
1a) using Onset HOBO RG3 rain gauges and data loggers
recording at 10 min intervals. The small watersheds (RB and
LB) were assigned the precipitation of the closest rain gauge,
while the larger LF and MR were assigned precipitation values
based on inverse distance weighting with the two closest rain
gauges. The rain gauges had on average ∼11% missing data;
however, the gauge near LB was swept away during a major
flooding event in April and had ∼30% of the data missing.
Missing data at each rain gauge were detected and filled using
double mass curves with adjacent rain gauges.
53
Open-channel streamwater levels and specific conductance
(SC), a measure of the ionic strength of a water sample, were
recorded at 10 min intervals with Onset HOBO Water Level
loggers and Onset HOBO Specific conductance loggers,
respectively, during the entire period, with redundant Decagon
CTD sensors connected to Campbell Scientific CR1000 data
loggers beginning January 2015. We developed stage−discharge
rating curves at each gauging site with >13 manual runoff(Q)
measurements over a range of observed discharge. At LB, we
manually measured the maximum observed runoffof the water
year. At RB, LF, and MR, bank-full Gauckler−Manning
54
estimates of Qwere used to restrict the rating curves at high
water levels. Water levels were above bank for <3% of the water
year at LF, and <1% at RB and MR. Missing data at RB from
10/14/2014 to 10/26/2014 was filled by interpolation since no
precipitation was recorded during this time. During a major
precipitation event on April 3, both RB and LF experienced
backflow from the Mud River Reservoir downstream of the
gauging sites, affecting the falling limbs of the hydrographs. The
affected time periods were corrected using two-term
exponential regression models. The LB gauging site experi-
enced backflow from the Mud River during the falling limb of
three storms (March 4, April 3, and July 14), which were
corrected using a regression with water level data from a sensor
approximately 100 m upstream.
Environmental Science & Technology Article
DOI: 10.1021/acs.est.7b02288
Environ. Sci. Technol. 2017, 51, 8324−8334
8326
Throughout the study period, the specific conductance
sensors experienced drift caused by either deposition of
dissolved particulates onto the electrodes or by becoming
covered in sediment. We assumed a linear drift in the SC data
and corrected the drift using measurements from a hand-held
SC meter at biweekly intervals.
From 06/27/2015 to 07/20/2015. the mining company
began intermittently pumping water out of a small retention
pond 150 m upstream of the gauging station, thereby affecting
the LB stream levels. The water was pumped to another pond
uphill of the LB valley fills. We believe the water remained in
the LB watershed.
Hydrologic Analysis. We used three different hydrograph-
based methods to separate baseflow from stormflow. The first
method was similar to the approach proposed by Hewlett and
Hibbert,
55
where baseflow rises at a constant rate after the
onset of precipitation. The rate of baseflow rise was the same
for each watershed and was chosen so that the small reference
watershed exhibited a baseflow percentage that corresponds to
the ∼30% USGS estimate of baseflow for the Mud River
watershed.
56
The second method was the “local minimum”
method, which searches the hydrograph for local minima over
specified periods of time.
57
For this method, the 10 min data
was aggregated to daily values. The third method was an
adaptation of the “constant-k”method proposed by Blume et
al.
58
The approach initially requires the computation of a
modified recession constant, k*,as
*= ×kQ
tQ
d
d
1
mean
Assuming an exponential recession curve in the case of a
linear groundwater reservoir, k*should become approximately
0 (or constant) when the stormflow portion of the hydrograph
ends. Baseflow was then delineated as a straight line with slope
0 from the beginning of the runoffevent to the end of
stormflow (the time during which k*≠0). Originally
developed for event-hydrograph separation, we extended the
method to the length of the study period. Since k*fluctuates
around 0 rather than becoming 0, even during long periods
without precipitation, we assumed constancy in k*when
−0.001 ≤k*≤0.001. To attenuate sensor-related jumps in the
Qtime series, we calculated k*using a 4-h running average.
Further, we calculated flow duration curves for all watersheds
as well as cumulative Qtotals to compare peak and baseflow
behavior between the reference and the mined sites as well as
potential changes in the seasonal timing of water delivery.
In addition to the hydrograph separation to distinguish
between stormflow and baseflow, we estimated the contribu-
tions from the mined and unmined areas in MR with a simple
two-component hydrograph separation using specific con-
ductance of RB (reference) and LB (mined) as the two
endmembers, following Pinder and Jones
59
=+
=+
=−
−
⎛
⎝
⎜⎞
⎠
⎟
QQ Q
QSC Q SC Q SC
QQ
SC SC
SC SC
MR unmined mined
MR MR unmined unmined mined mined
mined MR
unmined MR7
unmined mined
with Qbeing runoffand SC being specific conductance. This
approach has been applied in many geographic regions using
different chemical signatures to distinguish endmembers,
60−62
including specific conductance.
63−67
Similar to the baseflow
separation we calculated cumulative fluxes for the mined and
unmined areas for the entire water year and broken up into
baseflow and stormflow periods using the Hewlett and Hibbert
baseflow separation.
■RESULTS
Spatial Analysis. The size of LB changed from 99 ha
premining to 68 ha postmining, a 31% reduction in area (Figure
1c−f). MR increased in size from 3582 ha premining to 3672
ha after mining. Both watersheds experienced a large reduction
in mean slope; the mean slope in LB decreased from 20.5°to
13.3°, the slopes in MR decreased from 21.1°to 17.3°. The
mean slopes in the reference watersheds RB and LF are 19.5°
and 17.5°, respectively. We estimate that 10−14 million m3of
mine spoil were deposited in VFs in LB, while the VFs in the
larger MR watershed contain 162−185 million m3of
overburden. Spread out over the watersheds the crushed rock
material would cover LB about 15 m and MR 4 m deep.
Hydrology. Precipitation for the study period ranged from
1254 mm in RB to 1358 mm in MR. The 104 mm difference
between the small reference watershed and the larger mined
watershed is likely due a 75 m difference in mean elevation
between the two watersheds and is consistent with the 1981−
2010 PRISM data that indicate an average 51 mm difference
between the two watersheds. Precipitation in Charleston, WV,
was 1166 mm for the 2015 WY (data provided by the Utah
Climate Center). The 1996−2015 annual mean at this station
(Charleston WSFO) is 1191 mm, which makes the 2015 WY
an average precipitation year.
Runoffin the mined first-order watershed was 68 mm
(11.2%) higher than runoffin the first-order reference
watershed (677 and 609 mm, respectively), and runoffin the
mined fourth-order watershed was 40 mm (7.3%) higher than
in the fourth-order reference watershed (585 mm and 545 mm,
respectively; see Table 1).
Table 1. Precipitation, Runoff, And RunoffRatios for the
2015 Water Year for the Four Study Watersheds as Well as
Baseflow and Event Flow Proportions Derived from Three
Different Empirical Baseflow Separation Methods
RB (1st
order
reference)
LB (1st
order
mined)
LF (4th
order
reference)
MR7 (4th
order
mined)
precipitation
(mm) 1254 1339 1293 1358
runoff(mm) 609 677 545 585
runoffRatio (−) 0.49 0.51 0.42 0.43
Hewlett and
Hibbert (1967)
baseflow 0.30 0.71 0.41 0.69
event
flow 0.70 0.29 0.59 0.31
Pettyjohn and
Henning
(1979)
baseflow 0.30 0.72 0.33 0.65
event
flow 0.70 0.28 0.67 0.35
Blume et al.
(2007)
baseflow 0.29 0.75 0.30 0.71
event
flow 0.71 0.25 0.70 0.29
Environmental Science & Technology Article
DOI: 10.1021/acs.est.7b02288
Environ. Sci. Technol. 2017, 51, 8324−8334
8327
The hydrographs demonstrate differences in hydrologic
response between the mined and reference watersheds, with
both reference watersheds exhibiting generally higher peakflows
than the mined watersheds during runoffevents (Figure 2).
The flow duration curves (FDCs) further highlight the mining
impact for both mined sites with increased low flows and
attenuated high flows (Figure 3, top panel). Flow in RB ceased
several times during the growing season (Figure 3, left top
panel), while the slightly smaller LB sustained streamflow
throughout the year.
Water export in the mined and reference first-order
watersheds was largely similar from December 10 through
April 30 (∼415 mm). However, the reference watershed
exported more water during larger runoffevents, while the
mined watershed exported more water following events
(enhanced hydrograph recessions) (Figure 2 and Figure 3
middle and bottom panels). The greatest differences occurred
from May 1 through the end of the study period in early
October, when the mined headwater−watershed exported 2.4
times more water than the headwater reference site (184 mm
from LB and 77 mm from RB). The mined watershed also
exported more water over the first 1.5 months of the study
period following the previous year’s low-flow period. The
dynamics in the fourth-order watersheds were generally similar
to the first-order watersheds, while the overall difference in Q
after 12 months was lower than in the first-order watersheds
(Figure 3).
The baseflow separations yielded consistent results for all
three baseflow separation methods and across watershed scales
(Table 1 and Figure 4). The baseflow portion across all three
methods in the reference watersheds was ∼30% of annual
streamflow in the small reference watershed and 35% in the
larger reference watershed, while baseflow constituted ∼73%
and ∼68% of annual streamflow in the small mined watershed
and the larger mined watershed, respectively.
Specific Conductance. Specific conductance (SC) was on
average 10 times (MR to LF) to 25 times (LB to RB) higher in
the mined watersheds than in the associated reference
watersheds (Figure 2 and Table 2). The lowest SC values
were observed in RB (<10 μS/cm), where SC never exceeded
111 μS/cm. SC in the fourth-order reference watershed was
slightly higher and more variable than the values for RB but
never exceeded 195 μS/cm. The highest SC values were
measured in LB, ranging from 660 to 1977 μS/cm (omitting
the period of greatest pump influence from 06/27/2015
through 07/20/2015). MR, the partially mined fourth-order
watershed, had SC values that ranged from a minimum value of
53 μS/cm during a major winter storm to a maximum of 1705
μS/cm during summer baseflows. In all watersheds, SC
Figure 2. Precipitation (P, top panel), runoff(Q, solid lines), and
specific conductance (SC, dotted lines) for the four watersheds. Mined
watersheds are denoted in red, unmined/reference watersheds in blue
hues. The gray shading denotes the time period when the mining
company pumped water out of the small sedimentation pond below
the valley fill and does not represent a natural decrease in conductivity.
Interactive versions of the figures and additional information
accompanying this publication can be found at https://mtm-hydro.
web.duke.edu/.
Figure 3. Flow duration curves (FDCs) for the first-order watersheds
(left column) and fourth-order watersheds (right column). The insets
are enlarged sections of the high flows denoted by the black rectangles
(top panel); cumulative runofffor the first-order watersheds (left
column) and fourth-order watersheds (right column) (middle panel);
runoffdifference between mined and reference watershed for the first-
order watersheds (left column) and fourth-order watersheds (right
column) (bottom panel). Note that the shaded portion in the top left
panel represents the time periods affected by the mining company
pumping water out of the retention pond below the valley fill and does
not represent a natural decrease in streamflow.
Environmental Science & Technology Article
DOI: 10.1021/acs.est.7b02288
Environ. Sci. Technol. 2017, 51, 8324−8334
8328
decreased during runoffevents. In the unmined watersheds,
event flows could dilute SC to as low as 8 μS/cm or RB and 19
μS/cm for LF. In contrast, even the largest storms were unable
to dilute the mining associated SC signal in LB to a similar
degree, where even during the highest flow event SC remained
above 650 μS/cm. Stormwater dilution in the larger mined
watershed was more effective than in in the small mined
watershed, diluting SC to as low as 53 μS/cm during the largest
storms (Figure 2). In all cases, SC values recovered rapidly to
pre-event levels. With the exception of storms, there was little
seasonal variation in SC for the small mined watershed, with SC
near ∼1500 μS/cm for all of the year. In contrast, SC varied
seasonally in the larger mined watershed, shifting from dormant
season values of ∼1000 μS/cm SC to highs near 1500 μS/cm
for the majority of the growing season (Figure 2).
This seasonal variation in specific conductance in the fourth
order Mud River is caused by a shift in the relative contribution
of mined and unmined portions of the watershed over the
water year. Hydrograph separations were used to determine
that mined areas contributed slightly more water to the overall
annual runoffthan unmined areas (53% vs 47%, respectively,
Figure 5, top panel), despite making up a smaller proportion of
the watershed (46% of the MR watershed is in MTMVFs). In
addition to having a higher water yield, the mined areas of MR
exported more water during the drier growing season and the
unmined portions exported more water during the dormant
season (Figure 5, bottom panels). During baseflow periods,
64% of runofforiginated from mined areas, but that percentage
decreased to just 44% during stormflow periods, indicating a
shift in contributions from mining-dominated baseflow periods
to stormflow periods dominated by runofffrom the unmined
portions (Figure 6). During the most extended baseflow
periods (e.g., 05/12/2015−06/27/2015) contributions from
the mined areas increased to 94% of total flow as unmined
headwaters ran dry. Only during the wettest portion of the year
(e.g., 02/21/2015−04/17/2015) did contributions from the
mined areas fall to levels (46% of total flow) that were
equivalent to their areal extent. At the peak of stormflows,
contributions from the mined areas frequently dropped below
30% and fell to the annual minimum of 8% during the year’s
largest storm (Figure 5, top panel).
■DISCUSSION
In our study, we found that watersheds affected by mountain-
top removal coal mining with valley fills (MTMVF) had
reduced stormflows and enhanced baseflows relative to
reference watersheds. In these MTMVF impacted watersheds,
both baseflows and stormflows export large quantities of total
dissolved solids derived from strong acid weathering of
carbonate bedrock. Because of their elevated baseflow,
MTMVF watersheds contribute disproportionally to the flow
of downstream rivers during low flow periods. These significant
alterations of both watershed hydrology and water chemistry
are likely to lead to both more perennial and saltier streamflows
throughout Appalachia where at least 7% of the ecoregion has
been converted to MTMVF mines.
24
Baseflow/Stormflow Ratios. Our paired watershed
analysis documented significant reductions in stormflow and
Figure 4. Hydrographs and baseflow separation with constant slope method (Hewlett and Hibbert, 1967)
55
for first-order watersheds (top half) and
fourth-order watersheds (bottom half). Reference watersheds are depicted in blue; mined watersheds in red.
Table 2. Specific Conductance (SC) Statistics for the Four
Experimental Watersheds
a
SC (μS/cm) RB
(unmined) LB
(mined) LF
(unmined) MR7
(mined)
mean 58 1504 102 1053
median 52 1530 89 1005
standard dev 20 198 43 367
minimum 8 660 19 53
maximum 111 1977 195 1705
a
LB statistics were calculated omitting the period of greatest pump
influence (06/27/2015−07/20/2015).
Environmental Science & Technology Article
DOI: 10.1021/acs.est.7b02288
Environ. Sci. Technol. 2017, 51, 8324−8334
8329
enhanced baseflow as a result of MTMVF activities. These
findings were robust, with similar proportional changes in
baseflow/stormflow ratios in the first-order and fourth-order
watershed pairs. These results support the suggestion that
valley fills lead to massive increases in porosity and water
holding capacity
26,68
and that as a result valley fills have much
greater impacts on downstream hydrology than surface
compaction during mine reclamation.
The effect of MTMVF streamflow was almost equally strong
for both the first- and fourth-order watersheds, despite the
difference in watershed size and the fraction of the watershed
impacted by mining. For both sets of watersheds, stormwater
runoffwas substantially lower for the mined watershed and high
baseflow contributions from the MTMVF watersheds suggest
increased infiltration into deep valley fill storage, from which
the water then slowly drains. Our findings that MTMVF
increases baseflow in both of these watersheds are consistent
with earlier studies in the region reporting increased runoff
ratios
30
and higher baseflows
31
from mined watersheds
elsewhere in West Virginia. The current study improves upon
these earlier studies by performing comparisons and hydro-
graph separations on paired watersheds in which the only
mining impacts are MTMVF, thus greatly enhancing our ability
to connect MTM to observed changes in hydrology and
biogeochemistry. Our observations that mining reduces storm-
flows contrast with prior work in which Negley and Eshleman
33
documented increased stormflows from several surface coal
mines in western Maryland. The difference between the
Maryland study and our study watersheds is easy to explain,
as Negley and Eshleman
33
watersheds did not include valley
fills. Negley and Eshleman
33
attributed the hydrologic alteration
in their study to increased overland runoffresulting from
surface compaction during mine reclamation. While to some
extent this mechanism may be acting in our WV mines, both
their hydrology and chemistry suggest increased infiltration into
deep valley fill storage, from which the water then slowly drains.
A notable attribute of mountaintop mined landscapes is the
emergence of flat areas
69
that are rare in the steep Appalachian
mountains. These newly created flat areas favor enhanced
infiltration due to low slope gradients, at least partially
offsetting the influence of surface compaction. Therefore,
instead of increasing stormflow because of surface compaction,
the valley fills increase the baseflow portion of total streamflow.
Additionally, preferential flowpaths along the spoil-bedrock
interface could enhance infiltration into the VF.
70
Unfortu-
nately, little published research provides insight on the internal
structure of valley fills and how settlement or sorting of material
may affect hydrologic flowpathways.
68,70,71
Greer et al.,
72
for
example, demonstrated high subsurface heterogeneity in a
valley fill in Virginia using electrical resistivity imaging. It is
reasonable to assume that the physical characteristics of the VFs
affect how much water can be stored in the VFs and how the
stored water is subsequently released to sustain streamflow.
Ross et al.
26
determined large variability in VF area, depth, and
volume among >1500 VFs in Central Appalachia. While we
reference the increased storage in the VFs as reason for the
baseflow increases, it is unfortunately not possible at this point
to make quantitative assessments on how different VF
characteristics would influence the hydrologic response of
mined watersheds. However, while this may be important for
individual small headwater−watersheds, over larger areas
responses of individual VFs of different sizes would likely be
Figure 5. Hydrograph separation for the partially mined MR watershed. Portions of the hydrograph originating from unmined areas are denoted in
blue; mined area contributions are denoted in red. The gray shading marks the time when the mining company actively pumped water out of the
small sedimentation pond above the LB instrumentation (top panel). Cumulative flux from mined and unmined areas of MR (bottom left panel).
Runoffdifferences between mined area runoffand unmined area runoffin MR (bottom right panel).
Figure 6. Frequency distributions of mined area contributions during
baseflow (red) and eventflow (blue) periods.
Environmental Science & Technology Article
DOI: 10.1021/acs.est.7b02288
Environ. Sci. Technol. 2017, 51, 8324−8334
8330
obscured by the combined response of all VFs present in the
watershed.
Impacts of MTMVF on Watershed Water Balances.
MTMVF watersheds have lower plant biomass and reduced
topographic relief relative to unmined watersheds in the region.
Both the loss of evapotranspiration by vegetation and the
change in runoffand infiltration associated with landscape
flattening are expected to exert strong influence on the annual
water budget.
Differences in Qbetween mined and reference watersheds
are commonly attributed to the lack of vegetation on mined
areas and the associated elimination/reduction of the
transpiration component. Clearcutting vegetation typically
results in decreased ET and subsequent increases in Q.
73−75
The same response might be expected on mined areas due to
deforestation, especially on younger valley fills. Yet the
differences in annual water export between the mined and
reference watersheds in our study (first-order: 68 mm; fourth-
order: 40 mm) were smaller than differences measured between
forested and clear-cut watersheds in other parts of the
Appalachians, e.g., 130 mm reduction after 85% clearcutting
in Fernow, WV,
76
or 150−400 mm after 100% clear-cutting at
Coweeta, NC.
77
While it is near certain that the rates of
evapotranspiration must be lower from recently mined and
deforested landscapes, the relatively small change in annual
water yield suggests that the loss of ET may be compensated
for by other components of MTMVF affecting the water
balance.
Reductions in watershed slope may be counterbalancing this
reduction in ET by increasing infiltration and water residence
times. Ross, McGlynn and Bernhardt
26
determined that across
southern West Virginia premining landscapes had a modal
slope of ∼28°, while postmining landscapes exhibited bimodal
slope distributions of ∼2°and ∼20°. Annual runoffratios are
typically positively correlated with watershed slope,
78,79
while
decreased slopes are typically associated with greater infiltration
into the subsurface and longer water residence times.
79,80
In
vegetated watersheds, these longer water residence times
should increase the potential for water uptake by vegetation
and subsequent losses through evapotranspiration.
78
We
suspect that the lower topographic relief coupled with reduced
evapotranspiration in MTMVF watersheds is changing the
flowpaths and the residence time of water in mined watersheds
without fundamentally altering the total water yield.
Contributions to Streamflow from Mined and
Unmined Areas. The spatial configuration of our study
watersheds allowed for separation of MR streamflow into
contributions from mined and unmined areas, using the first-
order mined and reference watersheds as end members. The
constantly high base SC values in LB, even during the wet
dormant season, suggest that baseflow in the mined water-
shed−or rather VF−is generated from a deeper water source
within the VF that is largely unaffected by incoming
precipitation. Dilution occurred during runoffevents, but
even then, the SC values in LB remained above 650 μS/cm.
These sustained high SC values suggest limited overland flow
on mine soil, which is contrary to previous conclusions about
the role of overland flow in surface mining environments
without VFs.
33,81
The runofffrom mined areas that contributed
to streamflow could be displaced water with varying
concentrations from within the valley fill that has not reached
maximum SC values, similar to differences often observed
between groundwater and soil water in undisturbed sys-
tems
82,83
or precipitation that infiltrated into the VF and
dissolved readily available solutes while moving rapidly via
preferential flowpathways.
In the partially mined fourth-order MR watershed, runoff
from the mined areas was 53% of annual runoff, which is
slightly greater than the fraction of the watershed that was
mined (∼46%). This is consistent with our finding that the
mined watersheds exhibited greater runoffthan the reference
watersheds, but less than would be expected if the mined areas
had simply been clear-cut (see the discussion point on water
balance comparisons). The hydrograph separation in the
partially mined watershed (MR) corroborates that reference
watersheds export more water during the wetter dormant
season (64% Qfrom unmined areas), with peak contributions
from unmined areas exceeding 80% of total streamflow. The
brief rise in SC immediately coincident with streamflow
increases (Figure 5, top panel) is likely caused by the spatial
arrangement of mined and unmined areas, with the mined areas
being closer located to the watershed outlet (Figure 1a).
Because of thisand especially during the wetter periodsthe
stream received brief inputs of mined water only (which is itself
diluted but still higher in SC than the MR streamwater) until
the runofffrom the unmined areas further upstream travels to
the watershed outlet.
During baseflow periods the majority of MR streamflow
originated from mined areas (64% Qfrom mined areas). High
contributions during long baseflow periods (up to 94%) suggest
that the unmined areas in MR contribute little water to
streamflow during the growing season, similar to the reference
sites LB and LF that frequently fall dry during the growing
season after longer periods without precipitation. This
highlights the strong effect that MTMVF runoffcan exert on
water quality and quantity, especially during low-flow periods
when it can be the dominant source of streamflow downstream.
Implications. This study highlights and further demon-
strates the cascading effects that mountaintop mining has on
the immediate location of the disturbance (i.e., the disturbed
areas themselves) as well as the surrounding ecosystems (in this
case downstream areas). The changes to the hydrologic
responses to rainfall and the seasonality of streamflow are
indicators of this massive critical zone disturbance. While the
hydrologic impacts of most disturbances are rather easily
identified and often predictable, assessing the balance of the
opposing effects associated with MTMVF can be challenging.
For example, deforestation (through insect infestations, wild-
fires, clear-cutting, etc.) typically lead to increases in annual Q
through reduced evapotranspiration
77
and urbanization or
decreased infiltration rates typically results in flashier hydro-
graphs and an increase in stormflow and associated reduction in
baseflow.
84
The effect of other forms of surface mining without
valley fills often resemble the effects of urbanization.
36,85
In
contrast, the effect of MTMVF with valley fills on simple
hydrologic response is perhaps more comparable to the effect
of dams on riverine systems, since dams typically are designed
or managed to reduce high flows and increase low flows.
9,86−88
However, the effect on hydrologic response is achieved via
completely different mechanisms. While damming impacts
hydrology by placing a structure within the river network and
directly regulating the stream/river, MTMVF can alter the
critical zone of entire landscapes hundreds of meters deep. This
deep impact thereby dramatically changes the runoffgeneration
processes themselves, i.e. how water moves through the system
once it reaches the ground surface. The consequences are both
Environmental Science & Technology Article
DOI: 10.1021/acs.est.7b02288
Environ. Sci. Technol. 2017, 51, 8324−8334
8331
an altered hydrologic regime as well as degradation of
streamwater quality through the export of weathering products.
Increased baseflow portion in mined watersheds and high
streamwater specific conductance indicate that rainfall spends
more time in the subsurface, especially in the VFs. This has
implications for two key issues: First, the water draining mined
watersheds has been in contact with VF material, with greatly
enhanced weatherable surfaces,
26
for extended periods of time.
This results in increased concentrations of weathering products
that contribute to downstream alkaline mine drainage and thus
impair aquatic ecosystems. This degradation in streamwater
quality following MTMVF and its effect on stream biota−even
decades after mine reclamation−has been documented across
Central Appalachia.
39,41,42,44,45,89,90
Recent research on mine
reclamation techniques, especially reforestation both on current
as well as former mines,
91
promises faster regrowth of native
vegetation on unconsolidated spoil material. While the positive
effect on tree growth has been demonstrated,
92
the effects on
hydrologic response are not as clear. Agouridis et al.
93
for
example measured sharply declining electrical conductivities in
several reforested plots on a mine in Kentucky over a three-year
period after plot establishment. However, the mine spoil plots
were only 2.5 m deep
94
and may not be representative of valley
fill spoils >100 m deep.
Second, enhanced baseflow itself, even in large partially
mined watersheds, can contribute to stream-impairment. Our
hydrograph separations in the partially mined MR watershed
(Figure 5) demonstrate that the water quality influence from
mined areas was most dominant during low flow periods.
Similar runoffrates and patterns between mined and reference
streams during the dormant season (Figures 4 and 5) indicate
that the downstream impact of mining was less dramatic during
the winter high-flow period. During the summer baseflow
period, the majority of streamwater originated from VF outflow
in the fully mined and 46% mined study watersheds. Because of
the disproportionate influence of mined areas on stream
baseflow, the effect of MTMVF on downstream systems would
extend further than a simple area-mixing model
40
would
predict, especially during baseflow periods that constitute
∼80% of the year.
■AUTHOR INFORMATION
Corresponding Author
*Tel: 307-766-5012; fax: 307-766-6403; e-mail: fnippgen@
uwyo.edu.
ORCID
Fabian Nippgen: 0000-0002-7428-9375
Notes
The authors declare no competing financial interest.
■ACKNOWLEDGMENTS
This research was funded by NSF Grant No. EAR-1417405 to
B.L.M. and E.S.B. and a NSF GRFP to M.R.V.R. Logistical
support was provided by staffof WV DNR District 5’s Upper
Mud River office. We thank Nick Huffman and Eric Moore for
help with field data collection and Anita and Stanley Miller for
granting us access to their property.
■REFERENCES
(1) Broushaki, F.; Thomas, M. G.; Link, V.; López, S.; van Dorp, L.;
Kirsanow, K.; Hofmanová
, Z.; Diekmann, Y.; Cassidy, L. M.; Díez-del-
Molino, D.; Kousathanas, A.; Sell, C.; Robson, H. K.; Martiniano, R.;
Blöcher, J.; Scheu, A.; Kreutzer, S.; Bollongino, R.; Bobo, D.; Davudi,
H.; Munoz, O.; Currat, M.; Abdi, K.; Biglari, F.; Craig, O. E.; Bradley,
D. G.; Shennan, S.; Veeramah, K. R.; Mashkour, M.; Wegmann, D.;
Hellenthal, G.; Burger, J. Early Neolithic genomes from the eastern
Fertile Crescent. Science 2016,353,1−15.
(2) Riehl, S.; Zeidi, M.; Conard, N. J. Emergence of Agriculture in the
Foothills of the Zagros Mountains of Iran. Science 2013,341 (6141),
65−67.
(3) Achard, F.; Eva, H. D.; Stibig, H.-J.; Mayaux, P.; Gallego, J.;
Richards, T.; Malingreau, J.-P. Determination of Deforestation Rates of
the World’s Humid Tropical Forests. Science 2002,297 (5583), 999−
1002.
(4)Pongratz,J.;Reick,C.;Raddatz,T.;Claussen,M.A
reconstruction of global agricultural areas and land cover for the last
millennium. Global Biogeochemical Cycles 2008,DOI: 10.1029/
2007GB003153.
(5) Townsend, P. A.; Helmers, D. P.; Kingdon, C. C.; McNeil, B. E.;
de Beurs, K. M.; Eshleman, K. N. Changes in the extent of surface
mining and reclamation in the Central Appalachians detected using a
1976−2006 Landsat time series. Remote Sensing of Environment 2009,
113 (1), 62−72.
(6) Dudka, S.; Adriano, D. C. Environmental impacts of metal ore
mining and processing: A review. Journal of Environmental Quality
1997,26 (3), 590−602.
(7) Schindler, D. Tar sands need solid science. Nature 2010,468
(7323), 499−501.
(8) Vörösmarty, C. J.; Meybeck, M.; Fekete, B.; Sharma, K.; Green,
P.; Syvitski, J. P. M. Anthropogenic sediment retention: major global
impact from registered river impoundments. Global and Planetary
Change 2003,39 (1−2), 169−190.
(9) Graf, W. L. Downstream hydrologic and geomorphic effects of
large dams on American rivers. Geomorphology 2006,79 (3−4), 336−
360.
(10) Meyer, W. B.; Turner, B. L. Human-Population Growth And
Global Land-Use Cover Change. Annu. Rev. Ecol. Syst. 1992,23,39−
61.
(11) Hesse, R. Geomorphological traces of conflict in high-resolution
elevation models. Applied Geography 2014,46,11−20.
(12) Hupy, J. P.; Schaetzl, R. J. Soil development on the WWI
battlefield of Verdun, France. Geoderma 2008,145 (1−2), 37−49.
(13) Brantley, S. L.; Goldhaber, M. B.; Ragnarsdottir, K. V. Crossing
disciplines and scales to understand the Critical Zone. Elements 2007,
3(5), 307−314.
(14) Foley, J. A.; DeFries, R.; Asner, G. P.; Barford, C.; Bonan, G.;
Carpenter, S. R.; Chapin, F. S.; Coe, M. T.; Daily, G. C.; Gibbs, H. K.;
Helkowski, J. H.; Holloway, T.; Howard, E. A.; Kucharik, C. J.;
Monfreda, C.; Patz, J. A.; Prentice, I. C.; Ramankutty, N.; Snyder, P. K.
Global Consequences of Land Use. Science 2005,309 (5734), 570−
574.
(15) Peterson, B. J.; Wollheim, W. M.; Mulholland, P. J.; Webster, J.
R.; Meyer, J. L.; Tank, J. L.; Martí, E.; Bowden, W. B.; Valett, H. M.;
Hershey, A. E.; McDowell, W. H.; Dodds, W. K.; Hamilton, S. K.;
Gregory, S.; Morrall, D. D. Control of Nitrogen Export from
Watersheds by Headwater Streams. Science 2001,292 (5514), 86−90.
(16) McGlynn, B. L.; McDonnell, J. J. Role of discrete landscape
units in controlling catchment dissolved organic carbon dynamics.
Water Resour. Res. 2003,39 (4), SWC 3-1−SWC 3-18.
(17) Konar, M.; Todd, M. J.; Muneepeerakul, R.; Rinaldo, A.;
Rodriguez-Iturbe, I. Hydrology as a driver of biodiversity: Controls on
carrying capacity, niche formation, and dispersal. Adv. Water Resour.
2013,51, 317−325.
(18) Bunn, S. E.; Arthington, A. H. Basic principles and ecological
consequences of altered flow regimes for aquatic biodiversity. Environ.
Manage. 2002,30 (4), 492−507.
(19) Gaffield, S. J.; Goo, R. L.; Richards, L. A.; Jackson, R. J. Public
health effects of inadequately managed stormwater runoff. Am. J. Public
Health 2003,93 (9), 1527−1533.
Environmental Science & Technology Article
DOI: 10.1021/acs.est.7b02288
Environ. Sci. Technol. 2017, 51, 8324−8334
8332
(20) Wellen, C. C.; Shatilla, N. J.; Carey, S. K. Regional scale
selenium loading associated with surface coal mining, Elk Valley,
British Columbia, Canada. Sci. Total Environ. 2015,532, 791−802.
(21) Yang, X. J.; Lin, A.; Li, X.-L.; Wu, Y.; Zhou, W.; Chen, Z.
China’s ion-adsorption rare earth resources, mining consequences and
preservation. Environmental Development 2013,8, 131−136.
(22) Slonecker, E. T.; Benger, M. J. Remote sensing and
mountaintop mining. Remote Sensing Reviews 2001,20 (4), 293−322.
(23) DEP. Standards and Specifications for Erosion and Sediment
Control Excess Spoil Disposal Haulageways; Department of Environ-
mental Protection Division of Mining and Reclamation: 1993.
(24) U.S. EPA. The Effects of Mountaintop Mines and Valley Fills on
Aquatic Ecosystems of the Central Appalachian Coalfields; United States
Environmental Protection Agency: Washington, DC, 2011.
(25) Ebel, B. A.; Mirus, B. B. Disturbance hydrology: challenges and
opportunities. Hydrological Processes 2014,28 (19), 5140−5148.
(26) Ross, M. R. V.; McGlynn, B. L.; Bernhardt, E. S. Deep Impact:
Effects of Mountaintop Mining on Surface Topography, Bedrock
Structure, and Downstream Waters. Environ. Sci. Technol. 2016,50 (4),
2064−2074.
(27) Miller, A. J.; Zegre, N. P. Mountaintop Removal Mining and
Catchment Hydrology. Water 2014,6(3), 472−499.
(28) Zegre, N. P.; Maxwell, A.; Lamont, S. Characterizing streamflow
response of a mountaintop-mined watershed to changing land use.
Applied Geography 2013,39,5−15.
(29) Messinger, T. Comparison of Storm Response of Streams in
Small, Unmined and Valley-Filled Watersheds, 1999−2001, Ballard
Fork, West Virginia. Water-Resources Investigations Report 02-4303; U.S.
Geological Survey: Charleston, WV, 2003.
(30) Messinger, T.; Paybins, K. S. Relations Between Precipitation and
Daily and Monthly Mean Flows in Gaged, Unmined and Valley-Filled
Watersheds, Ballard Fork, West Virginia, 1999−2001; U.S. Department
of the Interior; U.S. Geological Survey: Charleston, WV, 2003.
(31) Zegre, N. P.; Miller, A. J.; Maxwell, A.; Lamont, S. J. Multiscale
Analysis Of Hydrology In A Mountaintop Mine-Impacted Watershed.
J. Am. Water Resour. Assoc. 2014,50 (5), 1257−1272.
(32) Evans, D. M.; Zipper, C. E.; Hester, E. T.; Schoenholtz, S. H.
Hydrologic Effects of Surface Coal Mining in Appalachia (US). J. Am.
Water Resour. Assoc. 2015,51 (5), 1436−1452.
(33) Negley, T. L.; Eshleman, K. N. Comparison of stormflow
responses of surface-mined andforestedwatershedsinthe
Appalachian Mountains, USA. Hydrol. Processes 2006,20 (16),
3467−3483.
(34) Simmons, J. A.; Currie, W. S.; Eshleman, K. N.; Kuers, K.;
Monteleone, S.; Negley, T. L.; Pohlad, B. R.; Thomas, C. L. Forest to
reclaimed mine land use change leads to altered ecosystem structure
and function. Ecological Applications 2008,18 (1), 104−118.
(35) McCormick, B. C.; Eshleman, K. N.; Griffith, J. L.; Townsend,
P. A. Detection of flooding responses at the river basin scale enhanced
by land use change. Water Resour. Res. 2009,45,1−15.
(36) Ferrari, J. R.; Lookingbill, T. R.; McCormick, B.; Townsend, P.
A.; Eshleman, K. N., Surface mining and reclamation effects on flood
response of watersheds in the central Appalachian Plateau region.
Water Resour. Res. 2009,45.10.1029/2008WR007109
(37) Miller, A. J.; Zegre, N. P. Landscape-Scale Disturbance: Insights
into the Complexity of Catchment Hydrology in the Mountaintop
Removal Mining Region of the Eastern United States. Land 2016,5
(3), 22.
(38) Griffith, M. B.; Norton, S. B.; Alexander, L. C.; Pollard, A. I.;
LeDuc, S. D. The effects of mountaintop mines and valley fills on the
physicochemical quality of stream ecosystems in the central
Appalachians: A review. Sci. Total Environ. 2012,417,1−12.
(39) Lindberg, T. T.; Bernhardt, E. S.; Bier, R.; Helton, A. M.;
Merola, R. B.; Vengosh, A.; Di Giulio, R. T. Cumulative impacts of
mountaintop mining on an Appalachian watershed. Proc. Natl. Acad.
Sci. U. S. A. 2011,108 (52), 20929−20934.
(40) Bernhardt, E. S.; Lutz, B. D.; King, R. S.; Fay, J. P.; Carter, C. E.;
Helton, A. M.; Campagna, D.; Amos, J. How Many Mountains Can
We Mine? Assessing the Regional Degradation of Central Appalachian
Rivers by Surface Coal Mining. Environ. Sci. Technol. 2012,46 (15),
8115−8122.
(41) Palmer, M. A.; Bernhardt, E. S.; Schlesinger, W. H.; Eshleman,
K. N.; Foufoula-Georgiou, E.; Hendryx, M. S.; Lemly, A. D.; Likens, G.
E.; Loucks, O. L.; Power, M. E.; White, P. S.; Wilcock, P. R.
Mountaintop Mining Consequences. Science 2010,327 (5962), 148−
149.
(42) Pond, G. J.; Passmore, M. E.; Borsuk, F. A.; Reynolds, L.; Rose,
C. J. Downstream effects of mountaintop coal mining: comparing
biological conditions using family- and genus-level macroinvertebrate
bioassessment tools. Journal of the North American Benthological Society
2008,27 (3), 717−737.
(43) Timpano, A. J.; Schoenholtz, S. H.; Soucek, D. J.; Zipper, C. E.
Salinity As A Limiting Factor For Biological Condition In Mining-
Influenced Central Appalachian Headwater Streams. J. Am. Water
Resour. Assoc. 2015,51 (1), 240−250.
(44) Bier, R. L.; Voss, K. A.; Bernhardt, E. S. Bacterial community
responses to a gradient of alkaline mountaintop mine drainage in
Central Appalachian streams. ISME J. 2015,9(6), 1378−1390.
(45) Voss, K. A.; King, R. S.; Bernhardt, E. S. From a line in the sand
to a landscape of decisions: a hierarchical diversity decision framework
for estimating and communicating biodiversity loss along anthro-
pogenic gradients. Methods in Ecology and Evolution 2015,6(7), 795−
805.
(46) Murphy, J. C.; Hornberger, G. M.; Liddle, R. G. Concentration-
discharge relationships in the coalmined region of the New River basin
and Indian Fork sub-basin, Tennessee, USA. Hydrological Processes
2014,28 (3), 718−728.
(47) NRCS. Soil Survey of Lincoln County, West Virginia; United
States Department of Agriculture; Natural Resources Conservation
Service in cooperation with the West Virginia Agricultural and
Forestry Experiment Station and West Virginia Conservation Agency,
2007.
(48) NRCS. Soil Survey Geographic Database (SSURGO 2.2); Natural
Resources Conservation Service, U.S. Department of Agriculture,
2015.
(49) Nicholson, S. W.; Dicken, C. L.; Horton, J. D.; Labay, K. A.;
Foose, M. P.; Mueller, J. A. L., Preliminary integrated geologic map
databases for the United States: Kentucky, Ohio, Tennessee, and West
Virginia; US Geological Survey: 2005.
(50) Braun, E. L. Deciduous forests of eastern North America; Free
Press: New York, 1974.
(51) PRISM. PRISM Gridded Climate Data; OSU PRISM Climate
Group, 2016.
(52) Adams, M.; Kochenderfer, J.; Edwards, P. The Fernow
Watershed Acidification Study: Ecosystem Acidification, Nitrogen
Saturation and Base Cation Leaching. Water, Air, Soil Pollut.: Focus
2007,7(1−3), 267−273.
(53) Searcy, J. K.; Hardison, C. H. Double Mass Curves. Geological
Survey Water-Supply Paper 1541-B; United States Government
Printing Office: 1960.
(54) Manning, R.; Griffith, J. P.; Pigot, T.; Vernon-Harcourt, L. F. On
the flow of water in open channels and pipes; Dublin, 1890.
(55) Hewlett, J. D.; Hibbert, A. R. Factors affecting the response of
small watersheds to precipitation in humid areas. Forest Hydrology
1966, 275−291.
(56) Wolock, D. M., Base-flow index grid for the conterminous United
States, 03-263 ed.; U.S. Geological Survey: Reston, VA, 2003.
(57) Pettyjohn, W. A.; Henning, R. Preliminary Estimate of Ground-
Water Recharge Rates, Related Streamflow and Water Quality in Ohio;
Water Resources Center, The Ohio State University: Columbus, 1979.
(58) Blume, T.; Zehe, E.; Bronstert, A. Rainfall-runoff response,
event-based runoff coefficients and hydrograph separation. Hydrol. Sci.
J. 2007,52 (5), 843−862.
(59) Pinder, G. F.; Jones, J. F. Determination of the ground-water
component of peak discharge from the chemistry of total runoff. Water
Resour. Res. 1969,5(2), 438−445.
Environmental Science & Technology Article
DOI: 10.1021/acs.est.7b02288
Environ. Sci. Technol. 2017, 51, 8324−8334
8333
(60) Sklash, M. G.; Farvolden, R. N.; Fritz, P. A conceptual model of
watershed response to rainfall, developed through the use of oxygen-
18 as a natural tracer. Can. J. Earth Sci. 1976,13, 271−283.
(61) Buttle, J. M. Isotope hydrograph separations and rapid delivery
of pre-event water from drainage basins. Progress in Physical Geography
1994,18 (1), 16−41.
(62) McGlynn, B. L.; McDonnell, J. J.; Seibert, J.; Kendall, C. Scale
effects on headwater catchment runoff timing, flow sources, and
groundwater-streamflow relations. Water Resour. Res. 2004,40 (7), 1−
14.
(63) Caissie, D.; Pollock, T. L.; Cunjak, R. A. Variation in stream
water chemistry and hydrograph separation in a small drainage basin. J.
Hydrol. 1996,178 (1−4), 137−157.
(64) Jencso, K. G.; McGlynn, B. L.; Gooseff, M. N.; Bencala, K. E.;
Wondzell, S. M. Hillslope hydrologic connectivity controls riparian
groundwater turnover: Implications of catchment structure for riparian
buffering and stream water sources. Water Resour. Res. 2010,46,1−18.
(65) Laudon, H.; Slaymaker, O. Hydrograph separation using stable
isotopes, silica and electrical conductivity: an alpine example. J. Hydrol.
1997,201 (1−4), 82−101.
(66) Pellerin, B. A.; Wollheim, W. M.; Feng, X.; Vorosmarty, C. J.
The application of electrical conductivity as a tracer for hydrograph
separation in urban catchments. Hydrol. Processes 2008,22 (12),
1810−1818.
(67) Kobayashi, D. Separation of a snowmelt hydrograph by specific
conductance. J. Hydrol. 1986,84 (1−2), 157−165.
(68) Wunsch, D. R.; Dinger, J. S.; Graham, C. D. R. Predicting
ground-water movement in large mine spoil areas in the Appalachian
Plateau. Int. J. Coal Geol. 1999,41 (1−2), 73−106.
(69) Maxwell, A.; Strager, M. Assessing landform alterations induced
by mountaintop mining. Nat. Sci. 2013,5, 229−237.
(70) Dinger, J. S.; Wunsch, D. R.; Kemp, J. E. Occurrence of
Groundwater in Mine Spoil, a Renewable Resource: Star Fire Tract,
Eastern Kentucky, Mining and Reclamation Conferene and Exhibition,
Charleston, WV, 1990; Charleston, WV, 1990.
(71) Wunsch, D. R.; Dinger, J. S.; Taylor, P. B.; Carey, D. I.; Graham,
C. D. R. Hydrogeology, Hydrogeochemistry, and Spoil Settlement at a
Large Mine-Spoil Area In Eastern Kentucky: Star Fire Tract; Kentucky
Geological Survey, University of Kentucky, Lexington, 1996.
(72) Greer, B. M.; Burbey, T. J.; Zipper, C. E.; Hester, E. T. Electrical
resistivity imaging of hydrologic flow through surface coal mine valley
fills with comparison to other landforms. Hydrological Processes 2017,
31 (12), 2244−2260.
(73) Hewlett, J. D.; Helvey, J. D. Effects of Forest Clear-Felling on
Storm Hydrograph. Water Resour. Res. 1970,6(3), 768−782.
(74) Moore, R. D.; Wondzell, S. M. Physical hydrology and the
effects of forest harvesting in the Pacific Northwest: A review. J. Am.
Water Resour. Assoc. 2005,41 (4), 763−784.
(75) Bosch, J. M.; Hewlett, J. D. A review of catchment experiments
to determine the effect of vegetation changes on water yield and
evapotranspiration. J. Hydrol. 1982,55 (1-4), 3−23.
(76) Reinhart, K. G.; Eschner, A. R.; Trimble, G. R., Jr. Effect On
Streamflow Of Four Forest Practices In The Mountains Of West Virginia;
Northeast Forest Experiment Station: Upper Darby, PA, 1963.
(77) Hewlett, J. D.; Hibbert, A. R. Increases In Water Yield After
Several Types Of Forest Cutting. International Association of Scientific
Hydrology. Bulletin 1961,6(3), 5−17.
(78) Nippgen, F.; McGlynn, B. L.; Emanuel, R. E.; Vose, J. M.
Watershed memory at the Coweeta Hydrologic Laboratory: The effect
of past precipitation and storage on hydrologic response. Water Resour.
Res. 2016,52 (3), 1673−1695.
(79) Nippgen, F.; McGlynn, B. L.; Marshall, L. A.; Emanuel, R. E.
Landscape structure and climate influences on hydrologic response.
Water Resour. Res. 2011,47,1−17.
(80) McGuire, K. J.; McDonnell, J. J.; Weiler, M.; Kendall, C.;
McGlynn, B. L.; Welker, J. M.; Seibert, J. The role of topography on
catchment-scale water residence time. Water Resour. Res. 2005,41 (5),
1−14.
(81) Ritter, J. B.; Gardner, T. W. Hydrologic Evolution Of Drainage
Basins Disturbed By Surface Mining, Central Pennsylvania. Geol. Soc.
Am. Bull. 1993,105 (1), 101−115.
(82) Hendershot, W. H.; Savoie, S.; Courchesne, F. Simulation Of
Stream-Water Chemistry With Soil Solution And Groundwater-Flow
Contributions. J. Hydrol. 1992,136 (1−4), 237−252.
(83) Scanlon, T. M.; Raffensperger, J. P.; Hornberger, G. M.
Modeling transport of dissolved silica in a forested headwater
catchment: Implications for defining the hydrochemical response of
observed flow pathways. Water Resour. Res. 2001,37 (4), 1071−1082.
(84) Dunne, T. Water in environmental planning. W. H. Freeman: San
Francisco, 1978.
(85) McCormick, B. C.; Eshleman, K. N. Assessing Hydrologic
Change in Surface-Mined Watersheds Using the Curve Number
Method. Journal of Hydrologic Engineering 2011,16 (7), 575−584.
(86) Magilligan, F. J.; Nislow, K. H. Changes in hydrologic regime by
dams. Geomorphology 2005,71 (1−2), 61−78.
(87) Singer, M. B. The influence of major dams on hydrology
through the drainage network of the Sacramento River basin,
California. River Research and Applications 2007,23 (1), 55−72.
(88) Poff, N. L.; Olden, J. D.; Merritt, D. M.; Pepin, D. M.
Homogenization of regional river dynamics by dams and global
biodiversity implications. Proc. Natl. Acad. Sci. U. S. A. 2007,104 (14),
5732−5737.
(89) Bernhardt, E. S.; Palmer, M. A., The environmental costs of
mountaintop mining valley fill operations for aquatic ecosystems of the
Central Appalachians. In Year in Ecology and Conservation Biology,
Ostfeld, R. S., Schlesinger, W. H., Eds.; Wiley, 2011; Vol. 1223,pp39−
57.
(90) Pond, G. J.; Passmore, M. E.; Pointon, N. D.; Felbinger, J. K.;
Walker, C. A.; Krock, K. J. G.; Fulton, J. B.; Nash, W. L. Long-Term
Impacts on Macroinvertebrates Downstream of Reclaimed Mountain-
top Mining Valley Fills in Central Appalachia. Environ. Manage. 2014,
54 (4), 919−933.
(91) Burger, J.; Graves, D.; Angel, P.; Davis, V.; Zipper, C. The
Forestry Reclamation Approach; The Appalachian Regional Reforesta-
tion Initiative (ARRI), 2005.
(92) Angel, P. N.; Burger, J. A.; Davis, V. M.; Barton, C. D.; Bower,
M.; Eggerud, S. D.; Rothman, P., The Forestry Reclamation Approach
And The Measure Of Its Success In Appalachia. In National Meeting of
the American Society of Mining and Reclamation; Barnhisel, R. I., Ed.;
ASMR: Lexington, 2009.
(93) Agouridis, C. T.; Angel, P. N.; Taylor, T. J.; Barton, C. D.;
Warner, R. C.; Yu, X.; Wood, C. Water Quality Characteristics of
Discharge from Reforested Loose-Dumped Mine Spoil in Eastern
Kentucky. Journal of Environmental Quality 2012,41 (2), 454−468.
(94) Taylor, T. J.; Agouridis, C. T.; Warner, R. C.; Barton, C. D.;
Angel, P. N. Hydrologic characteristics of Appalachian loose-dumped
spoil in the Cumberland Plateau of eastern Kentucky. Hydrol. Processes
2009,23 (23), 3372−3381.
Environmental Science & Technology Article
DOI: 10.1021/acs.est.7b02288
Environ. Sci. Technol. 2017, 51, 8324−8334
8334
