Journal of Environmental Management 197 (2017) 559e570

Contents lists available at ScienceDirect

Journal of Environmental Management

journal homepage: www.elsevier.com/locate/jenvman

Research article

Dissolved oxygen, stream temperature, and sh habitat response to

environmental water purchases
Sarah E. Null*, Nathaniel R. Mouzon, Logan R. Elmore
Department of Watershed Sciences, Utah State University, Logan, UT., 84322, USA

a r t i c l e i n f o a b s t r a c t

Article history: Environmental water purchases are increasingly used for ecological protection. In Nevada's Walker Basin
Received 1 December 2016 (western USA), environmental water purchases augment streamow in the Walker River and increase
Received in revised form lake elevation of terminal Walker Lake. However, water quality impairments like elevated stream tem-
11 March 2017
peratures and low dissolved oxygen concentrations also limit ecosystems and species, including
Accepted 6 April 2017
federally-threatened Lahontan cutthroat trout. In this paper, we prioritize water volumes and locations
that most enhance water quality for riverine habitat from potential environmental water rights pur-
chases. We monitored and modeled streamows, stream temperatures, and dissolved oxygen concen-
Keywords:
Dissolved oxygen
trations using River Modeling System, an hourly, physically-based hydrodynamic and water quality
Walker river model. Modeled environmental water purchases ranged from average daily increases of 0.11e1.41 cubic
Restoration meters per second (m3/s) during 2014 and 2015, two critically dry years. Results suggest that water
Modeling purchases consistently cooled maximum daily stream temperatures and warmed nightly minimum
Water temperature temperatures. This prevented extremely low dissolved oxygen concentrations below 5.0 mg/L, but
Environmental water quality increased the duration of moderate conditions between 5.5 and 6.0 mg/L. Small water purchases less
Water transfer than approximately 0.71 m3/s per day had little benet for Walker River habitat. Dissolved oxygen
concentrations were affected by upstream environmental conditions, where suitable upstream water
quality improved downstream conditions and vice versa. Overall, this study showed that critically dry
water years degrade environmental water quality and habitat, but environmental water purchases of at
least 0.71 m3/s were promising for river restoration.
Published by Elsevier Ltd.

1. Introduction streamow regulation, water diversions, and water quality im-

Water regulation and diversions from Nevada's Walker River
cause low streamows, high stream temperatures and low dis-
solved oxygen (DO) concentrations, which fragment aquatic habi-
tats and limit native sh species, like Lahontan cutthroat trout (LCT)
(Oncorhynchus clarki henshawi) (USFWS, 2003). LCT are a federally
threatened species (USFWS, 1975) that have been reduced to
headwater streams in the Walker Basin and occupy approximately
3% of their historical habitat in the western Great Basin (Dunham
et al., 1999). Cumulative effects of water diversions have also
impacted terminal Walker Lake, causing lake elevation to fall and
salinity to increase, and extirpating LCT from the lake (Beutel et al.,
2001; Sharpe et al., 2007). The problems in the Walker Basin are
common in arid, semi-arid, and Mediterranean climates, where
* Corresponding author.
E-mail address: sarah.null@usu.edu (S.E. Null).
pairments degrade and fragment cold water sheries and habitats
(Baron et al., 2002). Examples are widespread, including rivers in
Spain (Bae et al., 2016) South America (Grantham et al., 2013), and
throughout the American West (Goniea et al., 2006; Nehlsen et al.,
1991; Moyle et al., 2011).
One Walker Basin restoration strategy is to increase streamow
and Walker Lake elevation through environmental water purchases
(Collopy and Thomas, 2009; Ise and Sunding, 1998). Congress
enacted the Energy and Water Development Appropriations Act
(H.R. 2419) in 2006 to acquire environmental water from willing
sellers (Collopy and Thomas, 2009). Purchased or leased water may
be available as farmers upgrade to more efcient irrigation infra-
structure (Sunding et al., 2002) or switch to less water-intensive
crops (Bishop et al., 2010). To date, nearly 25 million cubic meters
(Mm3) (20,000 acre feet) of water have been purchased for
instream uses and to restore Walker Lake, approximately one third
of the goal (NFWF, 2016). Environmental water purchases (also

http://dx.doi.org/10.1016/j.jenvman.2017.04.016
0301-4797/Published by Elsevier Ltd.
560 S.E. Null et al. / Journal of Environmental Management 197 (2017) 559e570
called water transfers or markets) are increasingly applied to
maintain or enhance aquatic ecosystems in the western US,
Australia, Chile, Mexico, China, and South Africa (Grafton et al.,
2011; Debaere et al., 2014).
Instream ow dedications and management are a major thrust
of national and international stream restoration research (Richter
and Thomas, 2007). However, water quality impairments like
warm stream temperatures and low DO concentrations also limit
ecosystems and habitats (USFWS, 2003). Stream temperatures
drive biological, physical, and chemical properties of rivers,
including metabolic rates and life histories of aquatic organisms,
productivity, chemical reactions, nutrient cycling, and DO concen-
trations (Poole and Berman, 2001). DO concentrations also affect
metabolic rates of aquatic organisms, predation risk, and inuence
organism behavior and sh community structure (Magoulick and
Kobza, 2003). Agricultural diversions reduce streamow volume
which raises stream temperatures by reducing thermal mass and
slowing water travel time (Poole and Berman, 2001). DO saturation
concentrations have an inverse relationship with stream temper-
atures, so that as temperatures increase, less DO is contained in
water (Chapra, 2008). DO concentrations have been reduced from
warm stream temperatures and are exacerbated by warm,
nutrient-rich agriculture returns ows (Jalali and Kolahchi, 2009).
Nutrients promote biological activity, creating dead and decaying
biomass which consumes DO as respiration (Odum, 1956).
No research has specically examined the effects of environ-
mental water purchases on DO concentrations in streams, although
Elmore et al. (2016) showed that environmental water purchases in
the Walker Basin may reduce maximum daily stream temperatures
by up to 3
C while increasing minimum daily stream temperatures
by approximately 1
C. This conrms previous research showing
increasing streamow may decrease stream temperature by
increasing the thermal capacity of rivers (Null et al., 2010; Gu and Li,
2002). Existing research also suggests that cooler stream temper-
atures could reduce DO concentrations by increasing the amount of
oxygen that may be dissolved in water (Magoulick and Kobza,
2003; Ficklin et al., 2013), and which supports the hypothesis
that environmental water purchases may improve river DO
concentrations.
We build on modeling research completed by Elmore et al.
(2016) by simulating the effects of environmental water pur-
chases on DO concentrations and stream temperatures in the
Walker River using River Modeling System, an hourly, 1-
dimensional hydrodynamic and water quality model. Objectives
for this research are to: 1) measure the extent and seasonality of
high stream temperatures and low DO concentrations that limit
aquatic habitat in the Walker River, and 2) simulate stream tem-
perature and DO concentration change from environmental water
purchases. Improving understanding of the locations and volumes
that environmental water purchases affect DO concentrations and
stream temperatures will allow for coordinated management of
streamow and environmental water quality for better river
restoration decision-making.
2. Background
2.1. Study site
The Walker Basin is 10,500 km2 and is located in western
Nevada and eastern California (Fig. 1). Headwaters are split be-
tween two major tributaries, the East Walker and West Walker
Rivers. Both tributaries drain California's east-slope Sierra Nevada
Mountains (Sharpe et al., 2007). Mountain snowmelt is the primary
water source (Yuan et al., 2004). The Walker Basin transitions from
conifer woodland vegetation in the Sierra Nevada Mountains to
sagebrush scrub in the Great Basin Desert (Jones, 1992). The river's
terminus is Walker Lake, one of only three terminal lakes in North
America that historically supported a freshwater shery (Collopy
and Thomas, 2009).
Three reservoirs have been built to provide reliable water for
agriculture. On the East Walker River, Bridgeport Reservoir has
storage capacity of approximately 52 Mm3. West Walker River
streamow is diverted to Topaz Reservoir, a 73 Mm3 off-stream
reservoir, straddling the California-Nevada state line. Both tribu-
tary reservoirs are bottom release reservoirs and are physical bar-
riers to sh passage (Jones, 1992). The mainstem Walker River is
impounded by Weber Reservoir, with storage capacity of approxi-
mately 15 Mm3. Weber Reservoir has a bottom release with a
roughened channel shway. The dam does not have outletworks
that can be managed for downstream temperatures (Rheinheimer
et al. 2014).
Irrigated agriculture began in the basin in 1852 (Horton, 1996).
Agriculture is the dominant land use and the main consumer of
water. Principal crops include irrigated pasture, alfalfa, root crops,
and grains. Within the basin, water is over-allocated and full water
demands are met only in wet years (Yardas, 2007). For an average
snowpack year, 84% of agricultural water rights are satised. A wet
year with at least 130% of normal snowpack is required to fulll all
water rights in the basin (Sharpe et al., 2007). During dry years,
supplemental groundwater rights provide additional water.
Groundwater pumping in agricultural valleys has lowered the
water table, increasing recharge from the Walker River to the
aquifer (Carroll et al., 2010).
2.2. Hydrologic conditions
The Sierra Nevada Range and the Walker Basin have experi-
enced a prolonged and ongoing drought since 2012. Water year
(October 1 through September 30) 2014 and 2015 had the lowest
recorded snowpack, as a percentage of April 1 average, of the last 14
years of available data. Of the four sites to measure snowpack in the
Walker Basin, only two had measureable snowpack on April 1, 2015
(CDEC, 2016). The river was dewatered downstream of Wabuska
(Fig. 1) from September 9 through November 7, 2014 and again
from August 28 through December 15, 2015. This was unprece-
dented in the Walker River.
2.3. Water quality and stream ecology
Stream temperatures largely determine the overall health of
aquatic ecosystems (Poole and Berman, 2001). Effects on aquatic
organisms are typically dened in relation to chronic 7-day tem-
perature stress. In laboratory studies, 60% of adult LCT died when
stream temperatures exceeded 26
C for a week and complete
mortality occurred within two days when temperatures exceeded
28
C (Dickerson and Vinyard, 1999). Optimal temperatures for
spawning cutthroat trout (all subspecies) range from 6
C to 17
C.
Embryos and juveniles prefer temperatures of 10
C and 15
C,
respectively (Hickman and Raleigh, 1982). In trout streams, LCT
avoid water temperatures exceeding 26
C and sh are limited by
maximum rather than mean daily temperatures (Dunham et al.,
1999). Here, we assume a lethal threshold of 28
C for LCT.
Low DO concentrations in milligrams per liter (mg/L) creates
hypoxic conditions for sh species and other aquatic organisms.
While sh may tolerate short-term exposure to hypoxia, few persist
in low oxygen conditions (Magoulick and Kobza, 2003). Trout growth
rates are impaired at DO concentrations below 8 mg/L, growth rate is
reduced up to 22% when concentrations are below 6 mg/L, and
mortality occurs at 3 mg/L (WDOE, 2002; Carter, 2005). Thresholds
for this paper assume moderate impairment begins at 6 mg/L and
 S.E. Null et al. / Journal of Environmental Management 197 (2017) 559e570 561

Fig. 1. Walker Basin with monitoring locations and major landmarks. Diversion arrows are scaled relative to average diversion rates.

acute impairment occurs at concentrations below 5 mg/L. dominated. We simulated irrigation season (April 1 e October 31)
of 2014 and 2015. Upper boundary conditions are Topaz Reservoir
on the West Walker River and Bridgeport Reservoir on the East
3. Methods
Walker River. The downstream boundary is Wabuska (Fig. 1)
because the Walker River dried downstream of this point for much
3.1. Model description and development
of irrigation season in 2014 and 2015. Model extent is 305 km, with
a spatial resolution of 0.3 km. RMS is composed of a hydrodynamic
We used Tennessee Valley Authority's River Modeling System
module (ADYN) and a water quality module (RQUAL) (Hauser and
(RMS) version 4 (Hauser and Schohl, 2002) because it is open
Schohl, 2002). The two modules are run consecutively.
source, has riparian shading logic, and is process-based. RMS is a
ADYN, the hydrodynamic module, simulates time-varying ve-
one-dimensional model that simulates hourly streamow, stream
locity, depth, ow, and water surface elevation (Hauser and Schohl,
temperature, and DO. Many water quality models exist, including
2002). ADYN solves one-dimensional equations for conservation of
QUAL, WASP, BASIN, MIKE, and RMA modeling software. Models
mass and momentum (St. Venant equations) using a four point
differ in dimensionality, constituents represented, point versus
implicit nite difference scheme with weighted spatial derivatives
non-point boundary conditions, hydrodynamic capability, scale,
(Hauser and Schohl, 2002). Input for this module includes channel
and resolution. See Wang et al. (2013) for a recent review of surface
geometry, streambed roughness coefcients, boundary and lateral
water quality models. One-dimensional stream models assume that
inows, and initial water surface elevations and discharges
streamow and water quality change longitudinally, but are well
(Supplemental Table 1).
mixed laterally and vertically. Time is divided into discrete time-
Velocities and depths are passed to RQUAL, the water quality
steps, and hydrodynamics and water quality are assumed constant
module. RQUAL solves the mass transport equation using the Holly-
within each timestep. Our model assumes rivers are advection-
562 S.E. Null et al. / Journal of Environmental Management 197 (2017) 559e570
Priessman numerical method (Hauser and Schohl, 2002). RQUAL
simulates stream temperature and DO in rivers and advection-
dominated reservoirs where the one-dimensional longitudinal
ow assumption is appropriate (Hauser and Schohl, 2002). Input
includes weather (air temperature, dew point temperature, wind
speed, cloud cover, barometric pressure, and solar radiation),
boundary condition water quality, and initial water quality through
the modeled stream reach.
3.2. Walker River input data
3.2.1. Channel geometry
Walker River channel geometry is described by ve-point river
cross sections represented at 999 nodes from the outlets of
Bridgeport and Topaz Reservoirs to the mouth of Walker Lake.
Nodes were evenly spaced throughout the East, West, and main-
stem Walker Rivers. Light Detection and Ranging (LiDAR) digital
terrain data were collected in 2011 and estimate lateral elevations
at 5 m and 25 m from the river centerline (Elmore et al., 2016). For
center point depth, 20 river cross sections were measured and bed
elevations between them were interpolated, resulting in an esti-
mated average depth of 0.94 m (Elmore et al., 2016). Stream bed
roughness (Manning's n) was set uniformly to 0.05, representing a
natural stream channel with weeds and pools (Chapra, 2008).
3.2.2. Streamow
Eighteen USGS streamgages provided daily streamows for
boundary and initial conditions, and model calibration. Streamow
differences between the eighteen gages outside of irrigation season
(November 1 e March 31) estimated accretions and depletions.
Accretions and depletions account for unquantied springs, seeps,
diversions, return ows, and groundwater recharge. There were
approximately 80% more depletions during irrigation season than
the rest of the year. These were attributed to irrigation diversions
(Elmore et al., 2016). For river reaches containing multiple diver-
sion canals, each was assigned a percentage of the depletion ac-
cording to relative diversion rate (Pahl, 2000; WRIT, 2003; Yardas,
2007; Sharpe et al., 2007). Final diversion percentages are pre-
sented in Supplemental Table 2 (Elmore et al., 2016). When accre-
tions occurred in diversion reaches during the irrigation season,
diversions were assumed to be zero.
3.2.3. Dissolved oxygen and stream temperature monitoring
Ten miniDOT by Precision Measurement Engineering, Inc. DO
sensors were deployed from November 2013 through November
2015 (Fig. 1). The sensor is an optode that measures lifetime-based
luminescence quenching of uorescence of a thin membrane.
Sensors are accurate to 5% in the 0e150% oxygen saturation range
and measure stream temperatures with an accuracy of 0.1
C in
the 0
Ce30
C range (PME, 2014). Fourteen iButton temperature
sensors were also deployed (Fig. 1). iButton Model DS1921G sensors
are accurate to 1
C, at a resolution of 0.5
C in the 40
C
to 85
C range (iButton Link Technology, 2016). Data were
downloaded monthly, or bimonthly during summer, and DO sensor
membranes were cleaned to discourage bio-growth (Mouzon,
2016).
3.2.4. Meteorological data
Weather data were from the Smith Valley, NV meteorological
station (Fig. 1), managed by the Desert Research Institute (DRI,
2016). Input data include cloud cover, air temperature, dew point
temperature, air pressure, wind speed, and solar radiation. All
variables were available except for cloud cover. We estimated cloud
cover by grouping incoming solar radiation into 7-day periods,
where the highest daily total was assumed to be a cloud free day
and other days were reduced as a percentage of this maximum
(Mouzon, 2016).
3.2.5. Riparian vegetation shading
Vegetation height and type determine solar radiation trans-
mittance through riparian canopies. LiDAR data estimated average
vegetation height for each bank at all model nodes (Dilts et al.,
2010). Solar radiation transmittance was measured for major
vegetation types along the Walker River, and included 1) 100% solar
radiation transmittance for no signicant vegetation, 2) 9% solar
radiation transmittance for medium-height mixed shrub vegeta-
tion consisting mainly of willows (5.57e9.14 m), and 3) 14% solar
radiation transmittance for tall vegetation consisting primarily of
large cottonwoods (>9.14 m) (Elmore et al., 2016). Approximately
5% of the East Walker River, 1% of the West Walker River, and 1.5% of
the mainstem Walker River is shaded by mixed shrub or tall
vegetation (Dilts et al., 2010).
3.3. Calibration and model t statistics
RMS was calibrated by adjusting daily streamow volumes and
diversion percentages, heat exchange coefcients, and DO con-
centration coefcients. Twelve USGS streamow gages, 10 iButton
temperature loggers, and ve miniDOT DO sensors were used for
calibration. Streamows and diversion percentages were adjusted
to maintain enough streamow during the extremely dry study
years so that models did not crash (at least 0.06e0.14 m3/s). These
reaches were not completely dry, rather streamow was excep-
tionally low, which compromised model stability. Added water was
subtracted at the next node downstream to conserve water mass.
Water temperature coefcients were unchanged from Elmore et al.
(2016) (Supplemental Table 1).
Photosynthesis rates (gO2/m2/hr) were adjusted to match
diurnal timing of measured DO concentrations. Walker River
photosynthesis rates remained within the range of previous studies
(maximum of 1.48 gO2/m2/hr and minimum of 0.29 gO2/m2/hr)
(Westphal and Lefkowitz, 2011; Geisler, 2005). Macrophyte respi-
ration rates (gO2/m2/hr) are typically between 10% and 30% of
photosynthesis rates (Chapra, 2008; Hauser and Schohl, 2002).
Generally, we used macrophyte respiration rates at 20% of photo-
synthesis, although one reach required a macrophyte respiration
value of 23% to maintain diurnal timing. These rates are similar to
estimated macrophyte respiration for a study using RMS in Cal-
ifornia's Shasta River (Geisler, 2005). Remaining differences be-
tween modeled and measured DO were attributed to sediment
oxygen demand (SOD) (gO2/m2/day). SOD ranged from 0.57 to 5.65
gO2/m2/day and bookended published maximum SOD rates of
2.0e2.3 gO2/m2/day (Westphal and Lefkowitz, 2011; Geisler, 2005).
Environmental conditions varied between the two years. In 2014,
photosynthesis, macrophyte respiration, and SOD were calibrated
with higher values on the East and West Walker tributaries than
the mainstem Walker River based on observed periphyton,
macrophyte cover, and SOD. In 2015, only the East Walker River was
calibrated with higher rates based on observed periphyton and
macrophyte cover (Mouzon, 2016). Parameters that were adjusted
to calibrate stream temperature and DO are summarized in
Supplemental Table 1. See Mouzon (2016) for additional detail.
Root mean square error (RMSE), Nash-Sutcliffe efciency (NSE),
ratio of root mean square error to standard deviation of measured
data (RSR) and percent bias (PBIAS) between measured and
modeled data were calculated to quantify model t for streamow,
stream temperature, and DO concentration (Moriasi et al., 2007).
We compared modeled data with all measured data to corroborate
model results, rather than split measured data into calibration and
verication periods. Typically, additional parameter adjustments
 S.E. Null et al. / Journal of Environmental Management 197 (2017) 559e570
are necessary following validation, which renders the process a
calibration (Oreskes et al., 1994). RMSE, RSR, and PBIAS values near
zero indicate excellent model performance. NSE ranges between 1
and -, with 1 a perfect t. PBIAS describes the average tendency to
over- or under-predict model variables. Together these statistics
provide a robust statistical description for assessing model t
(Moriasi et al., 2007).
3.4. Model runs
Environmental water purchases were represented as reduced
diversions or increased reservoir releases for stored water rights
during irrigation season. Eight model runs were completed for 2014
and 2015, and daily average water purchases for the irrigation
season are provided in Table 1. The Historical Conditions model
run estimates current conditions of ow, stream temperature, and
DO concentrations. Some water has been purchased in the Walker
Basin, although purchases vary by water right seniority, and thus by
water year type. In dry years, approximately 2 Mm3 (1650 acre feet)
of water has been purchased, for daily average Existing Purchases
totaling 0.11 m3/s. We also assess potential future environmental
water purchases. Average daily additions of 0.71 m3/s (25 cfs) and
1.41 m3/s (50 cfs) were modeled as stored water rights from
Bridgeport and Topaz Reservoirs and the large SSWJD diversion
(Fig. 1). These are the only locations on the Walker River where
large (0.71e1.41 m3/s) environmental water purchases could be
acquired at single location. Additional model runs with smaller
environmental water purchases were described in Mouzon (2016);
however, they were ineffective for managing environmental water
quality and aquatic habitat.
Sensitivity analyses of respiration and SOD rates on DO con-
centrations were performed. Nine model runs for each year rep-
resented reductions of 10, 20, and 30% from respiration and/or SOD
oxygen consumption on DO concentrations.
4. Results and discussion
Results are organized by presenting observed aquatic habitat
conditions to address our rst research objective of measuring the
extent and seasonality of high stream temperatures and low DO
concentrations that currently limit aquatic habitat in the Walker
River. Next, we describe RMS calibration results and quantify model
t. We present modeled historical conditions as a baseline to
compare simulated environmental water purchases. Then we
describe stream temperature and DO concentration changes from
environmental water purchases to address our second research
objective. We conclude results by quantifying the sensitivity of DO
concentrations to reduced respiration and/or SOD rates, and dis-
cussing study limitations.
4.1. Observed aquatic habitat conditions
Measured data indicate that warm summer water temperatures
Table 1
Model runs with locations and average daily non-diversion volumes of environmental water purchases (m3/s) for 2014 and 2015.
Model Run Run Name Bridgeport
1 Historical Conditions
2 Existing Purchases 0.02
3 Bridgeport 0.71 m3/s 0.71
4 Bridgeport 1.41 m3/s 1.41
5 Topaz 0.71 m3/s
6 Topaz 1.41 m3/s
7 SSWJD 0.71 m3/s
8 SSWJD 1.41 m3/s
563
coincide with high air temperatures and low streamows, reiter-
ating ndings from Elmore et al. (2016). Stream temperatures
generally warm longitudinally during summer, although Weber
Reservoir moderates stream temperatures somewhat, demon-
strating temperature regulation from increased thermal capacity.
Acute (>28
C) temperature thresholds are exceeded in the Walker
River during irrigation season. Exceedances occur more often and
to a greater extent in the lower reaches of both the East and West
Walker tributaries and the mainstem Walker River (Fig. 2). High
stream temperatures persisted from early June to mid-August in
2014 and from June to mid-July in 2015.
DO concentrations fell below moderate production impairment
thresholds of 6.0 mg/L at all measured locations in the Walker River
during irrigation season in both years (Fig. 2). In 2014, DO period-
ically dropped below 3.0 mg/L in all but the uppermost reach of the
East Walker River. DO concentrations generally degraded longitu-
dinally (Fig. 3) and DO limitations existed in the 115 km upstream of
Walker Lake from approximately May through September. During
2015, DO concentrations in West Walker and upper East Walker
Rivers occasionally dropped below 5.0 mg/L. In the lower East
Walker and mainstem rivers, DO concentrations below 3.0 mg/L
limited habitat periodically during irrigation season. Suitable
habitat existed in the East and West Walker Rivers.
4.2. Model calibration results
Overall, both modeled years t measured data well. Average
annual 2014 irrigation season streamow had an RMSE of 0.12 m3/
s, RSR of 0.17, NSE of 0.96, and PBIAS of 3.5%, and 2015 irrigation
season streamow had an RMSE of 0.14 m3/s, RSR of 0.14, NSE of
0.97, and PBIAS of 3.6% (Table 2). Calibration parameters from
Elmore et al. (2016), in which one wet year and one dry year were
modeled, were applied for stream temperature without re-
calibrating so we did not bias the model toward dry years. Model
t was within about 2
C, although daily variability was slightly
muted in both years. Modeled historical stream temperatures were
consistently underestimated for 2014, remaining cooler than
measured temperatures throughout irrigation season. Modeled
stream temperatures in critically dry years are thus conservative
estimates. Average annual 2014 stream temperature had an RMSE
of 2.3
C, RSR of 0.28, NSE of 0.92, and PBIAS of 6.6%, and 2015
stream temperature had an RMSE of 1.9
C, RSR of 0.37, NSE of 0.86,
and PBIAS of 1.0% (Table 2). Modeled DO had a good to very good t
with measured data (Moriasi et al., 2007). The model captured the
timing and magnitude of diurnal swings accurately for both years.
Average annual 2014 irrigation season DO had an RMSE of 0.6 mg/L,
RSR of 0.29, NSE of 0.91, and a PBIAS of 0.5%, and 2015 had an RMSE
of 0.6 mg/L, RSR of 0.56, NSE of 0.68, and PBIAS of 0.8% (Table 2).
4.3. Modeled historical conditions
The lower East Walker River exhibited the highest maximum
daily stream temperatures (30.8
C and 31.9
C for 2014 and 2015,
Topaz COLONY GAGE SSWJD
0.01 0.05 0.03
0.71
1.41
0.71
1.41
564 S.E. Null et al. / Journal of Environmental Management 197 (2017) 559e570

Fig. 2. Measured streamows (top), stream temperatures (middle) and DO concentrations (bottom). Lighter hues are toward headwaters, and darker hues are toward Walker Laker
for 2014 (left) and 2015 (right). Red to gray gradient represents 28
C upper temperature limit (Dickerson and Vinyard 1999) and DO impairment below 6 mg/L (Carter 2005).

respectively) and 7-day average stream temperatures (22.6
C and
24.8
C for 2014 and 2015, respectively) in the Walker Basin. On
average, over the irrigation season, the FOX and BNGHH irrigation
canals divert 40% of East Walker River streamow. However, on
some days, more than 90% of streamow is diverted, allowing at-
mospheric conditions to inuence this reach substantially. Average
7-day stream temperatures were approximately 2e3.5
C cooler in
the West Walker River and neared or exceeded thermal thresholds
infrequently. The West Walker River contributed an average of
2.2 m3/s and 2.5 m3/s to the mainstem Walker River during the
irrigation season in 2014 and 2015, respectively, compared to East
Walker River's 0.4 m3/s in 2014 and 0.3 m3/s in 2015. Stream
temperatures just below the conuence of these rivers are pri-
marily driven by West Walker River.
In 2014, DO concentrations dropped below 6.0 mg/L on 108 days
and below 5.0 mg/L on 8 days in the East Walker River in 2014
(Table 3). DO concentrations in the West Walker River were below
6.0 mg/L on 64 days, below 5.5 mg/L on 25 days, and below 5.0 mg/L
on two days. Generally, DO concentrations just downstream of the
conuence were suitable for aquatic ecosystems, exceeding 6.0 mg/
L throughout 2014. DO concentrations degraded longitudinally
from the conuence. In 2015, DO concentrations were below
6.0 mg/L for nearly all of the East Walker River, with 20 days below
5.0 mg/L at the downstream reach. DO in the West Walker River
and Walker River below the conuence generally remained above
6.0 mg/L, although DO was below 6.0 mg/L for 35 days near
Wabuska. These reaches exceeded 5.5 mg/L for the 2015 irrigation
season.
4.4. Modeled environmental water purchases
Environmental water purchase alternatives evaluate increased
streamow effects on stream temperatures and DO concentrations.
Environmental water purchases decreased daily maximum stream
temperatures, increased daily minimum stream temperatures, and
had little effect on average stream temperatures. These changes
occurred from increased thermal mass provided by streamow and
conrm results by Elmore et al. (2016). Bridgeport Reservoir re-
leases of 1.41 m3/s decreased daily maximum stream temperatures
most, up to 5.0
C and 4.1
C in 2014 and 2015, respectively, and
 S.E. Null et al. / Journal of Environmental Management 197 (2017) 559e570 565

Fig. 3. Measured minimum daily DO concentrations during water year 2014. Black arrows show diversions scaled relative to average diversion rate. Spotted inow arrows are not
scaled with inow rate.

increased nightly minimum temperatures by up to 2.7
C and 2.8
C
in 2014 and 2015, respectively, in the downstream reach of East
Walker River for the week starting July 1 (Fig. 4).
Modeled environmental water purchases exhibited more
nuanced changes to DO concentrations than to stream tempera-
tures, although some clear patterns developed. First, small water
purchases were ineffective to manage DO or sometimes worsened
DO concentrations as minimum nightly stream temperatures
increased. Larger water purchases of 0.71 and 1.41 m3/s often
improved DO concentrations. Second, water purchases largely
prevented acute DO concentrations (<5.0 mg/L), but increased the
duration of moderate DO concentrations (5.5e6.0 mg/L). Finally,
upstream environmental conditions had a profound effect on
downstream reaches as the volume of water purchases increased.
Upstream environmental conditions (photosynthesis, respiration,
and SOD) that were better than downstream conditions improved
water quality downstream. But, the reverse was also true, wherein
upstream conditions could worsen downstream conditions from
excess nutrients, algal blooms, or high SOD. We discuss these
ndings in more detail in the following paragraphs.
The Existing Purchases model alternative deteriorated DO con-
centrations in the lower East Walker River just upstream of the
conuence in 2015. Adding a seasonal total of 371,000 m3 (average
daily release of 0.02 m3/s from Bridgeport Reservoir) to the East
566 S.E. Null et al. / Journal of Environmental Management 197 (2017) 559e570

Fig. 4. July 1, 2014 modeled maximum and minimum temperatures (top) and DO
concentrations (bottom). Black lines are Historical Conditions and grey lines are
Bridgeport 1.41 m3/s. Tan gradient represents acute temperature and DO thresholds
for LCT. West Walker River shown on top gure only.
Walker River increased the number of hours below both moderate
(<6.0 mg/L) and acute (<5.0 mg/L) DO thresholds. This small vol-
ume of added streamow was not enough to dilute the effects of
respiration and SOD, but slightly increased minimum nightly
stream temperatures so that less DO was contained in the water
column.
Large environmental water purchases of 1.41 m3/s as stored
water rights from Bridgeport and Topaz Reservoirs decreased the
number of days with acute DO concentrations below 5.5 mg/L by an
average of 38% and a maximum of 87% when water was purchased
from Bridgeport Reservoir in 2015. However, moderate DO con-
centrations below 6.0 mg/L (between 5.5 and 6.0 mg/L) could last
for an average of 15% longer, and up to 185% longer in the mainstem
just below the conuence when water was released from Topaz
Reservoir (Table 3). In other words, DO was below 5.5 mg/L for
fewer days, but for a longer number of hours when moderately
impaired DO conditions occurred. Also, consecutive hours below
6.0 mg/L could occur for up to two hours longer. At the downstream
reach of West Walker River, daily additions of 1.41 m3/s from Topaz
Reservoir increased both the number of days and number of hours
that DO concentrations under 6.0 mg/L and 5.5 mg/L occurred in
2014. This was caused by excessively low DO concentrations in the
West Walker River as respiration and SOD rates were higher in this
reach and time period.
The inverse relationship between stream temperature and DO
partly controls DO saturation in rivers. DO is likely driven by stream
temperatures for model alternatives that decreased the total
number of days with extremely low (acute) DO, but also increased
the duration of time with moderate DO concentrations (Fig. 5). The
timing of daily DO highs and lows indicate that stream tempera-
tures drove DO response (Butcher and Covington, 1995). However,
the pattern in Fig. 5 is not explained by temperature alone.
Photosynthesis, respiration, and SOD were held constant be-
tween model runs for a given year. With added streamow from
water purchases, the degree to which biological processes affected
DO concentrations was diluted, reducing their overall effect on DO
concentrations. For example, with 1.41 m3/s water purchases from
Bridgeport Reservoir, photosynthesis did not produce enough DO to
raise DO concentrations as high as occurred with the Historical
Conditions model run. Conversely, respiration and SOD did not
deplete DO concentrations as much. Also, during the day, photo-
synthesis added DO to streams, even though warmer stream tem-
peratures lowered DO saturation concentration, forcing oxygen out
of solution and into the atmosphere (Butcher and Covington, 1995).
In the Walker River, DO concentration maxima typically occurred
between 13:00 and 14:00. Once sunlight was no longer available,
respiration and SOD consumed DO, which occurred throughout the
night (Allan and Castillo, 2007). Nighttime lows for DO occurred
between approximately 21:00 and 22:00 in the Walker River. DO
concentrations rose as cooling stream temperatures increased DO
saturation concentration, overriding the rate of consumption from

Table 2
Average 2014 and 2015 reach streamow, temperature, and dissolved oxygen model t statistics.

Streamow Stream Temperature Dissolved Oxygen

RMSE RSR NSE PBIAS RMSE RSR NSE PBIAS RMSE RSR NSE PBIAS
(m3/s) (unitless) (unitless) (%) (
C) (unitless) (unitless) (%) (mg/ (unitless) (unitless) (%)
L)

2014
East Walker 0.06 0.23 0.95 2.5 2.4 0.31 0.90 5.2 0.7 0.33 0.89 0.6
Mainstem 0.17 0.22 0.95 6.3 2.5 0.28 0.92 9.0 0.5 0.26 0.93 1.8
Walker
West Walker 0.11 0.05 1.00 0.4 1.6 0.22 0.95 3.7 0.6 0.26 0.93 2.3
2014 Average 0.12 0.17 0.96 3.5 2.3 0.28 0.92 6.6 0.6 0.29 0.91 0.5

2015
East Walker 0.15 0.32 0.90 11.1 2.1 0.45 0.79 5.5 0.8 0.62 0.60 0.6
Mainstem 0.14 0.11 0.98 2.9 1.8 0.34 0.88 3.3 0.5 0.54 0.70 2.4
Walker
West Walker 0.16 0.05 1.00 0.3 1.5 0.30 0.91 2.8 0.4 0.47 0.78 0.0
2015 Average 0.14 0.14 0.97 3.6 1.9 0.37 0.86 1.0 0.6 0.56 0.68 0.8
 S.E. Null et al. / Journal of Environmental Management 197 (2017) 559e570 567

Table 3
DO concentration metrics for 2014 and 2015 with Historical Conditions and 1.41 m3/s releases from Bridgeport and Topaz Reservoirs. Bold values indicate DO concentration
improvement from Historical Conditions, and red underlined values indicate DO concentration deterioration from Historical Conditions. Historical Conditions are shaded
white, purchases from Bridgeport Reservoir are shaded gray, and purchases from Topaz Reservoir are shaded tan.

4.5. Sensitivity analyses

Results from sensitivity analyses suggest that low DO concen-

Fig. 5. DO concentrations and stream temperatures near Wabuska with Historical
Conditions and Bridgeport 1.41 m3/s model runs in 2014.
respiration and SOD. Greater stream volume from environmental
water purchases warmed nightly minimum temperatures and
decreased the rate at which stream temperatures cooled. For this
reason, moderate DO concentrations (<6.0 mg/L) sometimes per-
sisted for longer periods of time.
trations could be further improved by reducing respiration rate,
SOD, or a combination of both (Table 4). Reducing SOD improved
DO concentrations most. In fact, lowering SOD by just 10%
improved DO conditions so that the number of days with DO below
6.0 mg/L fell by 27%, total hours under 6.0 mg/L concentration
dropped by 29%, and the longest consecutive period under 6.0 mg/L
was shortened by one hour. Reducing both respiration and SOD
rates by 30% completely eliminated instances when DO fell below
5.0 mg/L in 2015 (Table 4). Understanding the conditions that
promote high SOD and respiration, like accumulation of decom-
posing organic material, high ammonia loads from livestock, and
excess nutrients from agriculture elds may enable managers to
enhance water quality to complement environmental water pur-
chases or provide promising restoration actions when environ-
mental water purchases are unavailable (Lee and Jones-Lee, 2007;
Butts and Evans, 1978).
4.6. Limitations
River modeling simplies systems. Some calibration parameters
were temporally and spatially uniform, so they inadequately rep-
resented all seasons and reaches of the 305 km Walker River.
568 S.E. Null et al. / Journal of Environmental Management 197 (2017) 559e570
Table 4
Sensitivity analyses of reduced respiration and SOD rates on DO concentrations for 2014 and 2015 in the lower East Walker River. Bold values indicate DO concentration
improvement from Historical Conditions, and red underlined values indicate DO concentration deterioration from Historical Conditions.
Accurate input data always improves modeling. This study did not
explicitly include groundwater interactions and contributions.
Seeps, springs, and recharge were lumped as accretions or de-
pletions. Groundwater moderates stream temperatures, resulting
in localized cooler patches of river during summer (Poole and
Berman, 2001). These inuences are likely localized or small-
scale, but may be important for thermal refuge for species. Accu-
rate measures of canal diversions would also improve model ac-
curacy. Water from bottom-release reservoirs may have low DO,
although we assumed that water would be mechanically aerated in
the relatively steeper reaches just below Topaz and Bridgeport
Reservoirs. Field measurements of DO stratication in the reser-
voirs, river SOD, and river nutrient concentrations would improve
model calibration, accuracy, and results.
Critically low ows from persistent drought resulted in little
hydrologic variability in the two years of this study and our results
assess environmental water purchase effects on DO concentrations
and instream habitat during critically dry years. Critically dry years
are increasingly likely with future climate change (Adam et al.,
2009). However, we did not explicitly model future climate alter-
natives that include snowfall shifting to rainfall (Ficklin et al., 2015),
which increases winter streamow, reduces baseow, and further
raises stream temperatures from warmer air temperatures (Null
et al., 2013). Lastly, our modeling did not consider the political,
legal, or economic feasibility of environmental water purchases.
5. Conclusions
River monitoring showed that the East and West Walker Rivers
(188.8 kilometers of habitat) and sometimes the mainstem Walker
River just below the conuence provide suitable habitat to support
native species. Aquatic habitat as a function of streamow, stream
temperature, and DO concentration is generally unsuitable down-
stream of the conuence to Walker Lake (111.7 kilometers of
habitat) in critically dry years, ndings reiterated using benthic
macroinvertebrate indices (Hogle et al., 2015; Mehler et al., 2015)
and from stream temperature modeling in a wet and a dry year
(Elmore et al., 2016). Water purchases should prioritize passage
through this reach rather than restore suitable habitat.
Understanding DO concentration and stream temperature
response to environmental water purchases is needed to address
water quality impairments which limit aquatic ecosystems (Hogle
et al., 2015). Increasing streamows from environmental water
purchases consistently reduced maximum daily stream tempera-
tures and increased minimum daily stream temperatures, with
variable changes in DO concentrations. Very small water purchases
had no benet for instream habitat because warmer nightly stream
temperatures depressed DO concentrations. Large environmental
water purchases (0.71 m3/s or larger) were most effective for
cooling maximum daily stream temperatures and preventing DO
concentrations below 5 mg/L, but increased the duration of mod-
erate DO concentrations between 5.5 and 6.0 mg/L. In biologically
productive reaches, environmental water purchases must be large
enough (approximately 0.71 m3/s or larger) to dilute the effect of
oxygen consuming processes like respiration and SOD. Upstream
reaches that had high respiration and SOD rates led to poor
downstream water quality as parcels of water moved through the
system. Overall, habitat in the West Walker River was most resilient
with existing conditions and habitat in the East Walker River had
the most improved DO concentrations from environmental water
purchases. Better understanding nutrients contributions to the
river and quantifying SOD would nuance environmental water
 S.E. Null et al. / Journal of Environmental Management 197 (2017) 559e570
purchase benets for aquatic ecosystems and highlight promising
restoration actions in addition to environmental water purchases.
Environmental water purchases are a useful river restoration
strategy in dry years. Dry year leasing is a promising avenue for
environmental water purchases and is being explored by buyers
and growers (Ise and Sunding, 1998). Walker River habitats and
ecosystems were hard hit by ongoing drought from 2012 to present.
Environmental water purchases were always benecial to maintain
cold water habitat and often benecial to maintain suitable DO
concentrations in dry years. The warm and dry conditions recently
experienced are anticipated to become more frequent in the future
from climate warming (Null and Viers, 2013; Adam et al., 2009).
Future droughts could also be longer. The current drought is more
severe, but signicantly shorter than medieval droughts (Hatchett
et al., 2015).
Water markets are increasingly implemented as a response to
water scarcity and one approach to augment streamows for
environmental protection (Debaere et al., 2014). Fractional water
sales are particularly useful so that farmers can sell or lease only a
portion of their water, incentivizing conservation and improved
water use efciency (Ise and Sunding, 1998; Sunding et al., 2002).
Our results improve understanding of water quality and habitat
effects of water purchases to enhance aquatic ecosystems while
maintaining human water uses. This research is also pertinent to
improve understanding of relationships between streamow and
water quality for other types of environmental ows like pulse
ows or reservoir releases (Richter and Thomas, 2007). River
restoration and monitoring that focuses entirely on streamow is
unlikely to improve habitat conditions, particularly where water
quality limits habitat and organisms (Null et al., 2010).
Acknowledgements
This research was funded by the National Fish and Wildlife
Foundation (grant number 2010-0059-201) as a component of the
Walker Basin Project. We thank Michelle Baker and Nancy Mesner
for feedback on this paper. We also thank Lisa Heki with the US Fish
and Wildlife Service and Peter Weisberg and Thomas Dilts at Uni-
versity of Nevada, Reno (UNR) for sharing LiDAR topographic data
of the Walker River riparian corridor. Finally, thank you to Sudeep
Chandra, eld crews from UNR's Aquatic Ecology Laboratory, and
eld crews from Utah State University's Aquatic Habitat, Climate,
and Water Analysis Laboratory for assisting with eld monitoring.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.jenvman.2017.04.016.
References
Adam, J.C., Hamlet, A.F., Lettenmaier, D.P., 2009. Implications of global climate
change for snowmelt hydrology in the twenty-rst century. Hydrol. Process. 23
(7).
Allan, J.D., Castillo, M.M., 2007. Stream Ecology: Structure and Function of Running
Waters. Springer Science & Business Media. ISBN: 1402055838.
Bae, M.J., Merciai, R., Benejam, L., Sabater, S., Garca-Berthou, E., 2016. Small weirs,
big effects: disruption of water temperature regimes with hydrological alter-
ation in a Mediterranean stream. River Res. Appl. 32 (3), 309e319.
Baron, J.S., Poff, N., Angermeier, P., et al., 2002. Meeting ecological and societal
needs for freshwater. Ecol. Appl. 12 (5).
Beutel, M.W., Horne, A.J., Roth, J.C., Barratt, N.J., 2001. Limnological effects of
anthropogenic desiccation of a large, saline lake, Walker Lake, Nevada. In: Sa-
line Lakes. Springer, pp. 91e105.
Bishop, C.D., Curtis, K.R., Kim, M., 2010. Conserving water in arid regions: exploring
the economic feasibility of alternative crops. Agric. Syst. 103 (8).
Butcher, J.B., Covington, S., 1995. Dissolved-oxygen analysis with temperature
dependence. J. Environ. Eng. 121 (10).
Butts, T.A., Evans, R.L., 1978. Sediment Oxygen Demand Studies of Selected
569
Northeastern Illinois Streams, Illinois State Water Survey.
Carroll, R.W.H., Pohl, G., McGraw, D., et al., 2010. Mason valley groundwater model:
linking surface water and groundwater in the Walker River basin, Nevada.
JAWRA 46 (3). http://dx.doi.org/10.1111/j.1752-1688.2010.00434.x.
Carter, K., 2005. The Effects of Dissolved Oxygen on Steelhead Trout, Coho Salmon,
and Chinook Salmon Biology and Function by Life Stage. California Regional
Water Quality Control Board, North Coast Region.
CDEC (California Department of Water Resources), 2016. California Data Exchange
Center. Accessed 6/2016.
Chapra, S.C., 2008. Surface Water-quality Modeling. Waveland press. ISBN:
1478608307.
Collopy, M.W., Thomas, J.M., 2009. Restoration of a Desert Lake in an Agriculturally
Dominated Watershed: the Walker Lake Basin. Bureau of Reclamation Report.
Debaere, P., et al., 2014. Water markets as a response to scarcity. Water Policy 16 (4).
Dickerson, B.R., Vinyard, G.L., 1999. Effects of high chronic temperatures and diel
temperature cycles on the survival and growth of Lahontan cutthroat trout.
Trans. Am. Fish. Soc. 128 (3).
Dilts, E., Yang, J., Weisberg, P.J., 2010. Mapping riparian vegetation with lidar data.
ArcUser Mag. 18e21.
DRI (Desert Research Institute), 2016. Smith Valley, Nevada Weather Station.
Available at. http://www.wrcc.dri.edu/weather/nsmv.html [Accessed November
1, 2016].
Dunham, J.B., et al., 1999. Local and geographic variability in the distribution of
stream-living Lahontan cutthroat trout. Trans. Am. Fish. Soc. 128 (5).
Elmore, L.R., Null, S.E., Mouzon, N.R., 2016. Effects of environmental water transfers
on stream temperatures. River Res. Appl. 32 (7) http://dx.doi.org/10.1002/
rra.2994.
Ficklin, D.L., et al., 2015. Assessing differences in snowmelt-dependent hydrologic
projections using CMIP3 and CMIP5 climate forcing data for the western United
States. Hydrol. Res. 47 (2) http://dx.doi.org/10.2166/nh.2015.101.
Ficklin, D.L., Stewart, I.T., Maurer, E.P., 2013. Effects of climate change on stream
temperature, dissolved oxygen, and sediment concentration in the Sierra
Nevada in California. Water Resour. Res. 49 (5).
Geisler, L., 2005. Shasta River Flow, Temperature, and Dissolved Oxygen Model
Calibration Technical Report. Prepared for ICE, UC Davis and NCRWQCB. UC
Davis and Watercourse Engineering, Inc. Available online. http://www.
waterboards.ca.gov/northcoast/programs/tmdl/shasta/060707/20AppendixD
ShastaRiverFlowTemperatureandDissolvedOxygenModelCalibrationTechnicalRe.
Grafton, R.Q., et al., 2011. An integrated assessment of water markets: a cross-
country comparison. Rev. Environ. Econ. Policy 5 (2).
Grantham, T.E., Figueroa, R., Prat, N., 2013. Water management in Mediterranean
river basins: a comparison of management frameworks, physical impacts, and
ecological responses. Hydrobiologia 719, 451.
Goniea, T.M., Keefer, M.L., Bjornn, T.C., Peery, C.A., Bennett, D.H., Stuehrenberg, L.C.,
2006. Behavioral thermoregulation and slowed migration by adult fall Chinook
salmon in response to high Columbia River water temperatures. Trans. Am. Fish.
Soc. 135 (2), 408e419.
Gu, R.R., Li, Y., 2002. River temperature sensitivity to hydraulic and meteorological
parameters. J. Environ. Manag. 66 (1).
Hatchett, B.J., et al., 2015. Placing the 2012e2015 California-Nevada drought into a
paleoclimatic context: insights from Walker Lake, California-Nevada, USA.
Geophys. Res. Lett. 42 (20).
Hauser, G.E., Schohl, G.A., 2002. River Modeling System V4duser Guide and
Technical Reference. Report No. WR28-1-164.
Hickman, T., Raleigh, R.F., 1982. Habitat Suitability Index Models: Cutthroat Trout.
USDI Fish and Wildlife Service, FWS/OBS-82/10.5.
Hogle, C., Sada, D., Rosamond, C., 2015. Using benthic indicator species and com-
munity gradients to optimize restoration in the Arid, Endorheic Walker River
Watershed, Western USA. River Res. Appl. 31 (6).
Horton, G.A., 1996. Walker River Chronology: a Chronological History of the Walker
River and Related Water Issues. Nevada Division of Water Planning, Department
of Conservation and Natural Resources.
iButton Link Technology, 2016. DS 1921G. Available at. http://www.ibuttonlink.com/
products/ds1921g [Accessed November 1, 2016].
Ise, S., Sunding, D.L., 1998. Reallocating water from agriculture to the environment
under a voluntary purchase program. Rev. Agric. Econ. 20 (1).
Jalali, M., Kolahchi, Z., 2009. Effect of irrigation water quality on the leaching and
desorption of phosphorous from soil. Soil Sediment Contam. 18 (5).
Jones, J., 1992. Walker River Atlas. Department of Water Resources, State of Cali-
fornia, Sacramento, CA.
Lee, G.F., Jones-Lee, A., 2007. Role of Aquatic Plant Nutrients in Causing Sediment
Oxygen Demand Part IIesediment Oxygen Demand, Report of G. Fred Lee &
Associates, El Macero, CA.
Magoulick, D.D., Kobza, R.M., 2003. The role of refugia for shes during drought: a
review and synthesis. Freshw. Biol. 48 (7).
Mehler, K., et al., 2015. Factors affecting spatiotemporal benthic macroinvertebrate
diversity and secondary production in a semi-arid watershed. J. Freshw. Ecol. 30
(2).
Moriasi, D.N., et al., 2007. Model evaluation guidelines for systematic quantication
of accuracy in watershed simulations. Trans. ASABE 50 (3).
Mouzon, N.R., 2016. Effects of Environmental Water Rights Purchases on Dissolved
Oxygen, Stream Temperature, and Fish Habitat. MS Thesis in Aquatic Ecology.
Utah State University.
Moyle, P.B., Katz, J.V.E., Quinones, R.M., 2011. Rapid decline of California's native
inland shes: a status assessment. Biol. Conserv. 144, 2414e2423.
570 S.E. Null et al. / Journal of Environmental Management 197 (2017) 559e570
Nehlsen, W., Williams, J.E., Lichatowich, J.A., 1991. Pacic salmon at the crossroads:
stocks at risk from California, Oregon, Idaho, and Washington. Fisheries 16 (2),
4e21.
NFWF (National Fish and Wildlife Foundation), 2016. Walker Basin Restoration
Program Overview - Legislative Commission's Subcommittee to Study Water. In
Walker Basin Restoration Program. Dyer, NV. Available at. https://www.leg.
state.nv.us/interim/78th2015/committee/statcom/lcwater/other/7-june-2016/
agendaivwalkerbasinrestorationprogram.pdf?rewrote1.
Null, S.E., et al., 2013. Stream temperature sensitivity to climate warming in Cal-
ifornia's Sierra Nevada: impacts to coldwater habitat. Clim. change 116 (1).
Null, S.E., Deas, M.L., Lund, J.R., 2010. Flow and water temperature simulation for
habitat restoration in the Shasta River, California. River Res. Appl. 26 (6).
Null, S.E., Viers, J.H., 2013. In bad waters: water year classication in nonstationary
climates. Water Resour. Res. 49 (2).
Odum, H.T., 1956. Primary production in owing waters. Limnol. Oceanogr. 1 (2).
Oreskes, N., Shrader-Frechette, K., Belitz, K., 1994. Verication, validation, and
conrmation of numerical models in the earth sciences. Science 263 (5147).
Pahl, P., 2000. Walker River Basin Irrigation Diversions, Prepared for Nevada Water
Basin Information and Chronology Series. State of Nevada, Department of Water
Resources, Carson City, NV.
PME (Precision Measurement Engineering, Inc, 2014. miniDOT. Available at. http://
pme.com/wp-content/uploads/2014/07/miniDOT2.pdf [Accessed November 1,
2016].
Poole, G.C., Berman, C.H., 2001. An ecological perspective on in-stream tempera-
ture: natural heat dynamics and mechanisms of human-caused thermal
degradation. Environ. Manag. 27 (6).
Rheinheimer, D.E., Null, S.E., Lund, J.R., 2014. Optimizing selective withdrawal from
reservoirs to manage downstream temperatures with climate warming. Journal
of Water Resources Planning and Management 141 (4).
Richter, B.D., Thomas, G.A., 2007. Restoring environmental ows by modifying dam
operations. Ecol. Soc. 12 (1).
Sharpe, S.E., Cablk, M.E., Thomas, J.M., 2007. The Walker Basin, Nevada and Cali-
fornia: Physical Environment, Hydrology, and Biology. Desert Research Institute.
Sunding, D., et al., 2002. Measuring the costs of reallocating water from agriculture:
a multi-model approach. Nat. Resour. Model. 15 (2).
USFWS (U.S. Fish and Wildlife Service), 1975. Reclassication of Lahontan cutthroat
trout to threatened. Fed. Regist. 40, 29864.
USFWS (U.S. Fish and Wildlife Service), 2003. Short-term Action Plan for Lahontan
Cutthroat Trout (Oncorhynchus clarki Henshawi) in the Walker River Basin,
Developed by Walker River Basin Recovery Implementation Team. Reno, NV.
Wang, Q., Li, S., Jia, P., Qu, C., Ding, F., 2013. A review of surface water quality models.
Sci. World J. 2013 http://dx.doi.org/10.1155/2013/231768.
WDOE (Washington State Department of Ecology), 2002. Evaluating Criteria for the
Protection of Freshwater Aquatic Life in Washington's Surface Water Quality
Standards: Dissolved Oxygen, Draft Discussion Paper and Literature Summary.
Publication number 00-10-071.
Westphal, K., Lefkowitz, J., 2011. Harpeth River Water Quality Model Comparison.
Available at. http://www.franklin-gov.com/home/showdocument?id9844
[Accessed November 1, 2016].
WRIT (Walker River Basin Implementation Team), 2003. Short-term Action Plan for
Lahontan Cutthroat Trout (Oncorhynchus Clarki Henshawi) in the Walker River
Basin. U.S. Fish and Wildlife Service, Reno, NV.
Yardas, D., 2007. Issues and Opportunities for Acquiring Water from Willing Sellers
to Increase Walker Lake Inows, Prepared for Natural Resources Conservation
Service. U.S. Department of Agriculture, Reno, NV.
Yuan, F., et al., 2004. A 1200 year record of hydrologic variability in the
Sierra Nevada from sediments in Walker Lake, Nevada. Geochem. Geophys.
Geosyst. 5 (3).