Incorporating Thermal Regimes into Environmental Flows Assessments: Modifying Dam Operations to Restore Freshwater Ecosystem Integrity

Abstract

1. Despite escalating conflict over fresh water, recent years have witnessed a growing realisation that human society must modify its behaviour to ensure long-term ecological vitality of riverine ecosystems. In response, ecologists have been increasingly asked to guide instream flow management by providing ‘environmental flow’ prescriptions for sustaining the ecological integrity of riverine systems.
2. Environmental flows are typically discussed in the context of water releases from dams and water allocation for extraction (such as for urban use or irrigation), where there is general agreement that rivers need to exhibit some resemblance of natural flow variability necessary to support a functioning ecosystem. Although productive dialogue continues on how best to define environmental flows, these discussions have been focused primarily on water quantity without explicit consideration of many components of water quality, including water temperature – a fundamental ecological variable.
3.Many human activities on the landscape have modified riverine thermal regimes. In particular, many dams have modified thermal regimes by selectively releasing hypolimnetic (cold) or epilimnetic (warm) water from thermally stratified reservoirs to the detriment of entire assemblages of native organisms. Despite the global scope of thermal alteration by dams, the prevention or mitigation of thermal degradation has not entered the conversation when environmental flows are discussed.
4.Here, we propose that a river’s thermal regime is a key, yet poorly acknowledged, component of environmental flows. This study explores the concept of the natural thermal regime, reviews how dam operations modify thermal regimes, and discusses the ecological implications of thermal alteration for freshwater ecosystems. We identify five major challenges for incorporating water temperatures into environmental flow assessments, and describe future research opportunities and some alternative approaches for confronting those challenges.
5. We encourage ecologists and water managers to broaden their perspective on environmental flows to include both water quantity and quality with respect to restoring natural thermal regimes. We suggest that scientific research should focus on the comprehensive characterisation of seasonality and variability in stream temperatures, quantification of the temporal and spatial impacts of dam operations on thermal regimes and clearer elucidation of the relative roles of altered flow and temperature in shaping ecological patterns and processes in riverine ecosystems. Future investigations should also concentrate on using this acquired knowledge to identify the ‘manageable’ components of the thermal regime, and develop optimisation models that evaluate management trade-offs and provide a range of optimal environmental flows that meet both ecosystem and human needs for fresh water.

Full text

Incorporating thermal regimes into environmental flows
assessments: modifying dam operations to restore
freshwater ecosystem integrity
JULIAN D. OLDEN AND ROBERT J. NAIMAN
School of Aquatic and Fishery Sciences, University of Washington, Seattle, WA, U.S.A.
SUMMARY
1. Despite escalating conflict over fresh water, recent years have witnessed a growing
realisation that human society must modify its behaviour to ensure long-term ecological
vitality of riverine ecosystems. In response, ecologists have been increasingly asked to
guide instream flow management by providing ‘environmental flow’ prescriptions for
sustaining the ecological integrity of riverine systems.
2. Environmental flows are typically discussed in the context of water releases from dams
and water allocation for extraction (such as for urban use or irrigation), where there is
general agreement that rivers need to exhibit some resemblance of natural flow variability
necessary to support a functioning ecosystem. Although productive dialogue continues on
how best to define environmental flows, these discussions have been focused primarily on
water quantity without explicit consideration of many components of water quality,
including water temperature – a fundamental ecological variable.
3. Many human activities on the landscape have modified riverine thermal regimes. In
particular, many dams have modified thermal regimes by selectively releasing hypo-
limnetic (cold) or epilimnetic (warm) water from thermally stratified reservoirs to the
detriment of entire assemblages of native organisms. Despite the global scope of thermal
alteration by dams, the prevention or mitigation of thermal degradation has not entered
the conversation when environmental flows are discussed.
4. Here, we propose that a river’s thermal regime is a key, yet poorly acknowledged,
component of environmental flows. This study explores the concept of the natural thermal
regime, reviews how dam operations modify thermal regimes, and discusses the ecological
implications of thermal alteration for freshwater ecosystems. We identify five major
challenges for incorporating water temperatures into environmental flow assessments, and
describe future research opportunities and some alternative approaches for confronting
those challenges.
5. We encourage ecologists and water managers to broaden their perspective on
environmental flows to include both water quantity and quality with respect to restoring
natural thermal regimes. We suggest that scientific research should focus on the
comprehensive characterisation of seasonality and variability in stream temperatures,
quantification of the temporal and spatial impacts of dam operations on thermal regimes
and clearer elucidation of the relative roles of altered flow and temperature in shaping
ecological patterns and processes in riverine ecosystems. Future investigations should also
concentrate on using this acquired knowledge to identify the ‘manageable’ components
of the thermal regime, and develop optimisation models that evaluate management
Correspondence: Julian Olden, School of Aquatic and Fishery Sciences, Box 355020, University of Washington, Seattle, WA 98195,
U.S.A. E-mail: olden@u.washington.edu
Freshwater Biology (2010) 55, 86–107 doi:10.1111/j.1365-2427.2009.02179.x
86 2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd

trade-offs and provide a range of optimal environmental flows that meet both ecosystem
and human needs for fresh water.
Keywords: coldwater pollution, hypolimnetic release, water quality, water temperature
Introduction
The availability of fresh water to simultaneously meet
the demands of a growing human population and
ensure freshwater ecosystem integrity has emerged as
one of the world’s primary resource issues (Alcamo
et al., 2008). As world water demand has more than
tripled over the last half-century, signs of water
scarcity have now become commonplace. Given the
realities of human dependency on freshwater ecosys-
tems in the future, ecological thought has evolved
from a focus on humans as exploiters of riverine
ecosystem services to a world where humans and
freshwater systems must coexist to ensure long-term
ecological sustainability (Palmer et al., 2004). This
represents a formidable challenge in the coming
decades, which will only get more difficult with
projected growth in the human population, intensified
land use and a changing climate (Poff et al., 2003;
Palmer et al., 2008).
Achieving compatibility between human and natu-
ral ecosystem needs is the ultimate challenge of
ecologically sustainable water management (sensu
Richter et al., 2003), in which scientists are becoming
increasingly engaged. One of the most promising
approaches to integrating human uses into the larger
scope of ecological sustainability is the concept of
environmental flows, or the provision of water within
rivers to conserve freshwater biodiversity while
maintaining the water needs of human society (Acr-
eman & Dunbar, 2004). Environmental flows are
typically discussed in the context of water releases
from dams and catchment abstraction management
(i.e. water allocation strategies for urban use or
irrigation), where there is general agreement that they
need to exhibit patterns of natural variability to
support a functioning riverine ecosystem (Dunbar,
Acreman & Kirk, 2004; Richter & Thomas, 2007).
Recent dialogue has emerged among scientists on
how best to define and prescribe environmental flows
in a managerial context (e.g. Acreman & Dunbar,
2004; Arthington et al., 2006; Richter et al., 2006; Poff
et al., 2010), yet these discussions have focused
primarily on water quantity without the explicit
consideration of water quality, such as water temper-
ature, pollutants, nutrients, organic matter and sedi-
ments and dissolved oxygen.
Hydrologic alteration is a major consequence of
river regulation associated with significant impacts on
aquatic biodiversity (Bunn & Arthington, 2002); how-
ever, dams and diversions have also greatly modified
riverine thermal regimes depending on their mode of
operation and specific mechanism and depth of water
release (Ward, 1985). For example, a substantial
number of large dams throughout the world have
intentionally managed thermal regimes by selectively
releasing cold water from deep reservoirs to establish
highly desirable fishing opportunities for trout,
salmon or walleye. Similarly, fluctuating releases by
hydroelectric dams from beneath the reservoir’s ther-
mocline can result in highly variable and frequently
depressed summer water temperatures. Smaller dams
and diversions, in contrast, can cause increases in
downstream temperatures by releasing warm water
directly from the reservoir surface. Dam-induced
modifications to a river’s thermal regime (also termed
thermal pollution) can have both direct and indirect
consequences for freshwater ecosystems, yet it has
been relatively unappreciated in discussions of in-
stream flow management.
While the ecological significance of water temper-
ature in riverine ecosystems is widely acknowledged
(reviewed in Magnuson, Crowder & Medvick, 1979;
Poole & Berman, 2001; Caissie, 2006), the prevention
or mitigation of thermal pollution below dams has
received little attention when environmental flow
assessments are conducted. This is despite the highly
recognised impact of dams on thermal regimes and
the inherent relationship between discharge, temper-
ature and a host of other water quality variables
(Nilsson & Reno
¨fa
¨lt, 2008). Clearly, both discharge
and temperature must be simultaneously considered
for the successful implementation of environmental
flow management below dams. We perceive this as a
Incorporating thermal regimes in environmental flows assessments 87
2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd, Freshwater Biology,55, 86–107

major gap in environmental flow assessments that
needs to be both elaborated upon and addressed in
the future. Our article aims to narrow this knowl-
edge gap by exploring the concept of the natural
thermal regime, reviewing how river regulation by
dams modify thermal regimes, and discussing the
ecological implications of thermal alteration for
freshwater ecosystems. We focus on the thermal
effects of dams, although we readily acknowledge
that many other human activities such as the direct
extraction of water for consumption, sanitation and
irrigation influence riverine thermal regimes and are
highly relevant in environmental flow assessments.
Next, we discuss five major challenges for incorpo-
rating water temperatures into environmental flow
assessments and highlight research opportunities
and some alternative approaches for confronting
those challenges. The perspectives developed in this
study are that by viewing environmental flows in a
holistic manner, in which both water quantity and
quality are considered in flow recommendations, we
will increase the chances of long-term success in
achieving ecologically sustainable water manage-
ment.
The natural thermal regime and its importance
for riverine ecosystems
Stream temperature depends on the amount of heat
energy added or lost to the channel (i.e. energy
budget) and the volume of water to be heated or
cooled (i.e. thermal capacity) (Poole & Berman, 2001;
Caissie, 2006). Heat energy is exchanged both at the
air⁄surface water interface and at the streambed⁄water
interface through a number of physical processes,
which are mediated by a multitude of external drivers
that control heat and water delivery to the stream
(Fig. 1). Heat flux at the air⁄surface water interface
occurs primarily through solar or short-wave radia-
tion, long-wave radiation, evaporative processes and
convective transfer resulting from temperature differ-
ences between the stream and the atmosphere. Heat
exchange at the streambed⁄water interface is mainly a
function of geothermal heating through conduction
and advective heat transfer through groundwater
inputs and hyporheic exchange. The rates of these
processes are controlled by a number of factors
related to atmospheric conditions, topography,
streambed and stream discharge (Webb, 1996; Caissie,
2006).
Water temperatures show marked annual and
diurnal fluctuations in response to seasonal and daily
rhythms in the amount and type of heat energy
gained and lost by a stream and the volume and
source of runoff contributing to discharge (Ward,
1985; Webb, 1996). A stream’s thermal regime
describes the distribution of the magnitude of water
temperatures, the frequency with which a given
temperature occurs, the time of the day or year when
a certain temperature occurs, and the duration of time
Fig. 1 Heat exchange processes responsi-
ble for variability in water temperatures
(bold font) and the physical drivers that
control the rate of heat and water delivery
to stream and river ecosystems (italic
font).
88 J. D. Olden and R. J. Naiman
2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd, Freshwater Biology,55, 86–107

for which a stream is above or below a given
temperature. We propose that the natural thermal
regime can be discomposed into its components of
magnitude, frequency, duration, timing and rate of
change. Therefore, much like the natural flow regime
(Poff et al., 1997) and the quantification of its charac-
teristic properties (Olden & Poff, 2003), thermal
regimes can be summarised using statistics that
describe the central tendency and variability in
distributions of temperature conditions (Fig. 2a). For
example, Jackson, Gibbins & Soulsby (2007) calculated
a suite of metrics describing what they deemed as
critical attributes of the thermal regime for inverte-
brate communities in the Lyon River (Scotland). Such
analyses could also include the characterisation of
thermal regimes at multiple temporal scales (Steel &
Lange, 2007). In addition, the physiological tolerances
of freshwater species to particular temperatures (see
below for further discussion) and the influence of
temperature on other aspects of water quality, such as
solute and pollutant fluxes, nutrient concentrations,
organic matter and sediments and dissolved oxygen
(Webb, 1996; Caissie, 2006), allow for the quantifica-
tion of more targeted metrics describing thermal
characteristics of streams and rivers. For example,
development schedules for freshwater fish and insects
respond to the summation of thermal units (i.e. the
accumulation of daily temperatures above some
threshold) as well as absolute temperatures, and fish
species have both chronic and acute temperature
(a)
(b)
Fig. 2 (a) A stream’s thermal regime de-
scribes the magnitude, frequency, dura-
tion, timing and variability in water
temperatures at different spatial and
temporal scales. Shown are a number of
quantitative metrics that describe different
components of the thermal regime,
including high thermal events (HTE) and
low thermal events (LTE) based on a
defined upper and lower threshold,
respectively. The threshold could be
ecologically derived (e.g. upper and lower
lethal temperature, threshold temperature
for cueing spawning or positive growth)
or based on the statistical distribution of
annual temperatures (e.g. 90th and 10th
percentile). Other metrics not illustrated
here could be appropriate for describing
thermal regimes for various objectives.
(b) A stream’s natural thermal regime
influences freshwater biodiversity via
multiple mechanisms that operate at
different spatial and temporal scales.
Shown are a number of examples in which
temperature influences the bioenergetic
and life-histories of fish (italic font),
insects (normal font) and riparian plants
(underlined font). Overall, the native
biotas of riverine ecosystems have
evolved in response to the natural thermal
regime.
Incorporating thermal regimes in environmental flows assessments 89
2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd, Freshwater Biology,55, 86–107

thresholds for survival, growth and reproduction
(Vannote & Sweeney, 1980; Coutant, 1987). This
provides the opportunity to derive synthetic metrics
that are ecologically relevant for the species or
community of interest.
The ecological integrity of lotic ecosystems depends
on the natural dynamics of the thermal regime. Water
temperature directly influences the metabolic rates,
physiology and life-history traits of aquatic species
and helps determine rates of important ecological
processes such as nutrient cycling and productivity,
and indirectly mediates biotic interactions (Magnuson
et al., 1979; Petts, 1986; Poole & Berman, 2001; Caissie,
2006; Webb et al., 2008). As ectotherms, freshwater
fishes and insects use a diverse array of thermal
habitats to meet their specific temperature require-
ments for survival, growth and reproduction (Cou-
tant, 1987; Vinson & Hawkins, 1998). Consequently,
specific components of the thermal regime have
specific ecological relevance for freshwater organisms
throughout their life history. We highlight a few
examples in Fig. 2b.
Water temperatures directly influence growth rates
of aquatic organisms and interacts with the lower and
upper incipient lethal temperatures to shape species
distributions. The availability of thermal heterogene-
ity also provides ecothermic organisms the opportu-
nity to select habitats that optimise their energy
intake relative to physiological costs and thus affect-
ing growth and survival. For example, species of
salmon (Oncorhynchus spp.) thermoregulate behavio-
urally by occupying cooler areas, such as seeps and
confluences with cold streams, when surrounding
temperatures exceed their upper tolerances (e.g.
Berman & Quinn, 1991). Similarly, temperate fishes
actively seek warm thermal refugia provided by
small streams and springs to escape cold waters and
take advantage of warmer conditions needed for
optimal feeding efficiency and growth (e.g. Peterson
& Rabeni, 1996). Development and life-history events
of locally adapted populations of aquatic insects and
fishes are closely timed to prevailing thermal regimes
(Vannote & Sweeney, 1980; Vinson & Hawkins, 1998;
Huryn & Wallace, 2000). For example, the natural
temperature regime of a river provides thermal cues
that stimulate fish migration, spawning and egg
hatching, and directly influences egg survivorship
and developmental time (Wootton, 1990; Fig. 2b).
Changes in temperature can also influence the
type and prevalence of diseases affecting fish. Lastly,
the accumulation of daily maximum temperatures
above a critical threshold has been shown to be a
fundamental parameter in shaping the distribution
and condition of many aquatic species (Armour,
1991).
Dam-induced modifications to riverine thermal
regimes
Spatial and temporal patterns in stream temperatures
are influenced by modifications to the energy budget
and⁄or the thermal capacity of the fluvial ecosystem
(Poole & Berman, 2001). Any natural process or
human activity that alters the external drivers of heat
load or stream discharge in a system will influence
spatiotemporal patterns of water temperature
(Fig. 1). For example, the discharge of heated indus-
trial effluents, such as cooling water from power
generating stations, can result in unnaturally elevated
water temperatures (Langford, 1990). A reduction in
stream discharge resulting from water extraction for
human consumption or irrigation practices can also
directly affect water temperatures by decreasing
thermal capacity and thus increasing the likelihood
of high temperature events (Sinokrot & Gulliver,
2000). In an indirect manner, stream temperatures
may be modified through changes in catchment land
use and channelisation, both of which alter the
energy budget and thermal capacity of the water
course (Moore, Spittlehouse & Story, 2005; Nelson &
Palmer, 2007).
Here, we focus on dams and diversions that have
both direct and indirect effects on riverine thermal
regimes. In general, the extent to which a dam affects
downstream thermal regimes depends on their mode
of operation and specific mechanism of water release.
Dams directly modify a river’s thermal regimes by
releasing water that differs greatly in temperature to
that occurring naturally in the river. The magnitude of
thermal alteration depends largely on stratification
behaviour of the reservoir (i.e. depth profile of the
thermocline), and the depth at which water is released
from the dam. Dams also modify water temperatures
indirectly by influencing processes controlling the
delivery, distribution and retention of heat within the
river channel. Changes to discharge and the volume
of water in a river, for example, affect the rate at
which water heats and cools in response to natural
90 J. D. Olden and R. J. Naiman
2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd, Freshwater Biology,55, 86–107

diurnal heat exchange both at the air⁄surface water
and streambed⁄water interface (Petts, 1986; Palmer &
O’Keeffe, 1989; Poole & Berman, 2001).
Many large dams throughout the world cause
downstream modifications to riverine thermal re-
gimes by altering flow regimes and releasing cold
water from below the thermocline of the reservoir (i.e.
the hypolimnetic layer). Coldwater thermal regimes
are established by releasing water from a single deep
portal (often associated with hydroelectric generation)
or selectively withdrawing water from different res-
ervoir depths. In many instances, hypolimnetic
releases of cold water provide unique and highly
desirable fishing opportunities for trout or salmon in
geographic regions that could otherwise not support
coldwater species (e.g. Brooker, 1981; Horne, Ruther-
ford & Wehrly, 2004; Krause, Newcomb & Orth, 2005).
Although dam operations provide optimal coldwater
conditions for tailwater fisheries, this is often at the
detriment of entire warmwater assemblages of native
fishes and other aquatic organisms (see discussion
below). Consequently, many ecologists consider this a
form of thermal depression or coldwater pollution
because water temperatures are typically much lower
during the spring and summer months compared to
free-flowing rivers (although during the winter
months, dam operations typically elevate water tem-
peratures in relation to natural conditions). Many
other dams (typically smaller in size and located in
cooler summer climates) release water from above the
thermocline of the reservoir (i.e. the epilimnetic layer)
resulting in elevated spring–summer water tempera-
tures (e.g. Lessard & Hayes, 2003).
The thermal impacts associated with dam opera-
tions in riverine ecosystems are well recognised
(Ward & Stanford, 1982; Ward, 1985; Petts, 1986;
Webb & Walling, 1996); although they are examined
much less often than river hydrology. A reservoir acts
as a buffer that reduces both the annual amplitude of,
and daily fluctuations in, downstream temperatures
because the reservoir’s mass warms and cools more
slowly than the free-flowing river. Thermal effects of
river regulation by dams vary depending on the
landscape position of the dam, mode of dam opera-
tion, release depth and environmental and geomor-
phologic setting. Despite these many factors, an
examination of rivers regulated by large, hypolimnet-
ic-release dams from different regions of the world
reveals a relatively consistent pattern (Fig. 3). In
general terms, reservoirs moderate downstream ther-
mal regimes where temperatures are lower in the
spring and summer months, higher in the winter
months, fluctuate less seasonally, and exhibit delayed
timing of maxima compared to natural conditions.
Finer-scale investigations also show that water tem-
peratures downstream from dams often have muted
diurnal variation because of the stable temperatures of
hypolimnetic waters (Lowney, 2000; Steel & Lange,
2007).
There are many similar examples to that presented
in Fig. 3 that could be cited (see references listed
above). In Australia, river regulation by hypolimnetic
dams has caused the annual thermal maxima of
numerous systems to be both reduced and displaced
in time; for example, downstream of Burrendong Dam
on the Macquarie River (8–12 C depression and 1–
3 month delay) and downstream of Dartmouth Dam
on the Mitta Mitta River (8–10 C depression) (re-
viewed in Lugg, 1999; Ryan et al., 2001; Preece, 2004).
Jackson et al. (2007) found that summer water tem-
peratures in the regulated Lyon River were 5–6 C
cooler than the adjacent unregulated Lochay River,
Scotland. Similarly, Angilletta et al. (2008) reported a
decrease in summer water temperature from 17.5 to
14.5 C after the construction of Hills Creek Dam on
the Willamette River, Oregon (U.S.A.). These exam-
ples highlight a general pattern toward summer
cooling and winter warming resulting from hypolim-
netic release operation below large dams. Not sur-
prisingly, a contrasting pattern emerges when we
examine epilimnetic-release dams. Lessard & Hayes
(2003) examined 10 small dams on Michigan streams
(U.S.A.) and found that summer temperature below
dams increased an average of 2.7 C, ranging from a
1.0 C cooling to a 5.5 C warming.
Scientists have also provided a wealth of informa-
tion concerning the extent to which the thermal effects
persist downstream beyond the immediate vicinity of
the dam. Thermal impacts can encompass relatively
short or extremely long distances below their respec-
tive dams depending on heat exchange with the
atmosphere, hydrologic inputs from tributaries and
groundwater recharge, and dam discharge (Palmer &
O’Keeffe, 1989). In the Murray-Darling River Basin
(Australia), depressed summer temperatures extend
up to several hundred kilometres downstream of
dams on a number of major rivers, including Mur-
rumbidgee, Marquarie, Mitta Mitta, Namoi and
Incorporating thermal regimes in environmental flows assessments 91
2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd, Freshwater Biology,55, 86–107

Murray (Ryan et al., 2001; Preece, 2004; Todd et al.,
2005). The thermal recovery of the River Svratka,
Czech Republic, required over 40 km (Ward, 1985),
and using rates of maximum summer warming,
Stevens, Shannon & Blinn (1997) estimated that a
mainstem distance of 930 km would be required for
water temperatures to fully recover (increase to pre-
dam conditions) below Glen Canyon Dam in the
Colorado River, U.S.A. (a distance that is currently
prevented by other downstream dams). In coolwater
systems affected by epilimnetic-release dams, streams
may not be able to shed added heat during the
summer and downstream water temperatures may
continue to warm due to normal stream processes.
Fraley (1979) found significant summer temperature
increases in the Madison River, Montana (U.S.A.) that
never returned to upstream temperatures even 56 km
downstream of a surface-release dam, although
Fig. 3 Annual thermographs from regulated rivers across the world illustrating summer cooling and winter warming caused by
hypolimnetic-release dams. Mean monthly temperatures are presented for the unregulated record (filled symbols and solid line)
and the regulated record (open symbols and dashed line). The unregulated record includes data recorded either in pre-dam years (i.e.
prior to construction), immediately upstream from the dam, or from an unregulated tributary, and the regulated record includes data
recorded in post-dam years. Details regarding the temperature data are as follows. Flaming Gorge Dam: unregulated (1958–62),
regulated (1963–77), data source (US Geological Survey); Kielder Dam: unregulated (1993–95), regulated (1993–95, River Rede),
data source (David Archer); Vilui Dam: unregulated (1950–65), regulated (1966–97), data source (Daqing Yang); Gathright Dam:
unregulated (1991–2006, upstream from dam), regulated (1991–2006, downstream from dam), data source (US Geological Survey);
Xinanjiang Dam: unregulated (1957–59), regulated (1960–68), data source (Zhong & Power, 1996); Keepit Dam: unregulated
(1976–2003, upstream from dam), regulated (1976–2003), data source (Australian Department of Natural Resources).
92 J. D. Olden and R. J. Naiman
2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd, Freshwater Biology,55, 86–107

diurnal temperature fluctuations did recover with
increasing distance.
Ecological consequences of modified thermal
regimes
Extensive dam construction throughout the world has
raised long overdue concerns regarding the potential
impacts of altered thermal regimes on biodiversity
and ecological processes. Empirical evidence suggests
that dam-induced thermal alteration by dams has
significant implications for stream productivity and
the reproduction, growth, distribution and assem-
blage structure of organisms (Haxton & Findlay,
2008). Despite this, our current level of understanding
continues to be much less than our knowledge of the
ecological implications of altered hydrology (Murchie
et al., 2008). Below we illustrate the ecological conse-
quences of modified thermal regimes through a series
of case studies.
In Australia, the ecological consequences of cold-
water pollution have been more widely recognised
compared to other countries. A number of empirical
and modelling studies have explored the relationship
between dam-induced thermal changes and fish pop-
ulations in rivers. Cooling and delayed timing of
maximum temperatures in the Namoi River below
Keepit Dam had significant consequences for the
spawning success of several native fish species (Preece
& Jones, 2002). Based on the percentage of time in the
spawning period that the mean daily water tempera-
ture exceeded the temperature threshold for spawning,
spawning opportunities for silver perch (Bidyanus
bidyanus Mitchell, 1838) and golden perch (Macquaria
ambigua Richardson, 1845) were reduced to 25–70%
and 44–87%of pre-dam years, respectively. Using a
stochastic population model, Todd et al. (2005) ex-
plored the impact of altered thermal regimes on the
population viability of Murray cod (Maccullochella peelii
peelii Mitchell, 1838) in the Mitta Mitta River down-
stream of the fourth largest dam in Australia, Dart-
mouth Dam. Model predictions showed that cold
water releases significantly threaten the post-spawning
survival of Murray cod by reducing the average
minimum female population size by 76%and increas-
ing population variability by 137%.
In the Qiantang River Basin of China, the construc-
tion of Xinanjiang Dam caused mean temperatures to
decrease from 19.0 to 13.5 C, and the annual sum of
degree days >15 C (considered the positive-growth
threshold temperature for warmwater fish species)
dropped 37%(Zhong & Power, 1996). As a result, the
majority of warmwater fishes were extirpated from
large sections of river below the dam. In the Hanjiang
River below the Danjiangkou Dam, China, the post-
poned timing of spring–summer warming caused by
dam operations resulted in fish spawning times that
were delayed by 20–30 days. Late hatching and lower
water temperatures reduced the first year growth
compared to an unregulated river (Zhou, Liang &
Huang, 1980). Similarly, significant declines of native
fishes in the Colorado River Basin (U.S.A.) have been
attributed, in part, to the depression of spring–
summer tailwater temperatures caused by hypolim-
nial-release dams. For example, the operation of
Flaming Gorge Dam on the Green River, lowered
spring–summer tailwater temperatures to nearly 6 C
from a previous range of 7–21 C, which contributed
to the extirpation of several native species including
the endangered Colorado pikeminnow (Ptychicheilus
lucius Girard, 1856), humpback chub (Gila cypha
Miller, 1946) and bonytail chub (G. elegans Baird &
Girard, 1853) (Clarkson & Childs, 2000). Similarly,
Olden (2004) found that depressed summer temper-
atures below Gathright Dam on the Jackson River,
Virginia (U.S.A.), resulted in complete species turn-
over from a warmwater fish assemblage to a river
dominated by cool- and coldwater fish species.
Modified temperature regimes also strongly influ-
ence invertebrate communities by eliminating key
developmental cues and influencing the rate of egg
development and juvenile growth. Consequently,
changes in temperature can cause the asynchrony of
the life cycle to seasonal patterns in resource and mate
availability (Vannote & Sweeney, 1980). Individual
species responses can generate significant changes in
assemblage structure by shifting the composition from
warm stenothermic species to cold-tolerant eury-
therms (Ward, 1974). For example, when a hypolim-
netic dam was constructed on the Saskatchewan River
(Canada), the cue to end egg diapause was eliminated
because winter temperatures were maintained near
4C and the insect biota no longer experienced a
prolonged period near freezing followed by a rapid
temperature increase (Lehmkuhl, 1974). This resulted
in the complete loss of an insect fauna, switching from
a river containing 12 orders, 30 families and 75 species
to one that comprised only the midge family Chiro-
Incorporating thermal regimes in environmental flows assessments 93
2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd, Freshwater Biology,55, 86–107

nomidae. Another potential consequence of warmer
winter temperatures is an increase in the growth of
aquatic insects, resulting in winter rather than spring
or early summer emergence (Ward & Stanford, 1982).
Emergence during the winter is either lethal or
impedes mating of aquatic insects. In another exam-
ple, Stevens et al. (1997) found that the macroinverte-
brate fauna of the Colorado River (U.S.A.) below Glen
Canyon Dam was highly depauperate compared with
other unregulated rivers of the basin. They collected
virtually no species from the Ephemeroptera (mayfly),
Plecoptera (stonefly) or Trichoptera (caddisfly) fami-
lies in the mainstream in 1991, suggesting that cold-
stenothermic releases may not permit these taxa to
complete their life cycle. Voelz & Ward (1991)
observed increasing insect species richness with dis-
tance downstream from Granby reservoir dam (Col-
orado River, U.S.A.) in response to a gradual recovery
(increasing) in maximum summer stream tempera-
tures. Additional examples of the effects of hypolim-
netic release on macroinvertebrate communities are
discussed by Haxton & Findlay (2008).
Dam-induced changes in thermal regimes may also
have long-term evolutionary consequences for river-
ine biota by inducing a mismatch between a species’
life-history and other critical environmental condi-
tions. For example, Angilletta et al. (2008) hypothes-
ised that warmer temperatures during the autumn
and winter below Lost Creek Dam (Rogue River,
U.S.A.) may indirectly influence the fitness of Chinook
salmon (Oncorhynchus tshawytscha Walbaum, 1792) by
accelerating the development of embryos, leading to
earlier timing of emergence. Shifts to earlier emer-
gence could lead to mortality from high flow events,
elevated predation or insufficient resources. Using an
age-based population model the authors predicted a
decrease in mean fitness of Chinook salmon after dam
construction. Although the likelihood for these im-
pacts is unknown (i.e. temperature changes may also
result in strong compensatory strategies, such as
delayed spawning by adults or slowed development
by embryos), the potential evolutionary consequences
of thermal alteration should not be overlooked.
Challenges and prospects for incorporating ther-
mal criteria into environmental flow assessments
We propose that a river’s thermal regime is a key, yet
poorly acknowledged, component of environmental
flows. The efficacy of environmental flows to advance
ecologically sustainable water management is predi-
cated upon a greater appreciation of the influence of
human activities on riverine thermal regimes and
other critical water quality variables, and requires
new research and tools that help define management
options that maintain key hydrologic and thermal
processes for the benefit of aquatic ecosystems. Below,
we discuss five overarching challenges for incorpo-
rating thermal criteria into environmental flow assess-
ments, and lay out some alternative approaches for
confronting those challenges. Some challenges are
deceptively simple, while others will require new
avenues of research. In the sections that follow we use
the Flaming Gorge Dam – a 91 m dam constructed in
1962 on Green River, U.S.A. (the largest tributary of
the Colorado River) for hydroelectricity generation
and flow control – as a working example. Although
our discussion, in part, focuses on the role of river
regulation by dams and diversions, we believe that
many of the challenges are equally pertinent for our
ability to incorporate water temperature targets into
environmental flows standards for water abstraction
practices.
Challenge #1: Advance our understanding of
dam-induced impacts to riverine thermal regimes
Continuing research has emphasised the complexity
of thermal responses to dams as a result of processes
occurring within reservoirs and downstream. Signif-
icant inter-annual variation in the impact of dam
operations on downstream temperatures has been
illustrated and attributed to year-to-year changes in
climate conditions, reservoir operation and behaviour,
and tributary inflows (Webb & Walling, 1993, 1996;
Preece & Jones, 2002; Todd et al., 2005). We high-
lighted a number of river systems around the world
with qualitatively very different annual thermographs
between pre- and post-dam time periods (Fig. 3).
Similarly, the majority of published studies reporting
the thermal impacts effects of dams have focused
almost exclusively on simple comparisons of annual
or monthly temperatures before and after dam con-
struction, often without supporting statistical analy-
ses. While such comparisons are informative for
descriptive purposes, we argue that ecologists must
better formalise how dams are altering the various
components of the thermal regimes, including the
94 J. D. Olden and R. J. Naiman
2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd, Freshwater Biology,55, 86–107

magnitude, frequency, duration, timing and rate of
change in temperature events. Each component of the
thermal regime has distinct ecological relevance for
freshwater organisms, communities and riverine pro-
cesses (Fig. 2b), which should ultimately guide the
appropriate characterisation of the thermal regime
and the selection of relevant descriptors (Fig. 2a).
Characterizing thermal regimes and assessing the
downstream impacts of dams will be challenging for
many of the same reasons that assessing hydrologic
variability and alteration continues to be difficult –
perhaps even more so. In the United States, water
temperature data are collected by a variety of federal,
state and local agencies, although the primary source
for long-term daily records is the U.S. Geological
Survey (also the main source for daily discharge data).
As of June 2008, the U.S. Geological Survey has
collected data from 24 856 gauges across the country,
yet only 7.4%of these gauges contain records of daily
water temperature for at least 1 year, and less than
1%of all gauges contain temperature data for at least
15 years. Moreover, most temperature records begin
after the construction of a dam or diversion. Given the
paucity and appropriateness of long-term water tem-
perature data (a problem common to most other
regions of the world), investigators may be forced to
rely more heavily on temperature models based on
unregulated systems to predict riverine thermal con-
ditions. These will include empirical models that use
statistical analyses (e.g. regression, neural networks)
to make temperature predictions from meteorological
data or catchment characteristics, physically based
modelling that involves solving the heat budget
equation and hydrodynamic models relating dis-
charge to water temperatures (Gu, McCutcheon &
Chen, 1999; Caissie, 2006; Webb et al., 2008). Scientists
will also benefit greatly from recent technological
developments that have facilitated the measurement
and monitoring of water temperature using program-
mable (and remotely downloadable) digital loggers
and remotely sensed thermal infrared imagery (Webb
et al., 2008), in addition to the creation of central
depositories or information systems containing water
temperature data.
In some cases, long-term daily water temperatures
will be available (or modelled) to quantify the impact
of river regulation on thermal regimes. As an exam-
ple, we quantitatively compared the thermal regimes
of the Green River (U.S.A.) before and after the
construction of Flaming Gorge Dam with respect to
metrics describing the magnitude, frequency, dura-
tion and timing of water temperatures. These thermal
metrics were selected a priori based on previous
empirical demonstrations of their ecological impor-
tance for species occurrence and community structure
in the Green River, or more broadly the Upper
Colorado River Basin (Clarkson & Childs, 2000;
Vinson, 2001). Flaming Gorge Dam uses a hypolim-
netic release schedule that has increased predictability
and substantially reduced annual variability in water
temperatures (Fig. 4). Average late spring and sum-
mer temperatures (May–August) are 66%lower in
post-dam years decreasing from 17.2 to 5.7 C,
whereas winter temperatures (December–March)
showed a marked increase of 802%from 0.7 to
5.4 C (Fig. 4). Magnitude of 1-, 7- and 30-day min-
imum temperatures increased from 0–0.2 to 3.4–3.9 C
after dam construction, whereas the magnitude of
maximum temperatures for the same duration de-
creased 56%from 21.1–24.0 C (pre-dam) to 9.6–
10.0 C (post-dam) (Fig. 4). Dam operations have
resulted in significantly lower frequency and duration
of low and high temperature events (Fig. 4), and
shifted the timing of average minimum temperatures
forward 3 weeks (12th February–4th March) and
average maximum temperatures ahead over 15 weeks
from the summer (24th July) to the autumn (7th
November) (Fig. 4). We recommend that future
research should focus on the quantitative assessment
of thermal alteration associated with dam operations,
including the statistical analysis of pre- and post-dam
time periods to better elucidate how different com-
ponents of the riverine thermal regime are modified
by dam operations and recovery downstream.
Without detailed and long-term temperature data
(either recorded or modelled), it is unlikely that
comprehensive evaluations of dam-induced thermal
alteration at the landscape scale will be possible.
However, in lieu of these data it is possible to
qualitatively forecast the thermal impacts of river
regulation according to dam⁄reservoir characteristics
(e.g. dam height, reservoir depth and bathymetry),
operation schedule (e.g. intake depth, discharge), and
geomorphic attributes of the receiving river system
(channel form, groundwater influences, position and
characteristics of tributaries). In Australia, the states of
Victoria, Queensland and New South Wales have
completed comprehensive reviews of large dams to
Incorporating thermal regimes in environmental flows assessments 95
2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd, Freshwater Biology,55, 86–107

identify and rank structures based on potential to
cause coldwater pollution. Ryan et al. (2001) ranked
dams in Victoria based on depth and frequency
(regular versus occasional) of release, and identified
24 top priority dams for which further assessment and
continuous temperature monitoring was recom-
mended. The New South Wales review used intake
depth and summer discharge to develop a short list of
nine dams (out of 93 in total) predicted to cause severe
coldwater pollution (Preece, 2004). With the exception
of Australia, the potential for coldwater pollution
associated with dams has not been systematically
assessed. These so-called ‘desktop assessments’ may
be particularly valuable for identifying those rivers
where thermal pollution may be prevalent, thus
helping guide future temperature surveillance efforts
and prioritizing the possible retrofitting of current
gauges to record water temperatures. Moreover, in
combination with additional modelling, this informa-
tion would enable the mapping of spatial patterns in
dam-induced thermal alteration across entire riverine
landscapes.
Challenge #2: Advance our understanding of the
ecological consequences of altered thermal regimes
Hydro-ecological studies are becoming increasingly
interdisciplinary, with research focusing on, among
other things, the relationships between water quantity
and quality, and their importance for riverine biota.
However, despite broad recognition of the importance
of flow and thermal regimes for river systems, there
are relatively few studies that have explicitly linked
dam-induced changes in both flow and thermal factors
to ecological structure and function. In a revealing
study, Murchie et al. (2008) conducted a systematic
review of the literature to identify studies that
examined the response of fish to modified flow
Fig. 4 Hypolimnetic release operations of
Flaming Gorge Dam have significantly
altered the thermal regime of the Green
River (U.S.A.). Comparisons of thermal
metrics computed for the pre-dam (1958–
62, circles) and post-dam records (1963–
77, squares) show considerable effects on
the thermal regime, including variability
and predictability of annual temperatures,
monthly temperatures, magnitude, fre-
quency and duration and timing of ther-
mal events. Low and high pulses (and
durations) are based on mean daily tem-
peratures below the 25th percentile and
above the 75th percentile of unregulated
temperatures, respectively. Bars represent
±1 SD and ‘CV’ refers to the coefficient of
variation.
96 J. D. Olden and R. J. Naiman
2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd, Freshwater Biology,55, 86–107

regimes in regulated rivers. Of the 131 studies
identified in their review, they found that although
almost half of the studies included the collection of
water temperature, the majority of these studies (57%)
failed to examine the potential consequences of
thermal alteration. Moreover, an extremely small
number of studies included temperature data in
formal statistical analysis to explore the potential
individual and interactive effects of flow and thermal
modification on fish assemblages. We cannot empha-
sise enough the need for this research for supporting a
truly holistic approach to environmental flow assess-
ments that incorporates both flow and temperature
targets.
The limited literature examining both thermal and
flow effects on freshwater organisms suggests that
both the direct and interactive influence of these
factors are critical. In the Jackson River (U.S.A.) below
Gathright Dam, Olden (2004) found that no one single
component of the flow or thermal regime explained
fish population- and community-level responses to
river regulation. Overall, species showed patterns of
numerical recovery below the dam that were largely
congruent with their ecological requirements; for
example, fluvial-specialist species responded posi-
tively to decreasing flow alteration and warmwater
species responded positively to decreasing thermal
alteration. Jackson et al. (2007) explored the relative
roles of discharge and temperature variation in
determining invertebrate community structure in a
regulated river in Scotland. The authors found that no
one flow or thermal metric exerted an overriding
influence on community structure suggesting that
whole regime changes, rather than alterations to one
particular aspect of the regime, were responsible for
the impoverishment of invertebrate communities
observed below the dam.
Although some additional examples do exist, in
general it remains extremely difficult to ascertain the
relative contributions of flow versus temperature (and
a number of other water quality variables influenced
by dams and human activities) to observed biotic
responses in riverine ecosystems (Bednarek & Hart,
2005). This is supported by Haxton & Findlay (2008)
who, after conducting a comprehensive review of the
literature, lamented the paucity of studies that quan-
tified the effects of particular drivers of dam-induced
change on downstream species responses. We view
this as a significant challenge that currently precludes
our ability to formally incorporate thermal criteria
into environmental flow assessments. Confronting
this challenge may be facilitated by the simple fact
that levels of hydrologic and thermal alteration and
rates of downstream recovery vary substantially
within and among regulated rivers. By synthesizing
and collecting data along these spatial gradients of
alteration, greater insight into the relative roles of
hydrologic and thermal factors in shaping stream
communities will be possible.
Challenge #3: Demonstrate the availability and success
of temperature management strategies
A number of management options exist for mitigating
the thermal impacts of dams. The indirect manage-
ment of water temperatures is possible by targeting
those drivers of thermal regimes that can be directly
manipulated, such as the appropriate management of
riparian zones and the maintenance of natural flow
variability (Poole & Berman, 2001; Fig. 1). The man-
agement of riverine landscapes for thermal integrity
will require a broad perspective that recognises the
heterogeneous nature in which the topology of the
drainage network controls the physical processes
shaping spatial and temporal variability in stream
temperatures. For example, the ecological importance
of tributaries for promoting physical heterogeneity is
well recognised (e.g. Rice, Ferguson & Hoey, 2006),
and we believe that understanding the dynamics of
confluence zones is an integral part of understanding
the role of tributaries in mediating the downstream
impacts of dams on thermal regimes (in addition to
hydrologic regimes and sediment processes). The
degree to which unregulated tributaries lessen the
thermal influences of dams will depend on the size of
the stream and its distance from the dam, and
characteristics of the tributary with respect to dis-
charge, sediment and water temperature (Petts, 1986).
The influence of tributaries on mainstem thermal
and biological recovery has been illustrated in a
number of studies (e.g. Ward, 1985; Stevens et al.,
1997; Vinson, 2001). Preece & Jones (2002) showed that
the pattern of thermal recovery downstream from
Keepit Dam on the Namoi River (Australia) was
significantly influenced by the contribution of two
unregulated tributaries; the magnitude of this effect
was greater in the spring and early summer when
tributary discharge was 50%of that from Keepit
Incorporating thermal regimes in environmental flows assessments 97
2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd, Freshwater Biology,55, 86–107

Dam, compared to the summer when the relative
contribution of the tributaries was small (<10%of
Keepit Dam). Sato et al. (2005) found that the repro-
ductive success of an important fisheries species,
Prochilodus argenteus Spix & Agassiz, 1829 in the cold
tailwaters below the Tres Marias Dam on the Sao
Francisco River, Brazil, was achievable only below the
confluence with a medium-sized, unregulated tribu-
tary. Vinson (2001) recorded greater insect richness
downstream from a tributary confluence on the Green
River below Flaming Gorge Dam, and Olden (2004)
found elevated richness of warmwater fish species in
the Jackson River below Gathright Dam in reaches
immediately downstream from incoming warmwater
tributaries. Understanding the role in which tributar-
ies contribute to achieving prescribed thermal bench-
marks in environmental flow assessments is a key
challenge that requires additional exploration.
In highly regulated systems, the management of
mainstem channel vegetation and tributary contribu-
tions may be insufficient to meet thermal regime
standards. In such cases, our ability to incorporate
temperature requirements in environmental flows
assessments will require suitable technologies for
directly modifying water temperature released from
dams. Current management options for mitigating
thermal impacts from dams divided into two general
categories: exploit the temperature stratification of
the reservoir by selective withdrawal of water of
the desired temperature, or artificially break up the
stratification prior to discharging water from the dam.
Sherman (2000) provided a review of seven general
strategies for mitigating thermal pollution; we briefly
summarise these below:
1Multi-level intake structures make use of the
stratification within the reservoir by permitting water
with desirable thermal attributes to be withdrawn
from defined regions within the water column
(Fig. 5a). We discuss the details of this method in
subsequent paragraphs.
2Trunnions (floating intakes) are a variation on the
theme of selective withdrawal that involves the intake
of epilimnetic and hypolimnetic water through two
pipes hinged at the dam wall (Fig. 5b).
3A variety of destratification approaches exist that
induce mixing of the water column to raise the water
temperatures of the hypolimnion immediate above
the dam prior to release. Local destratification is
(a) (b)
(c) (d)
(e) (f)
Fig. 5 Current management options for
mitigating thermal impacts from dams
include: (a) selective withdrawal using
multi-level intake structures or (b) trun-
nions, (c) local destratification using aer-
ation systems, (d) surface pumps or (e)
draft tubes and (f) submerged weirs or
curtains to redirect either hypolimnetic or
epilimnetic water. Figures were provided
by Brad Sherman (Australia’s Common-
wealth Scientific and Industrial Research
Organisation: CSIRO).
98 J. D. Olden and R. J. Naiman
2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd, Freshwater Biology,55, 86–107

accomplished by pushing cold water up to the surface
using aeration systems that introduce bubble plumes
or directly mixing using pumps (Fig. 5c). Bubble
plume destratification systems have been typically
applied in eutrophic reservoirs to improve overall
water quality, particularly to increase dissolved oxy-
gen content and decrease the potential for cyanobac-
terial blooms, however this approach can also greatly
increase hypolimnetic temperature water adjacent to
the intake structure.
4Similarly, surface pumps on floating platforms are
also used to locally destratify reservoirs by sending
surface water downwards into the withdrawal zone of
the intake structure (Fig. 5d).
5Draft tubes can be anchored in the water column
to increase the distance travelled by the plume of
water from the propeller (Fig. 5e).
6Submerged weirs or curtains suspended at vari-
ous depths can be used to provide a barrier to the
passage of water and force warm or cold water above
or below the curtain, respectively (Fig. 5f). Modifying
the topography of the channel feeding into the dam
can also result in a similar effect.
7Lastly, stilling basins (i.e. large shallow ponds)
have been used to delay the downstream release of
water so thermal equilibrium may be reached with the
atmosphere.
An insightful comparison of these technologies for
the Burrendong Dam (Australia) is provided by
Sherman (2000).
Selective withdrawal using a multi-level intake
structure is the most common and most effective
means of controlling the water temperature of dam
releases. A selective withdrawal system (also called a
temperature control device) can extract water from
selected depths of a thermally stratified reservoir to
produce a release with desired characteristics (Price &
Meyer, 1992). This technology provides the flexibility
to increase water temperatures by preferentially
selecting warm epilimnetic water from the surface,
or decrease water temperature by drawing cold
hypolimnetic water from below the thermocline. For
example, Flaming Gorge Dam (Green River, U.S.A.)
was installed with a multi-level intake structure in
1978 with the goal of increasing summer water
temperatures for native species in the trout-domi-
nated tailwaters. The partial success of this thermal
restoration is evident by examining the annual ther-
mographs below the dam during different time
periods (Fig. 6). After the installation of the intake
structure (1978–2007), average stream temperatures
during the summer months of May to August almost
doubled from a pre-installation (1963–77) temperature
of 5.7 to 11.0 C. Pre-dam temperatures (1958–62) in
the Green River during the summer averaged 17.2 C,
but restoring the tailwater to historical temperatures
was not the management goal (Fig. 6). This translates
to a dramatic decline in per cent thermal alteration
during the summer from )66%(pre-dam versus
pre-installation: Fig. 4b) to )36%(pre-dam versus
Fig. 6 Selective discharge below Flaming
Gorge Dam using a multi-level intake
structure has markedly decreased the
degree of thermal alteration in the Green
River (U.S.A.). Comparisons of monthly
water temperature during pre-dam (1958–
62, circles), post-dam (1963–77, squares)
and post-thermal restoration years (1978–
2007, triangles) shows significant in-
creases in spring–summer temperatures
toward unregulated conditions. Notably,
current operations of the temperature-
control device have not decreased the
degree of thermal alteration during the
winter months.
Incorporating thermal regimes in environmental flows assessments 99
2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd, Freshwater Biology,55, 86–107

post-installation). Reasonable improvements were
also observed for the frequency and duration of
thermal events, and the timing of extreme tempera-
tures. Minimum temperature after thermal restoration
occurred, on average, on 8th February (only 4 days
earlier than the pre-dam average date of 12th Febru-
ary), and mean day of year of maximum temperatures
recovered by moving 10 weeks earlier from 7th
November (pre-installment) to 18th August (post-
installation); now representing a 5-week delayed
timing compared to pre-dam years (24th July). The
biological implications of these thermal improve-
ments are discussed later. With another management
goal in mind, Shasta Dam (Sacramento River, U.S.A.)
was retrofitted with a multi-level intake structure in
1997 to improve downstream temperatures for endan-
gered coldwater salmonids. Dam management is
focused on releasing warmer surface waters in the
winter⁄spring and colder deep waters in the sum-
mer⁄autumn. These two examples illustrate the flexi-
bility of multi-level intake structures for controlling
downstream temperatures for a defined ecological
goal.
Although most selective withdrawal intake struc-
tures are built during initial reservoir construction
(ironically, most were originally designed to support
tailwater trout fisheries), release structures can also be
successfully modified for selective withdrawal later
following initial construction (see both example
above). However, the capital costs of refitting dams
with a multi-level intake structure may be prohibitive.
For example, the installation cost for several large
dams in Australia was estimated to range between 3
and 30 million AUD (Sherman, 2000), and the actual
cost for retrofitting Shasta Dam was 80 million USD.
Estimated installation costs for modifying eight intake
portals of Glen Canyon Dam (Colorado River, U.S.A.)
is 15 million USD. Assuming operations occur during
the months of June through August and all releases
are surface withdrawals, the economic value of the
additional increases in head loss (resulting from the
modified intake) are estimated to be 228 000 USD -
year
-1
(Vermeyen, 1999). Economic costs aside, there
are also a host of other operational, physical⁄chemical
and biological considerations that must be assessed
when decided whether to refit a dam with a temper-
ature control device.
Despite the capital costs of installing multi-level
intakes, recent investigations suggest that these struc-
tures provide the most flexible means of modifying
downstream water temperatures, even at low to
medium release volumes (Sherman, 2000), and
may represent the best opportunities for ecological
restoration. Sherman et al. (2007) predicted that the
installation of a multi-level intake structure in Hume
Dam, Australia, would increase discharge tempera-
tures by 4–6 C during the spring–early summer post-
spawning period for Murray cod. These temperature
increases were forecasted to increase minimum
female population abundance in the Murray River
by 30–300%depending on the assumed spawning
behaviour. Research continues on how epilimnetic
versus hypolimnetic withdrawal affects reservoir
water quality (e.g. vertical distribution of tempera-
ture, dissolved oxygen and nutrients) throughout the
year (Johnson et al., 2004), which will help determine
the optimal selective withdrawal release strategy.
Future research is required to compare the ability of
different mitigation strategies (Fig. 5) to meet re-
quired temperature standards defined in environmen-
tal flow prescriptions.
There is also opportunity to manage reservoir
operations to directly affect average daily down-
stream temperatures through adjustments in the
magnitude and timing of discharge releases. Polehn
& Kinsel (1997) examined this analytically and sug-
gested that changes in flow may be used to adjust the
diurnal temperature cycle at specific locations down-
stream of a reservoir. However, Lowney (2000)
showed that an interesting pattern of ‘nodes’ and
‘antinodes’ in the extent of diel variation may develop
at different distances downstream from large dams if
any particular flow is sustained for more than a few
days. Consequently, the control of nodes and antin-
odes through flow change may require continuous
flow adjustment, possibly in conflict with project
objectives. In large systems where constant flow is
desired and a temperature control system may
already be in place, it is possible that a well designed
temperature release pattern could restore diurnal
variation to that expected under pre-dam conditions
(McMahon & Finlayson, 1995). However, meeting diel
thermal targets using modified release schedules will
be complicated in regulated systems because it will
not be possible to mimic the diel temperature varia-
tions of a natural system if patterns in discharge are
not also mimicked. Quite simply, variation in diel
temperatures decreases with increasing discharge
100 J. D. Olden and R. J. Naiman
2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd, Freshwater Biology,55, 86–107

due to the increase in thermal mass of the river.
Typically, flows released during the summer, either
for irrigation or recreation, exceed the natural flow for
a river. While it may be possible to modify the mean
temperature of the release using a particular mitiga-
tion technique, it is unlikely that the nature temper-
ature range will also be restored. In general, the
feasibility of controlling river temperature through
streamflow management has been recognised for
some time, yet the quantitative temperature : flow
relationships required for implementing instream
flow requirements based on this principle have not
be successfully developed (Gu et al., 1999).
Challenge #4: Incorporate thermal-criteria into environ-
mental flow assessments
Defining environmental flows with respect to either
flow or temperature in isolation is highly unlikely to
support ecologically sustainable water management.
Consequently, the benefits of environmental flows
will be realised only if riverine thermal regimes are
also considered. One example supporting this notion
is the study of King, Cambray & Impson (1998) who
examined patterns of temperature and discharge
resulting from an experimental dam release, and their
relative importance in triggering successful spawning
of a threatened large cyprinid endemic, the Clanwil-
liam yellowfish Barbus capensis Smith, 1841, down-
stream of Clanwilliam Dam (Olifants River, Africa).
A critical flow component for many rivers in the
winter-rainfall region of South Africa, including the
Olifants River, is the small pulses of higher flow that
occur in the dry season (called freshes). Experimental
freshes released from Clanwilliam Dam in the early
1990s were strongly correlated with the suspected
spawning success of B. capensis (Cambray, King &
Bruwer, 1997); however, in subsequent years freshes
delivered during the species’ breeding season failed to
induce spawning. King et al. (1998) found that differ-
ences in the temperatures of the water release were
the key factor related to spawning success. Specifi-
cally, warm epilimnetic freshes (19–21 C) triggered
fish spawning behaviour and the movement of indi-
viduals onto spawning beds, whereas cool hypolim-
netic baseflows (16–18 C) released immediately after
the experimental freshets caused fish to abort spawn-
ing activities. Moreover, the occurrence of dead and
deformed young suggested that the cold water may
have had a detrimental effect on offspring of those
individuals who spawned during the warmwater
events. This research showed that freshets released
from Clamwilliam Dam at the appropriate time
should be able to induce spawning and support early
life stages of B. capensis only if water temperatures
at the spawning sites exceeded 19 C. Furthermore,
King et al. (1998) suggest that successful spawning
will lead to high recruitment only if water tempera-
tures are maintained at these levels for an extended
period after spawning to provide for the development
of the embryos and larvae.
Similar examples exist in the lower Mississippi River
(U.S.A.) where research has shown that growth and
abundance of juvenile fishes are only linked to flood-
plain inundation when water temperatures are greater
than a particular threshold. Schramm & Eggleton (2006)
reported that the growth of catfishes (Ictaluridae spp.)
was significantly related to the extent of floodplain
inundation only when water temperature exceeded
15 C; a threshold temperature for active feeding and
growth by catfishes. Under the current hydrographic
conditions in the lower Mississippi River, the authors
report that the duration of floodplain inundation when
water temperature exceeds the threshold is only about
1 month year
)1
on average. Such a brief period of time
is believed to be insufficient for floodplain-foraging
catfishes to achieve a detectable energetic benefit
(Schramm & Eggleton, 2006). These results are consis-
tent with the ‘thermal coupling’ hypothesis offered by
Junk, Bayley & Sparks (1989) whereby the concordance
of both hydrologic and thermal cycles is required for
maximum ecological benefit.
Similarly, the advantages of thermal restoration
may be realised only if flow is also actively managed
to mimic natural regimes. As previously discussed,
the operation of Flaming Gorge Dam after installation
of a multi-level intake structure resulted in significant
warming of summer temperatures (Fig. 6) and in-
creased the number of annual degree days from 2340
to 3200 (Vinson, 2001). While these improvements in
thermal regimes were predicted to increase inverte-
brate richness, Vinson (2001) found that the number of
taxa collected after partial thermal restoration was
similar to or lower than that observed before temper-
ature manipulations. The lack of appreciable increase
in richness can be attributed, in part, to the remaining
differences in the thermal and hydrologic (and asso-
ciated sedimentation processes) regimes of the Green
Incorporating thermal regimes in environmental flows assessments 101
2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd, Freshwater Biology,55, 86–107

River below Flaming Gorge Dam. In summary, the
studies discussed in the previous paragraphs provide
striking examples emphasizing that both flow and
thermal regimes must be managed simultaneously to
have the desired effects on riverine biota.
A key challenge in environmental flow assessments
is to determine the critical flow and thermal require-
ments of riverine ecosystems, while continuing to
provide the goods and services expected by human
society. Will there be conflicts between flow and
thermal targets in environmental flow prescriptions?
In the absence of a temperature control device, we
expect the answer to this question will be yes, quite
simply because stream temperature co-varies with
flow. For example, Krause et al. (2005) found that
proposed modifications to a regulated discharge
regime below Philpott Dam (Smith River, U.S.A.)
designed to improve physical habitat for brown trout
(Salmo trutta Linnaeus, 1758) would concurrently
reduce the frequency of optimal growth temperatures.
This study illustrated that the best management
option for improving flow regimes (e.g. for providing
critical fish habitat) was not necessarily the best for
improving thermal regimes (e.g. for providing opti-
mal growth conditions). Consequently, different
aspects of the temperature regime for any stream
would require careful analysis before any environ-
mental flow requirements could be safely prescribed.
Empirical data and predictive models are needed to
help identify and incorporate the potential interactive
effects of discharge and temperature recommenda-
tions in environmental flow assessments.
Synchronised management efforts are needed to
meet both flow and temperature requirements of
entire ecosystems by recoupling the natural hydro-
logic and thermal cycles. Given the complexities and
trade-offs regarding alternative dam strategies for
meeting both downstream flow and thermal stan-
dards, particularly in the absence of a temperature
control device, we believe the environmental flow
assessments would benefit from the use of formal
optimisation frameworks incorporated into adaptive
management strategies. Pareto efficiency (or Pareto
optimality), for example, is a central concept in
economics that is defined as the efficiency of a market
which is unable to produce more from the same level
of inputs without reducing the output of another
product. Recent efforts have developed Pareto-
optimal solutions for environmental flow schemes
that incorporate variability in flow regimes and
provide for human needs (e.g. Suen & Eheart, 2006;
Shiau & Wu, 2007), and we believe that the appli-
cation of Pareto optimality holds promise for incor-
porating thermal criteria into environmental flow
assessments. For example, this multi-objective opti-
misation approach could be applied to identify a set of
flow schemes (i.e. those contained in the Pareto
frontier) that are considered efficient with respect to
meeting both the temperature and flow needs of a
riverine ecosystem while also ensuring the water
requirements of human users.
Pareto optimisation could account for the fact that
particular flow events that may be deemed critical for
riverine ecosystems (i.e. high discharge events for
mobilizing sediment and shaping channel forming
processes) may not be optimal for restoring critical
components of the thermal regime (i.e. high discharge
events can increase the magnitude and spatial extent of
coldwater pollution). To illustrate this, we examine the
Coosa River tailwater below Jordan Dam (U.S.A.),
where dam influences on tailwater temperatures
extend approximately 4 km downstream under low
flow regimes and nearly 15 km downstream under high
flow regimes (Jackson & Davies, 1988). In this instance,
environmental flow assessments that are considered
important for environmental flows are detrimental
with respect to thermal targets. By identifying a set of
efficient and acceptable environmental flows, an opti-
mal solution could be selected to represent what would
be considered the optimal environmental flow pre-
scription for a given river system. Relationships
between the conflicting flow and thermal objectives
would offer decision makers with the marginal trade-
offs useful for the selection of the preferred solutions. In
summary, there is continued need for the collection of
more empirical data and development of pragmatic
models that will provide the basis for adaptive man-
agement schemes involving both riverine flow and
thermal regimes (Rivers-Moore & Jewitt, 2007).
Challenge #5: Designing temperature-enlightened
environmental flow assessments in a changing climate
The prospect of dramatic climate change over the next
century underscores the need for adaptive manage-
ment strategies for effectively designing and imple-
menting environmental flows (Palmer et al., 2008).
Projected increases in air temperatures, combined
102 J. D. Olden and R. J. Naiman
2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd, Freshwater Biology,55, 86–107

with lower accumulation in winter snowpack, earlier
onset of spring peak flows, and lower summer
baseflows, will have direct implications for the ther-
mal regimes of streams and rivers (Poff, Brinson &
Day, 2002). Equally relevant is the increasingly large
and thirsty human population which is rapidly
changing the geography of water demand and dra-
matically shifting current land use practices (Poff
et al., 2003; Fitzhugh & Richter, 2004), both of which
have direct and indirect implications for water tem-
peratures. By failing to account for projected climate
change, ecologists run the risk of making flow
recommendations based on the characteristics of past
thermal regimes that are not favoured by present-day
or future environments.
Future climate change will significantly influence
how dams are operated to achieve environmental
flows that incorporate water temperatures. For exam-
ple, Sinokrot et al. (1995) found that the predicted
impact of climate change on stream temperatures
below dams is more pronounced when water is
released from the reservoir surface compared to
deeper depths. Results from studies like this have
been used by some to advocate that management
agencies could use dams with deep, coldwater
releases to control thermal regimes in a manner that
offsets climate warming effects. We do not question
that innovative mitigation strategies are needed to
prepare for a rapidly changing climate; however,
we have a number of concerns with this pro-
posed management approach. First, climate-induced
changes in stream discharge are expected to have
direct (i.e. thermal capacity) and indirect (i.e. changes
to channel geomorphology and riparian vegetation)
consequences for seasonal and annual patterns in
water temperature, which will be very difficult to
predict. Secondly, as previously discussed, hypolim-
netic releases by dams from thermally stratified
reservoirs often result in abnormally warmer water
temperatures during the winter. This warming effect
is likely to be compounded in those regions predicted
to experience higher air temperatures during the
winter season; a trend already observed for some
rivers (Durance & Ormerod, 2009). Taken together, we
believe that our ability to mitigate the thermal
consequences of climate change through the manage-
ment of dam operations is likely to be unsuccessful. In
summary, we urge ecologists to ground their knowl-
edge in the past, but to look to the future in their
scientific endeavours. There is no doubt that environ-
mental flow assessments will benefit greatly from
directly incorporating the predicted response of
freshwater ecosystems to projected changes in flow
and thermal regimes from climate change.
Prospectus
Recent decades have witnessed an increasing recogni-
tion of the importance of variability of ecosystem
drivers, such as river flow and water temperature in
maintaining river integrity. Even though there has been
tremendous progress in environmental flow research
and implementation, what remains unknown is formi-
dable. Is more water alone sufficient without restoring
other essential water quality variables, such as recou-
pling the natural flow and thermal regimes? As we
continue to learn about these issues, clear and complete
answers remain elusive because of the complexity of
the scientific questions and management options.
Clearly, new strategies are needed that account for
the natural dynamics in temperature-related ecosystem
processes by using the natural thermal regime as one
template for environmental flow management.
From its name, the concept of ‘environmental flows’
would seem to offer the features of ecologically
sustainable water management, but in its current
form it has evolved to include only water quantity
and it is not comprehensive enough to solve the broad
water resource problems that will challenge us in the
future. We firmly believe that the benefit of using
environmental flows as a paradigm in water manage-
ment will be its focus on the blending of both water
quantity and quality. For this reason, many additional
aspects of water quality must be recognised, including
pollutants, nutrients, organic matter and sediments
and dissolved oxygen (Nilsson & Reno
¨fa
¨lt, 2008),
which can interact with water temperature and
discharge in complex ways.
Our objective was to add clarity to the increasing
popular but still poorly understood concept of an
environmental flow. The issue is not that ecologists do
not recognise the importance of water temperature for
riverine ecosystems; rather to date we have not
formally integrated it into environmental flow assess-
ments. This study began with an overview of the
natural thermal regime and its ecological impor-
tance, and ended with a discussion of some over-
arching research opportunities and challenges to
Incorporating thermal regimes in environmental flows assessments 103
2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd, Freshwater Biology,55, 86–107

incorporating thermal criteria into the prescription of
environmental flows.
Based on our assessment, we suggest that scientific
research should focus on the ongoing collection of
long-term temperature records, comprehensive char-
acterisation of seasonality and variability in stream
temperatures, quantification of the temporal and
spatial impacts of dam operations on thermal regimes,
and clearer elucidation of the relative roles of altered
flow and temperature in shaping ecological patterns
and processes in riverine ecosystems. Future investi-
gations should also concentrate on using this acquired
knowledge to identify the ‘manageable’ components
of the thermal regime, and develop optimisation
models that evaluate management trade-offs and
provide a range of optimal environmental flows that
meet both ecosystem and human needs for fresh
water. Lastly, ecologists and water managers must
maintain a nimble capacity to incorporate results from
ongoing climate change research into adaptive man-
agement strategies involving environmental flows.
Acknowledgments
We thank participants of the 3rd International Sympo-
sium on Riverine Landscapes (South Stradbroke Island,
Queensland, Australia) for insightful conversations,
and Mark Kennard, Angela Arthington and Stuart
Bunn for their kind hospitality during our visit. We
gratefully acknowledge Steve Ormerod and Christer
Nilsson for comments on the manuscript, Bruce Webb
for discussions regarding the thermal impacts of dam
operation, David Archer, Mark Vinson and Daqing
Yang for providing temperature data, and Brad Sher-
man for generously providing the drawings in Fig. 5. J.
D. Olden thanks Land and Water Australia and the
Tropical Rivers and Coastal Knowledge (TRaCK)
research consortium for partial funding support, and
R. J. Naiman acknowledges the Freshwater Committee
of DIVERSITAS and the Global Water System Project. J.
D. Olden conceived and developed the idea for the
manuscript and conducted the data analysis, and J. D.
Olden and R. J. Naiman wrote the manuscript.
References
Acreman M. & Dunbar M.J. (2004) Defining
environmental river flow requirements: a review.
Hydrology and Earth System Sciences,8, 861–876.
Alcamo J., Vo
¨ro
¨smarty C., Naiman R.J., Lettenmaier D. &
Pahl-Wostl C. (2008) A grand challenge for freshwater
research: understanding the global water system.
Environmental Research Letters,3, 1–6.
Angilletta M.J., Steel E.A., Bartz K.K., Kingsolver J.G.,
Scheurell M.D., Beckman B.R. & Crozier L.G. (2008)
Big dams and salmon evolution: changes in thermal
regimes and their potential evolutionary conse-
quences. Evolutionary Applications,1, 286–299.
Armour C.L. (1991) Guidance for evaluating and
recommending temperature regimes to protect fish.
U.S. Fish and Wildlife Service Biological Report,90, 1–13.
Arthington A.H., Bunn S.E., Poff N.L. & Naiman R.J.
(2006) The challenge of providing environmental flow
rules to sustain river ecosystems. Ecological Applications,
16, 1311–1318.
Bednarek A.T. & Hart D.D. (2005) Modifying dam
operations to restore rivers: ecological responses to
Tennessee River dam mitigation. Ecological Applications,
15, 997–1008.
Berman C.H. & Quinn T.P. (1991) Behavioral
thermoregulation and homing by spring Chinook
salmon, Oncorhynchus tshawytscha (Walbaum), in the
Yakima River. Journal of Fish Biology,39, 301–312.
Brooker M.P. (1981) The impact of impoundments on the
downstream fisheries and general ecology of rivers.
Advances in Applied Biology,6, 91–152.
Bunn S.E. & Arthington A.H. (2002) Basic principles and
ecological consequences of altered flow regimes for
aquatic biodiversity. Environmental Management,30,
492–507.
Caissie D. (2006) The thermal regime of rivers: a review.
Freshwater Biology,51, 1389–1406.
Cambray J.A., King J. & Bruwer C. (1997) Spawning
behaviour and early development of the Clanwilliam
yellowfish (Barbus capensis; Cyprinidae), linked to
experimental dam releases in the Olifants River, South
Africa. Regulated Rivers: Research and Management,13,
579–602.
Clarkson R.W. & Childs M.R. (2000) Temperature effects
of hypolimnial-release dams on early life stages of
Colorado River Basin big-river fishes. Copeia,2000,
402–412.
Coutant C.C. (1987) Thermal preference: when does an
asset become a liability? Environmental Biology of Fishes,
18, 161–172.
Dunbar M.J., Acreman M. & Kirk S. (2004)
Environmental flow setting in England and Wales:
strategies for managing abstraction in catchments.
Water and Environment Journal,18, 5–10.
Durance I. & Ormerod S.J. (2009) Trends in water quality
and discharge offset long-term warming effects on river
macroinvertebrates. Freshwater Biology,54, 388–405.
104 J. D. Olden and R. J. Naiman
2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd, Freshwater Biology,55, 86–107

Fitzhugh T.W. & Richter B.D. (2004) Quenching urban
thirst: growing cities and their impacts on freshwater
ecosystems. BioScience,54, 741–754.
Fraley J.J. (1979) Effects of elevated stream temperatures
below a shallow reservoir on cold-water
macroinvertebrate fauna. In: The Ecology of Regulated
Streams (Eds J.V. Ward & J.A. Stanford), pp. 257–272.
Plenum, New York.
Gu R., McCutcheon S. & Chen C.-J. (1999) Development
of water-dependent flow requirements for river
temperature control. Environmental Management,24,
529–540.
Haxton T.J. & Findlay C.S. (2008) Meta-analysis of the
impacts of water management on aquatic
communities. Canadian Journal of Fisheries and Aquatic
Sciences,65, 437–447.
Horne B.D., Rutherford E.S. & Wehrly K.E. (2004)
Simulating effects of hydro-dam alteration on
thermal regime and wild steelhead recruitment in a
stable-flow Lake Michigan tributary. River Research and
Applications,20, 185–203.
Huryn A.D. & Wallace J.B. (2000) Life history and
production of stream insects. Annual Review of
Entomology,45, 83–110.
Jackson D.C. & Davies W.D. (1988) Environmental
factors influencing summer angler effort on the
Jordan Dam tailwater. North American Journal of
Fisheries Management,8, 305–309.
Jackson H.M., Gibbins C.N. & Soulsby C. (2007) Role of
discharge and temperature variation in determining
invertebrate community structure in a regulated river.
River Research and Applications,23, 651–669.
Johnson B.M., Saito L., Anderson M.A., Weiss P., Andre
M. & Fontane D.G. (2004) Effects of climate and dam
operations on reservoir thermal structure. Journal of
Water Resources Planning and Management,130, 112–122.
Junk W.J., Bayley P.B. & Sparks R.E. (1989) The flood pulse
concept in river–floodplain systems. In: Proceedings of
the International Large River Symposium (Ed. D.P. Dodge),
pp. 110–127. Canadian Special Publications in Fisheries
and Aquatic Sciences 106. Toronto, Canada.
King J., Cambray J.A. & Impson N.D. (1998) Linked
effects of dam-released floods and water temperature
on spawning of the Clanwilliam yellowfish Barbus
capensis.Hydrobiologia,384, 245–265.
Krause C.W., Newcomb T.J. & Orth D.J. (2005) Thermal
habitat assessment of alternative flow scenarios in a
tailwater fishery. River Research and Applications,21,
581–593.
Langford T.E.L. (1990) Ecological Effects of Thermal Dis-
charges. Elsevier, London.
Lehmkuhl D.M. (1974) Thermal regime alterations and
vital environmental physiological signals in aquatic
systems. In: Thermal Ecology (Eds J.W. Gibbons & R.R.
Sharitz), pp. 216–222. AEC Symposium Series,
Springfield, Virginia.
Lessard J.L. & Hayes D.B. (2003) Effects of elevated water
temperature on fish and macroinvertebrate comm-
unities below small dams. River Research and
Applications,19, 721–732.
Lowney C.L. (2000) Stream temperature variation in
regulated rivers: evidence for a spatial pattern in daily
minimum and maximum magnitudes. Water Resource
Research,36, 2947–2955.
Lugg A. (1999) Eternal Winter in our Rivers: Addressing the
Issue of Water Pollution. New South Wales Fisheries,
Nowra, NSWa.
Magnuson J.J., Crowder L.B. & Medvick P.A. (1979)
Temperature as an ecological resource. American
Zoologist,19, 331–343.
McMahon T.A. & Finlayson B.L. (1995) Reservoir system
management and environmental flows. Lakes and
Reservoirs: Research and Management,1, 65–76.
Moore R.D., Spittlehouse D.L. & Story A. (2005) Riparian
microclimate and stream temperature response to
forest harvesting: a review. Journal of the American
Water Resources Association,41, 813–834.
Murchie K.J., Hair K.P.E., Pullen C.E., Redpath T.D.,
Stephens H.R. & Cooke S.J. (2008) Fish response to
modified flow regimes in regulated rivers: research
methods, effects and opportunities. River Research and
Applications,24, 197–217.
Nelson K.C. & Palmer M.A. (2007) Stream temperature
surges under urbanization and climate change: data,
models, and responses. Journal of the American Water
Resources Association,43, 440–452.
Nilsson C. & Reno
¨fa
¨lt B.M. (2008) Linking flow regime
and water quality in rivers: a challenge to adaptive
catchment management. Ecology and Society,13.
Available at: http://www.ecologyandsociety.org/
vol13/iss2/art18/.
Olden J.D. (2004) Fish Fauna Homogenization of the United
States, Life-History Correlates of Native Extinction and
Nonnative Invasions in the American Southwest, and the
Bidirectional Impacts of Dams in the American Southeast.
Dissertation. Colorado State University, Fort Collins,
CO.
Olden J.D. & Poff N.L. (2003) Redundancy and the choice
of hydrologic indices for characterizing streamflow
regimes. River Research and Applications,19, 101–121.
Palmer R.W. & O’Keeffe J.H. (1989) Temperature
characteristics of an impounded river. Archiv fu
¨r
Hydrobiologie,116, 471–485.
Palmer M.A., Bernhardt E., Chornesky E. et al. (2004)
Ecology for a crowded planet. Science,304, 1251–
1252.
Incorporating thermal regimes in environmental flows assessments 105
2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd, Freshwater Biology,55, 86–107

Palmer M.A., Reidy Liermann C., Nilsson C., Flo
¨rke M.,
Alcamo J., Lake P.S. & Bond N. (2008) Climate change
and world’s river basins: anticipating management
options. Frontiers in Ecology and the Environment,6,
81–89.
Peterson J.T. & Rabeni C.F. (1996) Natural thermal
refugia for temperate warmwater stream fishes. North
American Journal of Fisheries Management,16, 738–746.
Petts G.E. (1986) Water quality characteristics of
regulated rivers. Progress in Physical Geography,10,
492–516.
Poff N.L., Allan J.D., Bain M.B., Karr J.R., Prestegaard
K.L., Richter B.D., Sparks R.E. & Stromberg J.C. (1997)
The natural flow regime: a paradigm for river
conservation and restoration. BioScience,47, 769–784.
Poff N.L., Brinson M.M. & Day J.W. Jr (2002) Aquatic
Ecosystems and Global Climate Change. Pew Center on
Global Climate Change. Arlington, VA.
Poff N.L., Allan J.D., Palmer M.A., Hart D.D., Richter
B.D., Arthington A.H., Rogers K.H., Meyer J.L. &
Stanford J.A. (2003) River flows and water wars?
Emerging science for environmental decision-
making. Frontiers in Ecology and the Environment,1,
298–306.
Poff N.L., Richter B.D., Arthington A.H. et al. (2010) The
ecological limits of hydrologic alteration (ELOHA):
a new framework for developing regional envi-
ronmental flow standards. Freshwater Biology,55, 147–
170.
Polehn R.A. & Kinsel W.C. (1997) Transient temperature
solution for stream flow from a controlled temperature
source. Water Resource Research,33, 261–265.
Poole G.C. & Berman C.H. (2001) An ecological
perspective on in-stream temperature: natural heat
dynamics and mechanisms of human-caused thermal
degradation. Environmental Management,27, 787–802.
Preece R.M. (2004) Cold Water Pollution Below Dams in
New South Wales: A Desktop Assessment. NSW Depart-
ment of Infrastructure, Planning and Natural Re-
sources, Sydney, NSW.
Preece R.M. & Jones H.A. (2002) The effect of Keepit Dam
on the temperature regime of the Namoi River,
Australia. River Research and Applications,18, 397–414.
Price R.E. & Meyer E.B. (1992) Water Quality Management
for Reservoirs and Tailwaters: Operational and Structural
Water Quality Techniques. Technical Report E-89-1. US
Army Engineer Waterways Experimental Station,
Vicksburg, MS.
Rice S.P., Ferguson R.L. & Hoey T.B. (2006) Tributary
control of physical heterogeneity and biological
diversity at river confluence. Canadian Journal of
Fisheries and Aquatic Sciences,63, 2553–2566.
Richter B.D. & Thomas G.A. (2007) Restoring
environmental flows by modifying dam operations.
Ecology and Society,12. Available at: http://www.
ecologyandsociety.org/vol12/iss1/art12/.
Richter B.D., Matthews R.A., Harrison D.L. & Wigington
R. (2003) Ecologically sustainable water management:
managing river flows for river integrity. Ecological
Applications,13, 206–224.
Richter B.D., Warner A.T., Meyer J.L. & Lutz K. (2006) A
collaborative and adaptive process for developing
environmental flow recommendations. River Research
and Applications,22, 297–318.
Rivers-Moore N.A. & Jewitt G.P.W. (2007) Adaptive
management and water temperature variability within
a South Africa river system: what are the management
options? Journal of Environmental Management,82, 39–
50.
Ryan T., Webb A., Lennie R. & Lyon J. (2001) Status of
Cold Water Releases from Victorian Dams. Victorian
Department of Natural Resources and Environment,
Heidelberg.
Sato Y., Bazzoli N., Rizzo E., Boschi M.B. & Miranda
M.O.T. (2005) Influence of the Abaete River on the
reproductive success of the neotropical migratory
teleost Prochilodus argenteus in the Sao Francisco
River, downstream from the Tres Marias Dam,
southeastern Brazil. River Research and Applications,
21, 939–950.
Schramm H.L. & Eggleton M.A. (2006) Applicability of
the flood-pulse concept in a temperate floodplain river
ecosystem: thermal and temporal components. River
Research and Applications,22, 543–553.
Sherman B. (2000) Scoping Options for Mitigating Cold
Water Discharges from Dams. CSIRO Land and Water,
Canberra.
Sherman B., Todd C.R., Koehn J.D. & Ryan T. (2007)
Modelling the impact and potential mitigation of cold
water pollution on Murray cod populations
downstream of Hume Dam, Australia. River Research
and Applications,23, 377–389.
Shiau J.-T. & Wu F.-C. (2007) Pareto-optimal solutions
for environmental flow schemes incorporating the
intra-annual and interannual variability of the
natural flow regime. Water Resource Research,43,
W06433.
Sinokrot B.A. & Gulliver J.S. (2000) In-stream flow impact
on river water temperatures. Journal of Hydraulic
Research,38, 339–349.
Sinokrot B.A., Stefan H.G., McCormick J.H. & Eaton J.G.
(1995) Modeling of climate change effects on stream
temperatures and fish habitats below dams and near
groundwater inputs. Climatic Change,30, 181–200.
106 J. D. Olden and R. J. Naiman
2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd, Freshwater Biology,55, 86–107

Steel E.A. & Lange I.A. (2007) Using wavelet analysis to
detect changes in water temperature regimes at
multiple scales: effects of multi-purpose dams in the
Willamette River basin. River Research and Applications,
23, 351–359.
Stevens L.E., Shannon J.P. & Blinn D.W. (1997) Colorado
River benthic ecology in Grand Canyon, Arizona, USA:
dam, tributary and geomorphological influences.
Regulated Rivers: Research and Management,13, 129–149.
Suen J.-P. & Eheart J.W. (2006) Reservoir management to
balance ecosystem and human needs: incorporating
the paradigm of the ecological flow regime. Water
Resource Research,42, W03417.
Todd C.R., Ryan T., Nicol S.J. & Bearlin A.R. (2005) The
impact of cold water releases on the critical period of
post-spawning survival and its implications for
Murray cod (Maccullochella peelii peelii): a case study
of the Mitta Mitta River, southeastern Australia. River
Research and Applications,21, 1035–1052.
Vannote R.L. & Sweeney B.W. (1980) Geographic
analysis of thermal equilibria: a conceptual model for
evaluating the effect of natural and modified thermal
regimes on aquatic insect communities. American
Naturalist,115, 667–695.
Vermeyen T.B. (1999) An Overview of the Design Concept
and Hydraulic Modeling of the Glen Canyon Dam Multi-
Level Intake Structure. ASCE’s Waterpower ‘99 Confer-
ence Proceedings, July 6–9, 1999, Las Vegas, NV.
Vinson M.R. (2001) Long-term dynamics of an
invertebrate assemblage downstream from a large
dam. Ecological Applications,11, 711–730.
Vinson M.R. & Hawkins C.P. (1998) Biodiversity of
stream insects: variation at local, basin, and regional
scales. Annual Review of Entomology,43, 271–293.
Voelz N.J. & Ward J.V. (1991) Biotic responses along
the recovery gradient of a regulated stream.
Canadian Journal of Fisheries and Aquatic Sciences,48,
2477–2490.
Ward J.V. (1974) A temperature-stressed stream
ecosystem below a hypolimnial release mountain
reservoir. Archive fu
¨r Hydrobiologie,74, 247–275.
Ward J.V. (1985) Thermal-characteristics of running
waters. Hydrobiologia,125, 31–46.
Ward J.V. & Stanford J.A. (1982) Thermal responses in
the evolutionary ecology of aquatic insects. Annual
Review of Entomology,27, 97–117.
Webb B.W. (1996) Trends in stream and river
temperature. Hydrological Processes,10, 205–226.
Webb B.W. & Walling D.E. (1993) Temporal variability in
the impact of river regulation on thermal regime and
some biological implications. Freshwater Biology,29,
167–182.
Webb B.W. & Walling D.E. (1996) Long-term variability
in the thermal impact of river impoundment and
regulation. Applied Geography,16, 211–223.
Webb B.W., Hannah D.W., Moore R.D., Brown L.E. &
Nobilis F. (2008) Recent advances in stream and river
temperature research. Hydrologic Processes,22, 902–918.
Wootton R.J. (1990) Ecology of Teleost Fishes. Chapman
and Hall, London.
Zhong Y. & Power G. (1996) Environmental impacts of
hydroelectric projects on fish resources in China.
Regulated Rivers: Research and Management,12, 81–98.
Zhou C., Liang Z. & Huang H. (1980) Ecological features
of the spawning of certain fishes in the Hanjiang River
after the construction of dams. Acta Hydrobiologia
Sinica,7, 175–188 [in Chinese with English abstract].
(Manuscript accepted 9 January 2009)
Incorporating thermal regimes in environmental flows assessments 107
2009 The Authors, Journal compilation 2009 Blackwell Publishing Ltd, Freshwater Biology,55, 86–107