Download A method to consider whether dams mitigate climate change effects

JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION
Vol. 49, No. 6
December 2013
AMERICAN WATER RESOURCES ASSOCIATION
A METHOD TO CONSIDER WHETHER DAMS MITIGATE
CLIMATE CHANGE EFFECTS ON STREAM TEMPERATURES1
Sarah E. Null, Scott T. Ligare, and Joshua H. Viers2
ABSTRACT: This article provides a method for examining mesoscale water quality objectives downstream of
dams with anticipated climate change using a multimodel approach. Coldwater habitat for species such as trout
and salmon has been reduced by water regulation, dam building, and land use change that alter stream temperatures. Climate change is an additional threat. Changing hydroclimatic conditions will likely impact water temperatures below dams and affect downstream ecology. We model reservoir thermal dynamics and release
operations (assuming that operations remain unchanged through time) of hypothetical reservoirs of different
sizes, elevations, and latitudes with climate-forced inflow hydrologies to examine the potential to manage water
temperatures for coldwater habitat. All models are one dimensional and operate on a weekly timestep. Results
are presented as water temperature change from the historical time period and indicate that reservoirs release
water that is cooler than upstream conditions, although the absolute temperatures of reaches below dams warm
with climate change. Stream temperatures are sensitive to changes in reservoir volume, elevation, and latitude.
Our approach is presented as a proof of concept study to evaluate reservoir regulation effects on stream
temperatures and coldwater habitat with climate change.
(KEY TERMS: water temperature; water management; climate change; regulated rivers; reservoirs; coldwater
habitat; Sierra Nevada.)
Null, Sarah E., Scott T. Ligare, and Joshua H. Viers, 2013. A Method to Consider Whether Dams Mitigate
Climate Change Effects on Stream Temperatures. Journal of the American Water Resources Association
(JAWRA) 49(6): 1456-1472. DOI: 10.1111/jawr.12102
atmosphere and streambed, as well as source temperatures (i.e., groundwater, snowmelt, overland flow,
and precipitation). Long-term data analysis has
shown that stream temperatures have already
increased as air temperatures have risen in recent
decades (Hari et al., 2006; Kaushal et al., 2011), and
modeling studies suggest that stream temperatures
will increase further with climate warming, potentially causing habitat reduction for coldwater species
(Jager et al., 1999; Battin et al., 2007; Null et al.,
2013).
INTRODUCTION AND RATIONALE
Stream temperature is an instrumental characteristic of lotic systems, directly influencing dissolved
oxygen levels, nutrient cycling, chemical reaction
rates, productivity, and mortality in aquatic ecosystems (Poole and Berman, 2001). Stream temperature
is influenced by climate, making it particularly vulnerable to climate change (Mohseni and Stefan,
1999). It is driven by heat exchange with the
1
Paper No. JAWRA-12-0146-P of the Journal of the American Water Resources Association (JAWRA). Received June 14, 2012; accepted
April 29, 2013. © 2013 American Water Resources Association. Discussions are open until six months from print publication.
2
Respectively, Assistant Professor (Null), Department of Watershed Sciences, Utah State University, 5210 Old Main Hill, NR 201, Logan,
Utah 84321; Water Resources Control Engineer (Ligare), California State Water Resources Control Board, Sacramento, California 95814;
and Associate Director (Viers), Center for Watershed Sciences, University of California, Davis, Davis, California 95616 (E-Mail/Null: sarah.
[email protected]).
JAWRA
1456
JOURNAL
OF THE
AMERICAN WATER RESOURCES ASSOCIATION
A METHOD
TO
CONSIDER WHETHER DAMS MITIGATE CLIMATE CHANGE EFFECTS
OF THE
AMERICAN WATER RESOURCES ASSOCIATION
STREAM TEMPERATURES
mitigating climate change effects on stream temperatures at a regional scale. Reservoirs of varying size,
latitude, and elevation in California’s west-slope
Sierra Nevada mountain region are used for this
proof of concept study. This research begins to quantify the extent to which dams may provide thermal
benefit under “normal operations” — a generic operating rule curve that does not change operations to
explicitly manage for coldwater. This research is
novel because it explores thermal management at the
mesoscale, rather than analyze temperatures on a
case-by-case basis for individual reservoirs. It is
intended to guide operations and help researchers
and water managers who are relied on to provide
water for multiple and competing uses.
This study begins with a description of the Sierra
Nevada study area, including a discussion of current
water development. We explain our modeling
approach, including hydroclimate input data,
the Water Evaluation and Planning (WEAP) rainfallrunoff model used to simulate water operations; the
Water Quality for River-Reservoir Systems (WQRRS)
water quality model used to estimate reservoir temperatures; and the Regional equilibrium stream TEMPerature (RTEMP) model. We discuss reservoir release
temperatures with climate change, compare unregulated and regulated stream temperatures, analyze the
effect of reservoirs of different sizes, elevations, and
latitudes on stream temperature longitudinal profiles,
and end with a discussion of data uncertainty. Overall,
the approach described here provides a method for
dam operators or other stakeholders to incorporate
water quality effects of climate change with water supply, hydropower, and flood control impacts for a more
complete understanding of climate change impacts of
water regulation. Aquatic and riparian habitats are
often directly or indirectly managed by water operations and reservoir releases which alter downstream
water quantity, quality, and timing. As water is managed more tightly in a future with climate change, this
type of research will help to evaluate how reservoir
releases may benefit ecosystems.
A growing literature links climate change specifically with water quality and quantity for salmonid species (salmon and trout), but does not always consider
regulated systems. Research in unregulated systems
indicates that stream warming could reduce the total
amount of coldwater habitat (Eaton and Scheller, 1996;
Bogan et al., 2003; Tung et al., 2006; Battin et al.,
2007; Isaak et al., 2010; Null et al., 2013) and shift it to
higher elevations or latitudes (Jager et al., 1999; Hari
et al., 2006; Isaak et al., 2010; Null et al., 2013), alter
the distribution and abundance of coldwater species
(Gooseff et al., 2005), or introduce pathways for invasive species (Rahel and Olden, 2008). Most climate
change research in regulated systems focuses on
impacts to people, such as water supply, hydropower
generation, and flood control (Hamlet and Lettenmaier,
1999; Barnett et al., 2005; Medellın-Azuara et al.,
2008; Madani and Lund, 2010; Das et al., 2011).
It is not well understood the extent to which dams
can provide coldwater releases to maintain coldwater
habitat in coming decades or how climate change
may impact release temperatures. Reservoir releases
may present an opportunity to manage water temperature and provide suitable coldwater habitat for fish
and wildlife, particularly by releasing water from the
coldwater pool of thermally stratified reservoirs. Reservoir releases provide new boundary conditions for
downstream reaches, with diminishing influence as
distance from the dam increases. It is possible that
dams could mitigate some of the stream temperature
impacts of climate change, particularly where reservoir releases are colder than unregulated conditions
or where winter water is stored in the coldwater pool
of deep reservoirs. A handful of articles focus on
potential climate warming impacts to stream temperatures in regulated systems (Sinokrot et al., 1995;
Johnson et al., 2004; Gooseff et al., 2005; Yates et al.,
2008). Sinokrot et al. (1995) found that climate
change effects regulated stream temperatures more if
reservoirs release from the surface rather than the
deeper coldwater pool. A study of the Blue Mesa Reservoir on the Colorado River showed that climate
plays the largest role in reservoir thermal structure
and that a warmer climate may reduce the volume of
the hypolimnion (Johnson et al., 2004). Yates et al.
(2008) modeled California’s Shasta Reservoir and
downstream Sacramento River, and determined the
dam could provide coldwater to offset the effects of
climate change and maintain downstream habitat for
native Chinook salmon. Gooseff et al. (2005) found
that stream temperatures downstream of a shallow
reservoir in Montana’s Lower Madison River are
expected to increase from climate change and that
they are driven by daytime warming.
The objective of this study was to develop a
method to evaluate the potential of reservoirs for
JOURNAL
ON
BACKGROUND
Study Area
California’s west-slope Sierra Nevada rivers flow
westward from the crest of the Sierra Nevada through
the confluence with the Sacramento-San Joaquin Delta
and to the Pacific Ocean (Figure 1). The Sierra Nevada
mountain region has a Mediterranean-montane climate with a dry season roughly from May to October
1457
JAWRA
NULL, LIGARE,
AND
VIERS
FIGURE 2. Average Annual Number of Weeks Unregulated
Stream Temperatures Exceed 24°C (lethal threshold for Chinook
salmon and steelhead trout (Eaton and Scheller, 1996)) with Incremental Uniform 2°C Air Temperature Increases (T0, T2, T4, and
T6 represent climate warming of 0, 2, 4, and 6°C, respectively)
(Null et al., 2013).
FIGURE 1. Study Watersheds and Topography (Null et al., 2010a).
ture regime, threatening the health and integrity of
freshwater ecosystems (Poff, 2009; Olden and Naiman,
2010). West-slope watersheds from the Feather
watershed in the north to the Kern watershed in the
south have about 24,590 million cubic meters (mcm) of
water storage for all dams greater than 1.2 mcm
(1 thousand acre feet [taf]), and approximately 8,751
MW of online hydropower capacity (Null et al., 2010a).
Infrastructure consists of a complex network of dams,
diversions, canals, powerhouses, and transmission lines.
In general, the distribution of reservoirs is approximately equal by elevation, although storage capacity
is much greater at lower elevations (Table 1).
Roughly 75% of California’s in-state hydropower is
produced from more than 150 dams in the Sierra
Nevada Mountains at higher elevations (Aspen Environmental and M.Cubed, 2005), which are mostly
privately owned and operated solely for hydropower
revenues (Vicuna et al., 2008). Winter rain and
and a wet season from November to April. Historically,
snowline was about 1,000 meters (m). Precipitation
averaged about 100 cm/yr for the region, but varied
greatly with latitude, elevation, and local weather patterns. Precipitation was greatest in the northern
watersheds and the highest elevations of the southern
Sierra Nevada. Typically, snowpack persisted late into
summer.
Modeling suggests that unregulated stream temperatures are anticipated to increase approximately 1.6°C
for each 2°C rise in air temperature in this region (simulated with uniform air temperature increases of 2, 4,
and 6°C as a sensitivity analysis of stream temperature
response to climate warming) (Null et al., 2013) (Figure 2). However, Sierra Nevada rivers have been extensively developed for water supply, hydropower
generation, flood protection, and recreation, which has
fundamentally altered the natural flow and tempera-
TABLE 1. Major Sierra Nevada Reservoirs by Elevation, Latitude, and Average Volume (mcm).
Elevation Group
Northern (Feather,
Yuba, Bear, American,
Cosumnes)
Very high (>1,600 m)
High (800-1,600 m)
Mid (<800 m)
Total
n = 12
avg. vol.
n = 14
avg. vol.
n=9
avg. vol.
n = 35
avg. vol.
= 50
= 206
= 800
= 305
Central (Mokelumne,
Calaveras, Stanislaus,
Tuolumne, Merced)
n=4
avg. vol.
n=7
avg. vol.
n=6
avg. vol.
n = 17
avg. vol.
Southern (San
Joaquin, Kings,
Kaweah, Tule, Kern)
n=6
avg. vol.
n=2
avg. vol.
n=7
avg. vol.
n = 15
avg. vol.
= 82
= 171
= 1,313
= 553
= 137
= 102
= 412
= 261
Total
n = 22
avg. vol.
n = 23
avg. vol.
n = 22
avg. vol.
n = 67
avg. vol.
= 80
= 186
= 813
= 357
Note: mcm, million cubic meters.
JAWRA
1458
JOURNAL
OF THE
AMERICAN WATER RESOURCES ASSOCIATION
A METHOD
TO
CONSIDER WHETHER DAMS MITIGATE CLIMATE CHANGE EFFECTS
ON
STREAM TEMPERATURES
spring snowmelt are stored in relatively small reservoirs that provide within-year storage for generation
primarily during summer. The remaining 25% of California’s in-state hydropower is produced in large, low
elevation, multiuse projects operated for water supply, hydropower, and flood protection, which are
owned and operated by the state and federal governments (Aspen Environmental and M.Cubed, 2005).
METHODS
Our approach was to model a simple, hypothetical
watershed that has many of the physical characteristics found in watersheds of the Sierra Nevada such
as hydropower, water supply, and flood control infrastructure, as well as unregulated tributaries. This
method required a series of models that, when linked
via parameterization and sequencing, can be applied
to a large spatial area such as California’s Sierra
Nevada and have the ability to represent multiple
years and climate scenarios quickly.
FIGURE 3. Hypothetical Study Watershed with Modeled
Infrastructure.
4,363 mcm (Oroville on the Feather River). We modeled a reservoir approximately the size of Folsom
Lake to assess climate change effects on rim dams
(Table 2). Flood control operations were developed by
simplifying the U.S. Army Corps of Engineers Folsom
flood control curve. The top of conservation pool was
drawn down to 500 mcm from week 41 to week 46
and held constant for the entire flood season regardless of storage space available upstream. From week
13 to week 21 the flood control space was reduced to
zero for the remainder of the year. We represented
water supply operations in the lower reservoir with
water demand that peaks in the summer and is at a
minimum during winter months, based on actual water
demand (Department of Water Resources, 2009).
We modeled hydropower operations by requiring a
minimum release from the reservoir for each week,
which was based on average weekly hydropower
releases from hydropower plants in the Sierra
Nevada (Madani and Lund, 2010). This method
assumes that hydropower demand does not change
with water year type or climate.
To assess the effects of climate change on stream
temperatures over a spectrum of latitudes, the entire
theoretical watershed was shifted from the central
Infrastructure Representation
Our hypothetical watershed had three main forks,
with a total drainage area of 2,341 km2 (Figure 3).
Elevation ranges 1,200 m with three reservoirs in
series. The middle fork is the largest of the three
forks and has two hydropower projects, which consist
of a reservoir and a diversion to a downstream hydropower plant. Reservoirs are 58.1 mcm (approximately
the size of Ice House Reservoir in the American
watershed) and 86.9 mcm (about the size of Bowman
Lake in the Yuba watershed), respectively (Table 2).
The north fork and the south fork are undammed
and have a natural flow regime.
The lowest reservoir is below the confluence of all
three forks near the bottom of the watershed
representing a multipurpose “rim” dam, managed for
water supply, hydropower, and flood control. The size
of rim reservoirs in the Sierra Nevada varies from
86.3 mcm (Englebright on the Yuba River) to over
TABLE 2. Physical Characteristics of Modeled Reservoirs.
Reservoir
Elevation
Upper
Middle
Lower
Reservoir
Size
Elevation
(m)
Storage
Capacity (mcm)
Hydropower
Capacity (MW)
Turbine
Capacity (m3/s)
S
M
L
1,615
1,291
300
58.1
86.9
1,233.0
12
65
200
9.0
25.5
86.8
Note: mcm, million cubic meters.
JOURNAL
OF THE
AMERICAN WATER RESOURCES ASSOCIATION
1459
JAWRA
NULL, LIGARE,
AND
WEAP, WQRRS, and RTEMP were run sequentially, and are described below (see modeling
sequence in right side of Figure 4). A weekly timestep
was used for all models. This study required 48 model
runs, representing two GCMs, two management scenarios for regulated and unregulated conditions,
three latitudes, and four 20-year time periods ranging
from 1970 to 2090, which were labeled as historical
(htp), near-term (ntp), mid-term (mtp), and far-term
(ftp) time periods (Table 4).
Sierra Nevada (in the Stanislaus River) about
200 km north to the approximate location of the
Feather watershed and about 100 km south to the
approximate location of the San Joaquin watershed
(Figure 4). Watershed size, shape, and infrastructure
remained constant and only climatic inputs and
hydrology were changed with latitude.
Input Data and Modeling Approach
Reservoir Operations and Outflows — Water
Evaluation and Planning. WEAP is a water management and hydrology software package developed
by the Stockholm Environment Institute. Here,
hydrology was modeled with the VIC model by the
California Nevada Applications Program/California
Climate Change Center and WEAP was used to
model water management including infrastructure,
water demands, and allocation priorities (Yates et al.,
2005). It operates on a weekly timestep with spatial
representation of 250-m elevation bands (Young
et al., 2009). We placed the highest allocation priority
on water supply, second on hydropower generation,
and third on reservoir storage. WEAP output data
were passed to RTEMP and WQRRS (Figure 4). See
Yates et al. (2005) for WEAP governing equations
and additional model detail, Young et al. (2009) for a
description of unregulated Sierra Nevada flow modeling, Null et al. (2010a) for hydrologic sensitivity to
climate warming in California’s Sierra Nevada, and
Mehta et al. (2011) for climate change impacts on
Sierra Nevada hydropower production.
We used two general circulation models (GCMs)
that represent extremes for possible precipitation and
temperature, MIROC32med and NCARPCM1, MIROC32med is warmer and drier whereas NCARPCM1
is cooler and wetter (Cayan et al., 2009). Both GCMs
used the A2 emissions scenario, which assumes moderately severe climate change and continuously increasing population growth (Weare, 2009). GCM output was
downscaled to 1/8° using the bias correction and spatial downscaling method described in Maurer et al.
(2010). Climate input data drove the Variable Infiltration Capacity (VIC) rainfall-runoff model to produce
surface runoff, base flow, evapotranspiration, and soil
moisture on a 1/8° spatially gridded format (Maurer,
2007). These data were provided by the California
Nevada Applications Program/California Climate
Change Center (Table 3).
Reservoir Water Quality Modeling — Water Quality
for Reservoir-River Systems. Reservoir temperatures
were simulated with WQRRS, a one-dimensional
river and reservoir model developed by the U.S. Army
Corps of Engineers (USACE-HEC, 1986). WQRRS is
a physically-based, finite difference model based on
the principles of conservation of heat and mass. The
model operates on a daily timestep, although we used
aggregated weekly input and output data (we input
average weekly inflow, stream temperature, and
weather data and used average weekly water temperature output). We chose WQRRS because of quick
run times (20 years were simulated in seconds). For
this study, we used only the reservoir simulation
component (not the river simulation component so
that unregulated stream temperatures previously
modeled with RTEMP were comparable). Water temperature was the only water quality parameter modeled and was estimated using the heat budget
method (USACE-HEC, 1986).
WQRRS has been used extensively through the
past three decades to model water quality in lakes
FIGURE 4. West-Slope Watersheds of the Sierra Nevada
Showing Locations of Hypothetical Watersheds and Reservoirs
(left side) and Model Run Flow chart (right side).
JAWRA
VIERS
1460
JOURNAL
OF THE
AMERICAN WATER RESOURCES ASSOCIATION
A METHOD
TO
CONSIDER WHETHER DAMS MITIGATE CLIMATE CHANGE EFFECTS
ON
STREAM TEMPERATURES
TABLE 3. Model Input Data and Source. RTEMP Input from Null et al. (2013), WEAP Input from Young et al. (2009).
Model
Input
VIC
Source
GCM (MIROC32med and NCARPCM1–A2
SRES) boundary conditions
Meteorology
Hydrology (surface runoff, snow accumulation,
snowmelt, groundwater)
Topography
Reservoir outflow
Reservoir release water temperature
Inflow hydrology
Reservoir operational rules
Meteorology
RTEMP
WEAP
WQRRS
Geometry
Flow (inflow, outflow, initial storage)
Inflow water temperature
California Climate Change Center
VIC (cloud cover user defined)
VIC
USGS 10 m digital elevation model
WEAP
WQRRS
VIC
User defined
VIC (wet bulb temperature calculated, cloud
cover user defined)
User defined
WEAP
RTEMP
Note: RTEMP, Regional equilibrium stream TEMPerature; WEAP, Water Evaluation and Planning; VIC, Variable Infiltration Capacity;
WQRRS, Water Quality for River-Reservoir Systems.
TABLE 4. Model Run Matrix.
Variable
Climate models
Time periods
Latitudes
Management1
Model Runs
Count
A2-MIROC32med
A2-NCARPCM1
htp (1970-1990)
ntp (2010-2030)
mtp (2040-2060)
ftp (2070-2090)
Northern
Central
Southern
Regulated
Unregulated
2
Total model runs
V
where C is thermal energy (kcal), V is volume (m3),
t is time (s), x is vertical distance in the reservoir
(m), Qx is advective flow (m3/s), Ax is surface area
(m2), Dc is the effective diffusion coefficient (m2/s), Qi
is lateral inflow (m3/s), Ci is inflow thermal energy
(kcal), Qo is lateral outflow (m3/s), and S are sources
and sinks (kcal/s). Internal transport of heat and
mass occurs only in the vertical direction through
advection and effective diffusion mechanisms, with
water well mixed laterally and longitudinally. We
assumed all reservoir inflow occurs at the upstream
end of the reservoir, making the rate of advection
slower than if the inflow occurred closer to the outlet.
Effective diffusion between layers is based on temperature and includes molecular and turbulent diffusion
as well as convective mixing based on density gradients. One-dimensional models such as WQRRS work
well for small to large reservoirs, but should not be
used for shallow reservoirs or reservoirs with small
residence times (i.e., vertically mixed systems
that would be better represented with a river model)
(USACE-HEC, 1986).
Reservoirs are represented as a series of onedimensional layers, with user-defined surface area,
thickness, and volume. For this application, layer
thickness was 1 m. Inflows are instantaneously dispersed throughout the reservoir layer of similar density. Outflows are modeled using the selective
withdrawal allocation method developed by the U.S.
Army Engineer Waterways Experiment Station (WES
method) (USACE-HEC, 1986) to describe the vertical
limits of the withdrawal zone and the vertical velocity
distribution within the zone. For additional model
4
3
2
48
Notes: htp, historical period; ntp, near-term period; mtp, mid-term
period; ftp, far-term period.
1
Reservoir operation identical for all regulated runs (generic operating rules).
and
reservoirs,
with
recent
publications
reviewed here. It was used to aid decision making by
evaluating management alternatives for sustainable
ecosystem management of Turkish coastal lagoons
(Gonenc et al., 1997), and to aid conflict resolution
over water allocation by modeling Iran’s Karkheh
river-reservoir system (Karamouz et al., 2008). Deas
and Orlob (1999) simulated water quantity and quality for restoration of anadromous fisheries below Iron
Gate Dam on the Klamath River. Lopes et al. (2003)
used it to simulate dissolved oxygen and water temperature in Portugal’s Aguieira Reservoir, and Meyer
et al. (1996) altered meteorologic and hydrologic
input conditions to estimate water quality effects on
Shasta Reservoir and part of the Sacramento River
from a doubling of CO2.
Water temperature and other water quality parameters are simulated using the one-dimensional form
of the advection-dispersion equation:
JOURNAL
OF THE
AMERICAN WATER RESOURCES ASSOCIATION
@C
@C
@2C
þ DxQx
¼ DxAx Dc 2 þ Qi Ci Qo C VS ð1Þ
@t
@x
@x
1461
JAWRA
NULL, LIGARE,
AND
detail, see USACE-HEC (1986). Here, we aimed to
represent reservoir outlet structures in a generic
way, and modeled withdrawal outlets as a small
(<1 m3/s) withdrawal structure near the bed of the
reservoir and a second outlet near the top of the coldwater pool for the remainder of outflow (modeled with
WEAP). Spill events occur from the reservoir surface.
In reality, there is quite a bit of variability in outlet
structure design. The quantity and quality of reservoir releases form the boundary conditions for downstream river temperature models. See Table 3 for a
list of input data and sources used for each reservoir
model. Results from WEAP and WQRRS are passed
to RTEMP to estimate stream temperatures, and the
models are run sequentially (Figure 4).
from WEAP and WQRRS, we assumed that snowmelt
was 1°C to account for slight heating during overland
flow from snowpatch to stream, precipitation temperature was the same as air temperature, and groundwater temperature was 7.5°C, which is the mean
annual air temperature for the west-slope Sierra
Nevada at 2,500 m.
Equilibrium temperature is useful because heat
flux at the water surface can be represented with
only meteorological input variables, and it can be
applied on a regional scale. RTEMP has previously
been used to assess unregulated stream temperature
sensitivity to climate warming in California’s westslope Sierra Nevada, with an emphasis on impacts to
coldwater habitat for anadromous salmon species
(Null et al., 2013). Other equilibrium temperature
studies exist. Notable examples include assessing the
effects of climate change by examining the relationship between stream and air temperatures (Mohseni
and Stefan, 1999), predicting stream temperatures in
Canada’s Miramichi River (Caissie et al., 2005), modeling Colorado River temperatures downstream of
Glen Canyon Dam (Wright et al., 2009), and quantifying the effects of stream discharge on summer river
temperature (Gu et al., 1998).
Stream Temperature Model — Regional Equilibrium Temperature Model. Stream temperatures
were estimated with RTEMP, a one-dimensional equilibrium temperature model developed by Watercourse
Engineering, Inc. (Null et al., 2013). We used a weekly
timestep and spatial resolution of 250-m vertical elevation bands for streams (the same resolution as WEAP).
RTEMP is a physically-based model, which uses meteorology, hydrology, boundary conditions, and stream
length (Table 3) to estimate stream temperatures
using equilibrium temperature theory (Edinger et al.,
1968).
Stream temperatures are estimated using a simplified form of the advection-dispersion equation:
@Tw
Hnet
¼
@t
Cp qd
Model Calibration and Testing
WEAP and RTEMP models were previously calibrated for the 15 west-slope Sierra Nevada watersheds with natural flow hydrologic conditions (Young
et al., 2009; Null et al., 2013). Reservoir water temperatures, modeled with WQRRS, were calibrated
using measured temperature profiles from Icehouse
Reservoir, Bowman Reservoir, and Folsom Reservoir
for the upper, middle, and lower reservoirs, respectively. Calibration dates differ between modeled reservoirs depending on when measured data were
available. Most stream and reservoir temperature
studies in the Sierra Nevada are part of Federal
Energy Regulatory Commission (FERC) relicensing,
and therefore are limited to short-term collection
efforts.
The upper reservoir calibration was completed
using 13 thermal profiles. Daily average stream
inflow and outflow temperatures from October 1,
2000 to October 19, 2004, were collected by Sacramento Municipal Utilities District (SMUD, unpublished data) and observed streamflow measurements
were from the U.S. Geological Survey (USGS) (USGS,
2012). The middle reservoir was calibrated with two
years of measured data from Nevada Irrigation District’s (NID) Yuba-Bear project (NID, unpublished
data). Inflow and outflow temperature data included
15-min temperature observations from June 3, 2008
ð2Þ
where Tw is water temperature (°C), t is time (wk),
Hnet is net heat exchange at both the bed- and airwater interfaces (W/m2), Cp is the specific heat of
water (4,185 J/kg/°C), q is water density (1,000 kg/m3),
and d is water depth (m). Equation (2) assumes that
rivers are advection dominated and that changes in
longitudinal stream temperatures are driven by meteorological conditions, which alter stream temperatures with respect to time. It further assumes a
rectangular channel where water surface is the same
width as the streambed. For a full derivation and
additional model detail, see Null et al. (2013). Equation (2) produces a time series of dynamic stream
temperatures where water seeks equilibrium temperature, but is constrained by boundary conditions,
travel time, and river geometry.
Stream temperatures are estimated separately for
tributaries and mainstem rivers. This allows attenuation of tributaries to be assessed. Mainstem and tributary channels mix at the bottom of each reach using
a mass balance approach. In addition to boundary
conditions for flow and stream temperature obtained
JAWRA
VIERS
1462
JOURNAL
OF THE
AMERICAN WATER RESOURCES ASSOCIATION
A METHOD
TO
CONSIDER WHETHER DAMS MITIGATE CLIMATE CHANGE EFFECTS
ON
STREAM TEMPERATURES
flow hydrologic conditions using monthly natural flow
estimates at watershed outlets, and snow water
equivalent measurements at upstream sites within
each watershed (Young et al., 2009). Overall, WEAP
represents flow hydrographs well at the watershed
scale, and mean bias between simulated and measured natural flow at watershed outlets was 1%,
with a range 6 to 2%. RTEMP models were also previously calibrated for unregulated conditions and are
described in detail in Null et al. (2013). Simulated
stream temperatures fit measured data well and the
timing of seasonal maximum temperatures was consistent between modeled and observed data. MAE
was within 1.2°C of measured maximum or minimum
temperatures, with a mean of 0.4°C.
to October 10, 2008, and June 12, 2009 to October 16,
2009. Reservoir thermal profile measurements were
collected by NID in July, August, September, and
October of 2008 and 2009. Inflow discharge was estimated with USGS measured daily outflows through
the Bowman-Spaulding Canal and Canyon Creek
below Spaulding (USGS, 2012). The lower reservoir
was calibrated with Folsom Lake inflow and outflow,
as well as stream temperature data from the American
River below Folsom Reservoir from September 30,
2000 to April 10, 2011 (CDEC, 2011). No thermal profiles for Folsom Reservoir were available for calibration. Evaporation coefficients and radiation absorption
coefficients were adjusted so modeled profiles and outlet temperatures reasonably matched observed data,
and represent observed stratification and mixing
within each reservoir.
Regulated Stream Temperatures in a Changing
Climate
Study findings focus on reservoir release temperatures with climate change for reservoirs of different
sizes and elevations. Results also compare regulated
stream temperatures with unregulated stream temperatures to highlight how climate change may affect
regulated systems differently than unregulated systems. We discuss study results by focusing on the
central latitude, although the northern, central, and
southern latitudes are compared with longitudinal
stream temperature profiles at the end of this section.
Major findings from this work illustrate the utility of
our approach as a proof of concept study. Overall,
modeled stream temperatures with anticipated climate change representing ftp conditions are cooler
than current observed stream temperatures. We
discuss this and provide a brief analysis of input data
at the end of this section.
RESULTS
Overall, vertical temperature profiles within reservoirs matched observations well, indicating that models capture the timing and depth of thermal
stratification (see Null et al., 2012 for figures of modeled and observed vertical temperature profiles).
Mean bias, mean absolute error (MAE), and root
mean square error (RMSE) were calculated for reservoir release temperatures to evaluate WQRRS performance for the upper, middle (measured data from
two similar sites were compared), and lower reservoirs. MAE was always less than 2.4°C and averaged
1.8°C for all sites with measured data. Calibration
statistics are summarized in Table 5. Reservoir
release temperatures are important for stream temperatures because they provide the boundary conditions for downstream temperatures and thus can
drive thermal response.
WEAP models were previously calibrated for the
15 west-slope Sierra Nevada watersheds with natural
Reservoir Release Temperatures. Modeling
suggests that average reservoir release temperatures
will be resilient to climate change through winter
(Figure 5). Reservoir release temperatures warm
TABLE 5. Calibration Model Fit Statistics by Reservoir.
Upper
Measured Location
n, count
Mean bias, °C
MAE, °C
RMSE, °C
Calibration date,
month/date/year
Middle
Lower
Icehouse Release
Bowman-Spaulding
Canal
Canyon Creek below
Spaulding
Folsom
Release
187
0.1
1.1
1.5
10/1/00-10/19/04
40
0.7
1.9
2.3
6/3/08-10/10/08, 6/12/09-10/16/09
66
0.9
2.4
3.0
6/3/08-10/10/08, 6/12/09-10/16/09
547
0.7
1.6
2.0
9/30/00-4/10/11
Note: MAE, mean absolute error; RMSE, root mean square error.
JOURNAL
OF THE
AMERICAN WATER RESOURCES ASSOCIATION
1463
JAWRA
NULL, LIGARE,
AND
VIERS
FIGURE 5. Average Weekly Central Latitude MIROC32med and NCARPCM1 Reservoir Release
Temperatures by Time Period for the (a) Upper, (b) Middle, and (c) Lower Reservoirs
combination of outlet works infrastructure, reservoir
operations, depth, and inflow temperatures. Because
the middle reservoir is calibrated to Bowman Reservoir, some of the above factors are implicitly included
in this analysis. In addition, the modeled middle reservoir (Figure 5b) is deeper than the upper reservoir
(Figure 5a), potentially affecting the volume of the
cold-water pool, with cooler release temperatures during summer. The middle reservoir illustrates the
importance of parameterizing and calibrating models
for specific systems or regions when using this
approach.
On average, the warmest water is released in
August to early September for the upper reservoir,
late July to early August for the middle reservoir,
and early October for the large, lower reservoir (Figure 5). However, there is considerable variability
between years that is not depicted in the annually
averaged figures. For the upper reservoir, the timing
of the warmest releases generally became more consistent with more severe climate change (mtp and
from climate change beginning in early May. Springtime warming was particularly evident at the upper
and middle reservoirs (Figures 5a and 5b), indicating
that the effects of climate change on reservoir
releases may be more severe at higher elevations.
Simulations using MIROC32med input climate data
were warmer than those using NCARPCM1 data
(MIROC32med predicts a warmer and drier future
than NCARPCM1). In fact, release temperatures for
NCARPCM1 ftp were similar to release temperatures
for MIROC32med ntp (Figure 5). This is indicative of
uncertainty in climate modeling, such as representation of physical processes and sensitivity to greenhouse gas forcings.
Reservoir release temperatures were cooler for the
middle reservoir than the upper reservoir, reflecting
model calibration and reservoir depth. The upper and
middle reservoirs were calibrated to Ice House and
Bowman Reservoirs, respectively. Bowman Reservoir
had exceptionally cold release temperatures (much
colder than Ice House Reservoir), a result of the
JAWRA
1464
JOURNAL
OF THE
AMERICAN WATER RESOURCES ASSOCIATION
A METHOD
TO
CONSIDER WHETHER DAMS MITIGATE CLIMATE CHANGE EFFECTS
ON
STREAM TEMPERATURES
outlook into the future (Figure 6, first and third columns, time progresses from the htp, ntp, mtp, and
ftp in the downward direction). For all unregulated
runs, the middle river that reaches from approximately 150-250 river kilometer (RKM) is the warmest, before tributaries subsequently join the
mainstem with cooler temperatures and higher flows
that moderate mainstem temperatures. Toward the
end of the 21st Century, modeling indicates that
streams not only become warmer, but that the average duration with warmer temperatures is extended
as well (x-axes in Figure 6).
Unregulated winter stream temperatures become
approximately 2°C warmer as well. For MIROC32med
unregulated simulations, maximum average weekly
ftp). For all reservoirs, the warmest release temperatures coincided with extended periods of low reservoir
storage which forced mixing of stratified waters
within the reservoir, or less commonly, with spill
events. The upper reservoir typically had less stored
water earlier in the year than the large, lower reservoir. For the lower reservoir, low reservoir levels
most commonly occurred in early fall, which is consistent with observed storage levels in large rim dams
(CDEC, 2011).
Effects of Reservoirs on Stream Temperatures. For unregulated conditions, modeling suggests that stream temperatures warm with more
severe climate change, represented as a farther
FIGURE 6. Modeled Average Weekly Stream Temperatures for the Central Latitude
with Regulated and Unregulated Conditions for Both GCMs and All Time Periods.
JOURNAL
OF THE
AMERICAN WATER RESOURCES ASSOCIATION
1465
JAWRA
NULL, LIGARE,
AND
MIROC32med and NCARPCM1 plots represent
uncertainty from climate modeling assumptions and
representation. This modeling indicates changes to
stream temperatures from reservoirs of different
sizes and elevations.
stream temperatures are 13.1, 14.4, 15.8, and 18.2°C
for htp, ntp, mtp, and ftp, respectively. For unregulated NCARPCM1 runs, stream temperatures are
13.2, 13.2, 14.3, and 14.8°C for htp, ntp, mtp, and ftp,
respectively. The warmer stream temperatures in the
MIROC32med runs are largely driven by air temperatures that increase at double the rate of NCARPCM1
air temperatures for the ntp, mtp, and ftp.
The relatively cool reservoir releases of the upper
and middle reservoirs (Figure 5) provide cooler stream
temperatures during summer and/or a shorter duration of warm stream temperatures than occur with
unregulated conditions. However, the longitudinal
length of cooler stream temperatures below dams is
variable (Figure 6 — second and fourth columns). For
example, cooler temperatures below the middle reservoir vary by approximately 10-90 RKM depending on
GCM input data and modeled time period. There is no
clear trend as to whether that distance is increasing
or decreasing with climate change and may be driven
largely by flow, with larger flow volumes taking longer
to reequilibrate with atmospheric conditions (Gu
et al., 1998). Below the large, lower dam there is a
clear shift in the timing of annual maximum stream
temperatures from midsummer toward autumn. For
both GCMs and all modeled time periods, the lower
dam cools summer stream temperatures, increases
autumn stream temperatures, extends the duration of
warm temperatures, and increases winter temperatures. This study reiterates existing research that
dams alter the thermal regime of rivers in both space
and time (Olden and Naiman, 2010).
Stream temperature differences between regulated
and unregulated conditions using MIROC32med and
NCARPCM1 climate data (Figure 7) further illustrate the effect of dam regulation. For MIROC32med
runs, reservoir releases cool downstream temperatures by up to 4°C during June through August;
although stream temperatures below reservoirs are
nearly 9°C warmer during autumn and early winter
with regulation, illustrating potential changes when
coldwater pools have been depleted. The progression
of time (htp, ntp, mtp, to ftp) shows that changes
become more pronounced with more severe climate
change. Reservoir releases during summer months
can cool stream temperatures relatively more with
more extreme climate change (ftp) than for the htp,
although climate change causes stream temperatures
to rise overall for both regulated and unregulated
conditions with extreme climate change. In other
words, in the ftp future, we should expect reservoir
releases to provide a more pronounced cooling effect
although stream temperature increases. The
NCARPCM1 runs show similar trends, but with less
pronounced summer cooling downstream of reservoirs.
Temperature
differences
between
the
JAWRA
VIERS
Longitudinal Stream Temperature Profiles
Longitudinal stream temperature profiles show that
dams abruptly cool downstream river reaches during
summer (Figure 8). In headwater reaches in July,
streams start with elevated temperatures, particularly
at the northern latitude (Figure 8a, RKM 0 on x-axis).
RTEMP struggles to accurately estimate stream temperature in headwater reaches with very low flow
conditions (<~0.3 m3/s). In real headwater systems,
microtopographic shading, hyporheic flow, partial
snow cover at high elevations, and snowmelt likely
moderate stream temperatures somewhat; although
observed August stream temperature data show warmer stream temperatures above high-elevation reservoirs such as French Meadows and Hell Hole than
below them (American watershed) (PCWA, 2007).
Downstream of the middle reservoir, stream temperatures warm in response to July atmospheric conditions. Stream temperatures warm longitudinally below
the lower reservoir, although at a slower rate than
occurred below the middle reservoir because flow volume is higher. Overall stream temperatures increase
more with the MIROC32med climate-driven input data
than the NCARPCM1 input data. Regulated streams
warm more quickly than unregulated streams for all
latitudes and time periods because flows are reduced
with regulation — flow in the reach below the middle
reservoir is 32-55 m3/s with unregulated conditions,
but is 7.5-30 m3/s with regulated conditions. Longitudinal heating occurs more rapidly for the regulated runs
because the river has a slower travel time (i.e., longer
exposure to atmospheric conditions) and less mass to be
heated.
Figure 8 implies that flow and temperature can be
managed with dam releases to maintain and enhance
suitable instream habitat for coldwater species with
climate change. Modeling suggests the highest, smallest dam reduced summer stream temperatures the
least because high, small dams generally empty each
year (do not have carryover storage) and contribute
relatively more flow to rivers than lower elevation
dams do (typically because river flow is low at upper
elevations).
Stream temperatures vary slightly by latitude,
although clear trends are not obvious. Differences in
stream temperatures appear to be sensitive to local
differences in climate input data rather than different rates of heating by latitude.
1466
JOURNAL
OF THE
AMERICAN WATER RESOURCES ASSOCIATION
A METHOD
TO
CONSIDER WHETHER DAMS MITIGATE CLIMATE CHANGE EFFECTS
ON
STREAM TEMPERATURES
FIGURE 7. Average Weekly Regulated Minus Unregulated Stream Temperature Difference for the Central Watershed for All Time Periods.
both GCMs used in this study as well as the Daymet
dataset used in Null et al. (2013) (Table 6). Daymet
data are daily, interpolated surface data (aggregated
to a weekly timestep here) (Thorton et al.,1997). We
also include projected future change for each model
projection. Future climate projects from Null et al.
(2013) simply model uniform air temperature
increases of 2, 4, and 6°C as a sensitivity analysis of
unregulated stream temperature response to climate
warming. Table 6 shows that for htp conditions,
mean annual air temperature for the central latitude
watershed is nearly 3°C warmer with the Daymet
dataset than either GCM, and air temperature is
cooler for the NCARPCM1 dataset by the end of the
21st Century than Daymet’s htp conditions.
Limitations
Input Data Uncertainty. Climate and hydrology
input data were used to estimate instream flow and
the atmospheric heat budget, both of which drive
stream temperatures. The cool stream temperatures
suggested by this modeling effort, where maximum
average annual stream temperatures remain below
approximately 18°C for all climate alternatives, time
periods, and management alternatives (regulated vs.
unregulated) are representative of these data.
It is beyond the scope of this study to thoroughly
analyze the climate datasets or to quantify their
accuracy. However, we have provided a comparison of
mean annual air temperature and precipitation for
JOURNAL
OF THE
AMERICAN WATER RESOURCES ASSOCIATION
1467
JAWRA
NULL, LIGARE,
AND
VIERS
FIGURE 8. Average July Stream Temperature Longitudinal Profile for MIROC32med and
NCARPCM1 htp and ftp for the (a) Northern, (b) Central, and (C) Southern Latitudes.
TABLE 6. Comparison of Climate Data (future values represent average change
between each model projection and that model’s simulated historical climate).
1970-1990 (historical)
Climate
Dataset
MIROC32med
NCARPCM1
Daymet
2010-2030
2040-2060
Air
Temp, °C
Precip,
mm/yr
DAir
Temp, °C
DPrecip,
%
DAir
Temp, °C
DPrecip,
%
DAir
Temp, °C
DPrecip,
%
8.1
8.2
10.9
197
205
161
1.3
0.2
2
0.8
7.5
0
2.7
1.1
4
13.3
6.0
0
4.7
2.1
6
20.1
12.4
0
Compounding
comparison
of
input
data,
MIROC32med and NCARPCM1 climate estimates
were input into the VIC hydrology model (WEAP was
JAWRA
2070-2090
only used to model infrastructure and priorities for
water use with regulated conditions) and Daymet climate data were input into the WEAP hydrologic
1468
JOURNAL
OF THE
AMERICAN WATER RESOURCES ASSOCIATION
A METHOD
TO
CONSIDER WHETHER DAMS MITIGATE CLIMATE CHANGE EFFECTS
This study focuses on stream temperatures not
changes in reservoir operations, and therefore we
held flood control, hydropower, and water supply
demands constant. A limitation of this simplification
is a lack of weekly variability in streamflow that
would be the result of a more complex hydropower
simulation model. Also, because water and power
demands do not change with future climate scenarios,
our approach does not represent realistic adaptations.
Climate change will affect reservoir operations, which
will further alter reservoir temperatures. Previous
hydropower studies in the Sierra Nevada have shown
that reservoirs fill earlier and empty earlier as inflow
shifts from springtime snowmelt to winter snowmelt
and rain events (Vicuna et al., 2008). The strength of
our method is that it complements existing reservoir
operations studies (that typically ignore water quality) and highlights water temperature changes from
climate as opposed to reservoir operations. A reservoir operations model that includes our stream temperature modeling module could provide valuable
insights into reservoir operations that include environmental uses with human uses. Finer temporal resolution of modeled stream temperatures would also
complement hydropower simulation models (which
typically use hourly or daily timesteps to simulate
peaking), but is not needed for water supply simulation models (which often use monthly timesteps).
DISCUSSION AND MAJOR FINDINGS
The thermal regime of rivers has been fundamentally altered by humans from water development,
land use changes, and increasingly, climate-induced
changes to atmospheric conditions. Many aquatic ecosystems are thermally sensitive and have had significant reductions in habitat with these changes. In
California’s Sierra Nevada, coldwater habitat has
generally decreased with water development and land
use changes (Yoshiyama et al., 1998). Maintaining
coldwater habitat with anticipated climate change is
a future challenge.
Overall, our approach is useful as a proof of concept study that can be incorporated with water operations models. The effects of dams on downstream
temperatures vary by season, stream length between
reservoirs, reservoir size, outlet structure, and reservoir elevation. During summer, dams provide a step
reduction in stream temperatures, which then warm
longitudinally in response to atmospheric conditions.
Increasingly severe climate change (represented with
FIGURE 9. Modeled Stream Temperatures at the Outlet of the
Stanislaus River (500 m elevation) for Historical and Future
Climate Conditions Using MIROC32med and Daymet Input Data
(T0 is historical conditions and T2, T4, and T6 are 2, 4, and 6°C
climate warming, respectively) (unregulated stream temperatures
with Daymet input data from Null et al., 2013).
OF THE
AMERICAN WATER RESOURCES ASSOCIATION
STREAM TEMPERATURES
Reservoir Operations
model (Young et al., 2009). Also, future climate
assumptions are different in Figure 9, the GCMs
explicitly model physical processes whereas Daymet
data are increased by 2°C increments as a sensitivity
analysis (all other data and parameters are identical).
Thus, both input data and hydrologic modeling were
different in this comparison, but it is relevant
because both methods are common for modeling climate impacts on water resources and to highlight the
uncertainty of modeling climate impacts. This comparison shows that different input data lead to very
different stream temperature results (Figure 9). Modeled htp stream temperatures with unregulated conditions (black lines) differ markedly and highlight the
compounding uncertainty inherent in modeling anticipated hydroclimate effects on water quality. Average
annual maximum stream temperature is 23.8°C using
the Daymet climate data, but is over 10°C cooler, at
13.1°C with the MIROC32med climate data. Stream
temperatures reach their average annual maximum
in the first week of August for both simulations.
Climate is notably difficult to model in mountain
regions due to orographic microclimates and these
data suggest future modeling and research should
focus on improving accuracy of climate predictions as
well as calibration efforts for mountain regions.
The input climate and hydrology data discussed
above should be noted as a limitation of this study.
More complete input and calibration data would
improve results. Calibration of climate and hydrology
input data has primarily been completed for lower
elevation locations in the Central Valley, with less
certainty for high-elevation locations of the Sierra
Nevada. While this compromises results, it also highlights future data needs and is a pertinent finding of
the study.
JOURNAL
ON
1469
JAWRA
NULL, LIGARE,
AND
ecosystems. Our focus was on relative changes rather
than absolute stream temperatures to complement
existing reservoir operations research. Incorporating
our approach with reservoir operations studies could
provide valuable insights into the potential of existing infrastructure to mitigate the effects of climate
change. Future research includes modifying dam
outlet works to highlight promising infrastructure
improvements and adaptation strategies for anticipated hydroclimatic changes. Our method could be
applied to other locations on a case-by-case basis
using input data for the new region, changing reservoir parameters (such as capacity, outlet structure,
depth, elevation, distance between reservoirs…) and
recalibrating models to improve understanding of
regional climate change effects on water quality.
a longer future outlook) indicates that although the
absolute temperature of stream temperatures
increase from htp conditions, the thermal reduction
from reservoir releases grows somewhat more pronounced and maintains cool water into summer
months. This implies that reservoirs may be able to
release cool water and offer a potential mitigation
strategy for climate change-induced stream warming
for short distances downstream. Changing water and
power demands will also impact operations and
should be considered in future work.
In nonstationary climates, the geographic distributions of habitats and ecosystems are also likely to
change (Hanak et al., 2011). This study indicates that
summertime coldwater habitat is anticipated to exist
immediately downstream from dams but stream temperatures rise with distance from dams. This indicates that coldwater habitat may persist in regulated
rivers with climate change (particularly if dams are
operated to preserve winter water in stratified reservoirs), although possibly not in the same locations as
coldwater habitat historically existed. This presents
future challenges for water managers since many of
the laws and regulations to protect ecosystems (such
as the Endangered Species Act) were created with
the assumption of ecological stationarity. Given the
generally cool stream temperatures estimated in this
effort with the MIROC32med- and NCARPCM-driven
VIC input data (and the incongruence with previously modeled unregulated conditions in west-slope
Sierra Nevada rivers) (Null et al., 2013), it is beyond
the scope of this study to analyze the extent of coldwater habitat from river regulation with climate
change.
In California, water is tightly managed. Economicengineering modeling suggests in a warmer and drier
future, competition for already scarce water could
increase by an order of magnitude (Harou et al.,
2010). This means that attention should focus on
managing water quality to maintain and enhance
aquatic ecosystems, especially when maintaining
water quality may reduce the need for environmental
instream flows (Null et al., 2010b). With increasing
water scarcity, it will be important to understand the
factors limiting aquatic ecosystems in Sierra Nevada
rivers and manage those specific problems.
Modeling studies such as this are needed to better
understand the thermal effects from dams with climate change and the potential to mitigate for anticipated changes. Here, we use a novel approach to
apply atmospheric projections on a global scale to regulated stream temperatures at the mesoscale. This
proof of concept study shows the regional modeling
approach is worthwhile for assessing potential for
thermal management of reservoir releases to maintain and enhance downstream coldwater aquatic
JAWRA
VIERS
ACKNOWLEDGMENTS
This research was made possible by funding from the California
Energy Commission and was submitted as a white paper to them
under the title “Water and Energy Sector Vulnerability to Climate
Warming in the Sierra Nevada: A Method to Consider Whether
Dams Mitigate Climate Change Effects on Stream Temperatures”.
We would like to thank Jeff Mount and Mike Deas for scholarly
contributions, David Rheinheimer for programming support, as
wells as Danielle Salt and Paige Dulberg for downloading measured calibration data, running models, and analyzing data. We
would also like to acknowledge the Sacramento Municipal Utility
District for sharing measured water temperature data used for
model testing. Thanks to Mary Tyree and the California Nevada
Applications Program/California Climate Change Center (directed
by the Climate Research Division of Scripps Institute for Oceanography) for sharing VIC climate data. Finally, we would like to
thank Guido Franco and PIER staff for their direction and oversight, as well as anonymous reviewers for their thoughtful comments and suggestions. All authors were affiliated with University
of California, Davis when the majority of this work took place.
LITERATURE CITED
Aspen Environmental and M.Cubed, 2005. Potential Changes in
Hydropower Production from Global Climate Change in California and the Western United States. Prepared for California
Energy Commission, Sacramento, California.
Barnett, T.P., J.C. Adam, and D.P. Lettenmaier, 2005. Potential
Impacts of a Warming Climate on Water Availability in SnowDominated Regions. Nature 438(7066):303-309, doi: 10.1038/
nature04141.
Battin, J., M.W. Wiley, M.H. Ruckelshaus, R.N. Palmer, E. Korb,
K.K. Bartz, and H. Imaki, 2007. Projected Impacts of Climate
Change on Salmon Habitat Restoration. Proceedings of the
National Academy of Sciences of the United States of America
140 (16):6720-6725, doi: 10.1073/pnas.0701685104.
Bogan, T., O. Mohseni, and H.G. Stefan, 2003. Stream Temperature-Equilibrium Temperature Relationship. Water Resources
Research 39(9):1245, doi: 10.1029/2003WR002034.
Caissie, D.M.G., M.G., Satish, and N. El-Jabi, 2005. Predicting
River Water Temperatures Using the Equilibrium Temperature
Concept with Application on Miramichi River Catchments (New
1470
JOURNAL
OF THE
AMERICAN WATER RESOURCES ASSOCIATION
A METHOD
TO
CONSIDER WHETHER DAMS MITIGATE CLIMATE CHANGE EFFECTS
OF THE
AMERICAN WATER RESOURCES ASSOCIATION
STREAM TEMPERATURES
Karamouz, M., A. Moridi, and H.M. Fayyazi, 2008. Dealing with
Conflict over Water Quality and Quantity Allocation: A Case
Study. Scientia Iranica 15(1):34-49.
Kaushal, S.S., G.E. Likens, N.A. Jaworski, M.L. Pace, A.M. Sides,
D. Seekell, K.T. Belt, D.H. Secor, and R.L. Wingate, 2011. Rising Stream and River Temperatures in the United States. Frontiers in Ecology and the Environment 8(9):461-466, doi: 10.1890/
090037.
Lopes, L.F.G., J.S.A. do Carmo, and R.M.V. Cortes, 2003. Influence
of Dam-Reservoirs Exploitation on the Water Quality. River
Basin Manage II 7: 221-230.
Madani, K. and J. Lund, 2010. Estimated Impacts of Climate
Warming on California’s High-Elevation Hydropower. Climatic
Change 102(3):521-538, doi: 10.1007/s10584-009-9750-8.
Maurer, E., 2007. Uncertainty in Hydrologic Impacts of Climate
Change in the Sierra Nevada, California, Under Two Emissions
Scenarios. Climatic Change 82(3):309-325.
Maurer, E.P., H.G. Hidalgo, T. Das, M.D. Dettinger, and D.R. Cayan, 2010. The Utility of Daily Large-Scale Climate Data in the
Assessment of Climate Change Impacts on Daily Streamflow in
California. Hydrology and Earth System Sciences 14:1125-1138.
Medellın-Azuara, J., J. Harou, M. Olivares, K. Madani, J. Lund, R.
Howitt, S. Tanaka, X. Marion Jenkins, and T. Zhu, 2008. Adaptability and Adaptations of California’s Water Supply System to
Dry Climate Warming. Climatic Change 87(0):75-90, doi: 10.
1007/s10584-007-9355-z.
Mehta, V.K., D.E. Rheinheimer, D. Yates, D.R. Purkey, J.H. Viers,
C.A. Young, and J.F. Mount, 2011. Potential Impacts on Hydrology and Hydropower Production Under Climate Warming of the
Sierra Nevada. Journal of Water and Climate Change 2(1):2943, doi: 10.2166/wcc.2011.054.
Meyer, G.K., G.T. Orlob, and C. Jokiel, 1996. Effects of Climate
Change on Water Quality in the Central Valley of California.
In: Water Resources Management in the Face of Climatic/
Hydrologic Uncertainties, Z. Kaczmarek, K.M. Strzepek,
L. Somlyody, V. Priazhinskaya (Editors). Kluwer Academic
Publishers, Dordrecht, The Netherlands, pp. 274-299.
Mohseni, O. and H.G. Stefan, 1999. Stream Temperature/Air Temperature Relationship: A Physical Interpretation. Journal of
Hydrology 218(3-4):128-141, doi: 10.1016/S0022-1694(99)00034-7.
Null, S.E., M.L. Deas, and J.R. Lund, 2010b. Flow and Water Temperature Simulation for Habitat Restoration in the Shasta
River, California. River Research and Applications 26(6):663681, doi: 10.1002/rra.1288.
Null, S.E., S.T. Ligare, and J.H. Viers, 2012. Water and Energy
Sector Vulnerability to Climate Warming in the Sierra Nevada:
A Method to Consider Whether Dams Mitigate Climate Change
Effects on Stream Temperatures. Prepared for the California
Energy Commission. http://watershed.ucdavis.edu/files/biblio/
CWS-2011-001.pdf, accessed January 2013.
Null, S.E., J. Viers, M. Deas, S. Tanaka, and J. Mount, 2013.
Stream Temperature Sensitivity to Climate Warming in California’s Sierra Nevada: Impacts to Coldwater Habitat. Climatic
Change 116(1):1-22, doi: 10.1007/s10584-012-0459-8.
Null, S.E., J.H. Viers, and J.F. Mount, 2010a. Hydrologic Response
and Watershed Sensitivity to Climate Warming in California’s
Sierra Nevada. PLoS ONE 5(4):e9932, doi: 10.1371/journal/pone.
0009932.
Olden, J.D. and R.J. Naiman, 2010. Incorporating Thermal Regimes
into Environmental Flows Assessments: Modifying Dam Operations to Restore Freshwater Ecosystem Integrity. Freshwater
Biology 55(1):86-107, doi: 10.1111/j.1365-2427.2009.02179.x.
PCWA, 2007. 2006 Water Temperature Study Report. Placer
County Water Agency. Auburn, California. June 2007.
Poff, N.L., 2009. Managing for Variability to Sustain Freshwater
Ecosystems. Journal of Water Resources Planning and Management 135(1):1-4.
Brunswick, Canada). Hydrological Processes 19:2137-2153, doi:
10.1002/hyp.5684.
Cayan, D.R., M. Tyree, M.D. Dettinger, H. Hidalgo, T. Das, E. Maurer, P. Bromirski, N. Graham, and R. Flick, 2009. Climate
Change Scenarios and Sea Level Rise Estimates for the California 2009 Climate Change Scenarios Assessment. Prepared for
California Energy Commission, Sacramento, California.
CDEC (California Data Exchange Center), 2011. California Data
Exchange Center. Operated by CA. Department of Water
Resources. http://cdec.water.ca.gov, accessed May 2011.
Das, T., M. Dettinger, D. Cayan, and H. Hidalgo, 2011. Potential
Increase in Floods in California’s Sierra Nevada Under Future
Climate Projections. Climatic Change 109(0):71-94.
Deas, M.L. and G.T. Orlob, 1999. Klamath River Modeling Project.
University of California, Davis, California.
Department of Water Resources, 2009. California Water Plan
Update. http://www.waterplan.water.ca.gov/cwpu2009/index.cfm,
accessed May 2011.
Eaton, J.G. and R.M. Scheller, 1996. Effects of Climate Warming
on Fish Thermal Habitat in Streams of the United States. Limnology and Oceanography 41(5):1109-1115.
Edinger, J.E., D.W. Duttweiler, and J.C. Geyer, 1968. The
Response of Water Temperature to Meteorological Conditions.
Water Resources Research 4(5):1137-1143.
Gonenc, I.E., B.B. Baykal, O. Ince, and O. Asikoglu, 1997. Ecosystem Modelling of Coastal Lagoons for Sustainable Management.
A Case Study: Kucukcekmece Lagoon, Turkey. International
Journal of Salt Lake Research 6:91-105.
Gooseff, M., K. Strzepek, and S. Chapra, 2005. Modeling the Potential Effects of Climate Change on Water Temperature Downstream of a Shallow Reservoir, Lower Madison River, MT.
Climatic Change 68(3):331-353.
Gu, R., S. Montgomery, and T. Austin, 1998. Quantifying the
Effects of Stream Discharge on Summer River Temperature.
Hydrological Sciences 43(6):885-904.
Hamlet, A.F. and D.P. Lettenmaier, 1999. Effects of Climate
Change on Hydrology and Water Resources in the Columbia
River Basin. Journal of the American Water Resources Association 35(6):1597-1623, doi: 10.1111/j.1752-1688.1999.tb04240.x.
Hanak, E., J. Lund, A. Dinar, R. Gray, R. Howitt, J. Mount, P. Moyle, and B. Thompson, 2011. Managing California’s Water: From
Conflict to Reconciliation. Public Policy Institute of California,
San Francisco, California.
Hari, R.E., D.M. Livingstone, R. Siber, P. Burkhardt-Holm, and H.
Guttinger, 2006. Consequences of Climatic Change for Water
Temperature and Brown Trout Populations in Alpine Rivers
and Streams. Global Change Biology 12:10-26, doi: 10.1111/
j.1365-2486.2005.01051.x.
Harou, J.J., J. Medellin-Azuara, T. Zhu, S.K. Tanaka, J.R. Lund,
S. Stine, M.A. Olivares, and M.W. Jenkins, 2010. Economic Consequences of Optimized Water Management for a Prolonged,
Severe Drought in California. Water Resources Research. 46(5):
W05522.
Isaak, D.J., C.H. Luce, B.E. Rieman, D.E. Nagel, E.E. Peterson,
D.L. Horan, S. Parkes, and G.L. Chandler, 2010. Effects of Climate Change and Wildfire on Stream Temperatures and Salmonid Thermal Habitat in a Mountain River Network. Ecological
Applications 20(5):1350-1371.
Jager, H.I., W. Van Winkle, and B.D. Holcomb, 1999. Would
Hydrologic Climate Changes in Sierra Nevada Streams Influence Trout Persistence? Transactions of the American Fisheries
Society 128:222-240.
Johnson, B.M., L. Saito, M.A. Anderson, P. Weiss, M. Andre, and
D.G. Fontane, 2004. Effects of Climate and Dam Operations on
Reservoir Thermal Structure. Journal of Water Resources Planning and Management 130(2):112-122, doi: 10.1061/(ASCE)
0733-9496.
JOURNAL
ON
1471
JAWRA
NULL, LIGARE,
AND
VIERS
Poole, G.C. and C.H. Berman, 2001. An Ecological Perspective on
In-Stream Temperature: Natural Heat Dynamics and Mechanisms of Human-Caused Thermal Degradation. Environmental
Management 27(6):797-802, doi: 10.1007/s002670010188.
Rahel, F.J. and J.D. Olden, 2008. Assessing the Effects of Climate Change on Aquatic Invasive Species. Conservation Biology
22(3):521-533, doi: 10.1111/j.1523-1739.2008.00950.x.
Sinokrot, B.A., H.G. Stefan, J.H. McCormick, and J.G. Eaton, 1995.
Modeling of Climate Change Effects on Stream Teperatures and
Fish Habitats Below Dams and Near Groundwater Inputs. Climatic Change 30:181-200.
Thorton, P.E., S.W. Running, and M.A. White, 1997. Generating
Surfaces of Daily Meteorological Variables over Large Regions
of Complex Terrain. Journal of Hydrology 190:214-251.
Tung, C., T. Lee, and Y. Yang, 2006. Modelling Climate-Change
Impacts on Stream Temperature of Formosan Landlocked Salmon Habitat. Hydrological Processes 20(7):1629-1649, doi: 10.
1002/hyp.5959.
USACE-HEC (US Army Corps of Engineers-Hydrologic Engineering Center), 1986. WQRRS Water Quality for River-Reservoir
Systems User’s Manual. Hydrologic Engineering Center, Davis,
California.
USGS (U.S. Geological Survey), 2012. Water Resources of the United States. http://water.usgs.gov/, accessed January 2013.
Vicuna, S., R. Leonardson, M. Hanemann, L. Dale, and J. Dracup,
2008. Climate Change Impacts on High Elevation Hydropower
Generation in California’s Sierra Nevada: A Case Study in the
Upper American River. Climatic Change 87(0):123-137.
Weare, B.C., 2009. How Will Changes in Global Climate Influence
California? California Agriculture 63(2):59-66, doi: 10.3733/ca.
v063n02p59.
Wright, S.A., C.R. Anderson, and N. Voichick, 2009. A Simplified
Water Temperature Model for the Colorado River Below Glen
Canyon Dam. River Research and Applications 25:675-686, doi:
10.1002/rra.1179.
Yates, D., H. Galbraith, D. Purkey, A. Huber-Lee, J. Sieber, J.
West, S. Herrod-Julius, and B. Joyce, 2008. Climate Warming,
Water Storage, and Chinook Salmon in California’s Sacramento
Valley. Climatic Change 91(3-4):335-350, doi: 10.1007/s10584008-9427-8.
Yates, D., J. Sieber, D. Purkey, and A. Huber-Lee, 2005. WEAP21—
A Demand-, Priority-, and Preference-Driven Water Planning
Model. Water International 30(4):487-500, doi: 10.1080/
02508060508691893.
Yoshiyama, R.M., F.W. Fisher, and P.B. Moyle, 1998. Historical
Abundance and Decline of Chinook Salmon in the Central Valley Region of California. North American Journal of Fisheries
Management 18, 487-521.
Young, C.A., M. Escobar, M. Fernandes, B. Joyce, M. Kiparsky,
J.F. Mount, V. Mehta, J.H. Viers, and D. Yates, 2009. Modeling
the Hydrology of California’s Sierra Nevada for Sub-Watershed
Scale Adaptation to Climate Change. Journal of the
American Water Resources Association 45(6):1409-1423, doi: 10.
1111/j.1752-1688.2009.00375x.
JAWRA
1472
JOURNAL
OF THE
AMERICAN WATER RESOURCES ASSOCIATION