RIVER RESEARCH AND APPLICATIONS

River Res. Applic. 29: 939–960 (2013)
Published online 3 July 2012 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/rra.2590

RESTORING SALMON HABITAT FOR A CHANGING CLIMATE
T. BEECHIEa*, H. IMAKIa, J. GREENEa, A. WADEb, H. WUc,d, G. PESSa, P. RONIa, J. KIMBALLe, J. STANFORDe,
P. KIFFNEYa AND N. MANTUAf
a

b

Northwest Fisheries Science Center, National Marine Fisheries Service, Seattle, Washington, USA
National Center for Ecological Analysis and Synthesis, University of California Santa Barbara, Santa Barbara, California, USA
c
Earth System Science Interdisciplinary Center, University of Maryland, College Park, Maryland, USA
d
NASA Goddard Space Flight Center, Greenbelt, Maryland, USA
e
Flathead Lake Biological Station, University of Montana, Missoula, Montana, USA
f
School of Aquatic and Fishery Sciences, University of Washington, Seattle, Washington, USA

ABSTRACT
An important question for salmon restoration efforts in the western USA is ‘How should habitat restoration plans be altered to accommodate
climate change effects on stream ﬂow and temperature?’ We developed a decision support process for adapting salmon recovery plans that
incorporates (1) local habitat factors limiting salmon recovery, (2) scenarios of climate change effects on stream ﬂow and temperature, (3) the
ability of restoration actions to ameliorate climate change effects, and (4) the ability of restoration actions to increase habitat diversity and
salmon population resilience. To facilitate the use of this decision support framework, we mapped scenarios of future stream ﬂow and
temperature in the Paciﬁc Northwest region and reviewed literature on habitat restoration actions to determine whether they ameliorate a
climate change effect or increase life history diversity and salmon resilience. Under the climate change scenarios considered here, summer
low ﬂows decrease by 35–75% west of the Cascade Mountains, maximum monthly ﬂows increase by 10–60% across most of the region,
and stream temperatures increase between 2 and 6  C by 2070–2099. On the basis of our literature review, we found that restoring ﬂoodplain
connectivity, restoring stream ﬂow regimes, and re-aggrading incised channels are most likely to ameliorate stream ﬂow and temperature changes
and increase habitat diversity and population resilience. By contrast, most restoration actions focused on in-stream rehabilitation are unlikely to
ameliorate climate change effects. Finally, we illustrate how the decision support process can be used to evaluate whether climate change should
alter the types or priority of restoration actions in a salmon habitat restoration plan. Copyright © 2012 John Wiley & Sons, Ltd.
key words: restoration; climate change; decision support; adaptation; salmon habitat; stream ﬂow; stream temperature
Received 24 April 2012; Accepted 30 May 2012

INTRODUCTION
Climate change is predicted to have signiﬁcant effects on
Paciﬁc salmon and their ecosystems in western North
America, and several reports suggest that restoring habitats
for salmon in some places may be pointless because
climate change will make their habitats inhospitable
(Lackey, 2003; Nelitz et al., 2007). By contrast, recent
modelling of the combined effects of climate change and
habitat restoration indicates restoration actions are likely
to result in a net beneﬁt to salmon populations despite future
shifts in temperature and hydrology (Battin et al., 2007). This
lack of consensus on how climate change will affect salmon
populations inhibits the development of clear guidance on
how to modify habitat restoration efforts in response to
climate change. With millions of dollars spent each year to
restore habitats for threatened and endangered salmon in the
western USA, there is increasing concern that climate change
*Correspondence to: T. Beechie, Northwest Fisheries Science Center, 2725
Montlake Blvd. E., Seattle, WA 98112, USA.
E-mail: tim.beechie@noaa.gov

Copyright © 2012 John Wiley & Sons, Ltd.

effects on freshwater habitats may limit the future effectiveness of certain salmon recovery efforts (Lackey et al., 2006;
Battin et al., 2007; Mantua et al., 2010) and that the priority
or design of speciﬁc restoration actions should be altered to
accommodate future climate change (Mote et al., 2003).
Making the decision to adapt a restoration plan for
climate change is not straightforward, as predicted climate
change effects vary widely throughout the Paciﬁc salmon
range, and some species have life histories that will likely
allow them to persist throughout most of their range despite
shifts in temperature and precipitation (Waples et al., 2009).
Stream temperatures are expected to increase in most rivers,
and the threat to salmon recovery is high where temperatures are near lethal or sub-lethal thresholds for salmon,
but low in many rivers with current temperatures well below
those thresholds. Furthermore, some rivers are expected to
see large increases in peak ﬂows, whereas other rivers are
expected to experience decreased low ﬂows (Arnell, 1999,
Mantua et al., 2010). However, past land uses and water
abstraction have often degraded habitats to a greater degree
than that predicted from climate change, presenting substantial

 940

T. BEECHIE ET AL.

opportunities to improve salmon habitats more than enough to
compensate for expected climate change effects over the next
several decades (Battin et al., 2007). For salmonids, variation
in life history strategies and habitat requirements—along with
their demonstrated capacity to adapt to new environments—
further complicates the development of general guidance for
modifying restoration strategies to accommodate climate
change (Quinn, 2005; Beechie et al., 2006; Bryant, 2009).
This complex interplay of climate effects, restoration opportunities, and potential salmon responses poses a considerable
challenge for effectively restoring salmon populations in a
changing climate.
In this paper, we present a simple logic framework and
data sets to assist managers in adapting salmon habitat
restoration efforts to climate change in the Paciﬁc Northwest,
USA (PNW). Our approach consists of four components: (1) a
set of guiding questions that serve as a starting point for evaluating the potential effects of climate change on freshwater
habitat restoration effectiveness; (2) maps showing future
stream ﬂow and temperature scenarios; (3) a review of the
ability of speciﬁc river restoration actions to ameliorate future
effects of climate change or to increase salmon resilience; and
(4) a simple decision support structure that integrates these

three components to help managers evaluate whether salmon
restoration actions should be reprioritized or redesigned for a
climate-altered future. Together, these components guide
decisions on whether and how to adapt or reprioritize actions
in light of expected climate-induced habitat changes.
STUDY AREA
The study area encompasses the Columbia River basin and
coastal drainages of Oregon and Washington (Figure 1), with
climatic and ecological conditions ranging from wet forests in
the Cascade Mountains to semi-arid and desert regions in the
central plateaus (Omernik and Bailey, 1997). The study area
is bordered by the Rocky Mountains to the east, and the
Cascade Mountains separate coastal drainages from the
interior Columbia basin. Mean annual precipitation ranges
from <200 mm/year in the central deserts to 3550 mm/year
in the Cascade Range (Daly et al., 2002), and elevations range
from sea level to over 3700 m in the Rocky Mountains and over
4200 m in the Cascade Mountains. Five anadromous salmon
species (Oncorhynchus spp.) and steelhead (O. mykiss) are
found in the study area, along with bull trout (Salvelinus
conﬂuentus), Dolly Varden char (S. malma), and rainbow and

Figure 1. Map of the study area indicating major rivers, mountain ranges, and current and historical ranges of salmon
Copyright © 2012 John Wiley & Sons, Ltd.

River Res. Applic. 29: 939–960 (2013)
DOI: 10.1002/rra

 RESTORING SALMON HABITAT FOR A CHANGING CLIMATE

cutthroat trout (O. mykiss, O. clarkii). Current ranges of salmon
and steelhead are limited to coastal rivers, the Columbia River
basin downstream of Chief Joseph dam and the Snake River
basin downstream of Hells Canyon Dam. Historical ranges
extended into the upper reaches of the Columbia River (Lake
Windermere) and into the Snake River basin up to Shoshone
Falls. Resident trout species (rainbow, cutthroat, and bull trout)
occupy streams throughout the region, with bull trout generally
restricted to colder streams at higher elevations.

FRAMEWORK AND GUIDING QUESTIONS
Determining how a speciﬁc change in stream ﬂow or
temperature will impact a salmon population depends in part
upon species-speciﬁc tolerances and life history requirements, and in part upon the expected change in stream ﬂow
and temperature relative to those tolerances. The timing of
important salmon life history events varies both within and
among species (e.g. Groot and Margolis, 1991; Quinn,
2005; Figure 2). For example, salmonids with ocean-type
life histories (e.g. pink, chum, and some Chinook) tend to
spawn in the fall and winter, do not rear in freshwater during
summer, and migrate to sea in the winter or early spring.
Salmonids with stream-type life histories (e.g. coho salmon,

941

steelhead, and some Chinook salmon) spawn between fall
and late spring, rear in freshwater for 1 or 2 years, and usually
migrate to sea in spring or early summer. Hence, each species
and life history strategy will encounter a different suite of
stream ﬂow and temperature effects because they occupy
different habitats and vary in timing of life history events.
Climate change effects will also vary among rivers (e.g. Mote
et al., 2003; Beechie et al., 2006; Rieman et al., 2007; Crozier
et al., 2008), adding additional complexity to understanding
how climate change will affect salmonid populations.
Because there are many possible combinations of climate
change effects and life history responses to evaluate across
the study region, we do not attempt a comprehensive review
of all possible effects, nor do we use detailed population
models to estimate climate change effects on restoration
actions (e.g. Battin et al., 2007), mainly because many types
of restoration actions cannot be modelled with any certainty
and evaluation of hundreds of salmon populations is not
feasible (Bartz et al., 2006; Scheuerell et al., 2006). Rather,
we summarize key temperature thresholds by species and
life history stage, summarize climate change scenarios for
stream temperature and ﬂow, and allow local practitioners
to relate local climate change scenarios to locally relevant
salmonid tolerances. We also review the likely effectiveness
of various restoration techniques in a climate-altered future.

Figure 2. Timing of climate change effects on stream ﬂow and temperature by life history stages of ocean type Chinook salmon, coho salmon,

and steelhead
Copyright © 2012 John Wiley & Sons, Ltd.

River Res. Applic. 29: 939–960 (2013)
DOI: 10.1002/rra

 942

T. BEECHIE ET AL.

For example, actions that create new summer rearing habitats
in an area expected to exceed upper lethal temperature limits
will not likely improve salmonid populations over the long
term. By contrast, actions that signiﬁcantly reduce stream
temperature or create thermal refugia for the same species
are more likely to retain their effectiveness in a future
with increasing temperatures. Therefore, we reviewed recent
literature to evaluate whether speciﬁc restoration action types
will likely ameliorate climate change effects on ﬂood ﬂows,
low ﬂows, or stream temperature.
We also developed four guiding questions to evaluate the
potential impacts of climate change on project prioritization
and design:
(1) What habitat restoration actions are necessary for recovery
of local salmon populations?
(2) Do future stream ﬂow and temperature scenarios alter
the types of habitat restoration actions that are necessary
for recovery?
(3) Does the restoration plan or action ameliorate a predicted
climate change effect on stream ﬂow or temperature?
(4) Will the restoration plan or action increase habitat diversity and salmon population resilience?
Systematic consideration of these questions will help determine whether restoration objectives or priorities should be
altered to accommodate future climate change. Answering
the ﬁrst question requires local information about restoration
plans and objectives, which restoration planners and practitioners can acquire from salmon recovery plans developed
under the Endangered Species Act. However, information
required to answer the last three questions is rarely readily
available to restoration groups. Therefore, in the following
sections, we summarize information needed to address
questions 2–4 for salmon restoration actions in the PNW.

SCENARIOS FOR CLIMATE CHANGE EFFECTS ON
STREAM FLOW AND TEMPERATURE IN THE
PACIFIC NORTHWEST
Recent climate change scenarios modelled for the PNW
suggest a clear warming trend, but the magnitude of estimated
temperature increase varies with choice of climate model and
emissions scenario (Elsner et al., 2010). By contrast, even the
sign of future precipitation changes is not consistent among
different scenarios, with some predicting precipitation increases
and others predicting decreases (e.g. Elsner et al., 2010). A
multi-model averaged climate change scenario under A1B
emissions indicates an average temperature increase in the
PNW of 3.5  C by 2080, with wetter winters, drier summers,
and an increase in average annual precipitation of 5% (Elsner
et al., 2010; Mote and Salathé, 2010). Because all the future
Copyright © 2012 John Wiley & Sons, Ltd.

climate scenarios evaluated by Elsner et al. (2010) predict
warming trends, the models predict that more precipitation will
fall as rain and less as snow and that this effect will be most
pronounced in mid-elevation areas (Hamlet and Lettenmaier,
1999). In this paper, we use this multi-model average from
Elsner et al. (2010) to drive a coupled stream ﬂow and
temperature model (Whited et al., in press) to produce
scenarios of stream temperature and ﬂow regimes that
may have signiﬁcant impacts on salmon populations and
food webs that support them.
Although the A1B multi-model average is commonly
considered to be an informative future climate scenario (i.e. it
is closest to most model estimates and the weighting scheme
discounts extreme values; Mote and Salathé, 2010), there
remains considerable uncertainty around any estimate of future
precipitation or air temperature. Uncertainties around future
temperature and precipitation predictions have three main
sources: (1) the factors that force climate change (including
future greenhouse gas and aerosol emissions), (2) global
climate model (GCM) errors, and (3) ‘natural’ variability in
the climate system (Deser et al., 2010; Hawkins and Sutton,
2011). In general, the variation in temperature or precipitation
predictions among different emissions scenarios is smaller than
the variation among different GCMs (Mote and Salathé, 2010).
For example, the multi-model average predictions of PNW
climate for three emissions scenarios in the 2040s indicate a
temperature increase of 1.7–2.4  C and a precipitation increase
of 1–2% (Figure 3); the 2080s simulations indicate a
2.7–4.7  C increase in temperature and 3–6% increase in
precipitation. However, the variation among GCMs for the
2040s is roughly 2  C for each emissions scenario (compared
with a range of <1  C among emission scenarios) and as high
as 3.5  C among GCMs in the 2080s (compared with ~2  C
among emissions scenarios). Variation among GCMs is even
greater for precipitation predictions, with a range as high as
À8% to +23% for precipitation by the 2080s (compared with
3–6% between emissions scenarios). The combined emissions
scenario, model uncertainties, and natural variability for air
temperature in the PNW suggest an increase of 1–3  C by the
2040s and 2–6  C by the 2080s (multi-model averages of 2  C
and 3.5  C, respectively). The combined uncertainties for
precipitation suggest a À6% to +14% change in precipitation
by the 2040s and a À8% to +23% change by the 2080s
(multi-model averages of +2% and +5%, respectively). Finally,
both air temperature and precipitation are expected to continue
increasing through the end of the 21st century regardless of the
GCM used, and our use of climate scenarios for the 2080s is not
intended to suggest a stable future climate.
Stream ﬂow and temperature methods
Changes in stream ﬂow and temperature were simulated
using a two-step modelling process that (1) predicted daily
River Res. Applic. 29: 939–960 (2013)
DOI: 10.1002/rra

 RESTORING SALMON HABITAT FOR A CHANGING CLIMATE

943

model was based on a hierarchical dominant river tracing
algorithm that deﬁnes the underlying hydrography for stream
ﬂow and temperature calculations (Wu et al., 2011). The
coupled model, called the dominant river tracing-based stream
ﬂow and temperature model, produces gridded daily stream
ﬂow and stream temperature data on the basis of water and heat
transport in river networks, thermal dynamics of stream water
and the surrounding environment, and the coupling of hydrologic routing processes and associated thermal dynamics
(Whited et al., in press). Stream ﬂow and temperature scenarios
were based on the multi-model average future climate scenario
described previously that provided daily gridded precipitation
and air temperature data at one-sixteenth degree resolution
(Elsner et al., 2010). Only cells with ﬂow accumulation areas
greater than six upstream cells (equivalent to a drainage area
of approximately 200 km2) were included in the regional mapping because smaller drainage areas were not considered reliable for future scenarios (Wu and Kimball, unpublished data).
Stream ﬂow and temperature were calibrated to measured
stream ﬂow and temperatures at seven US Geological
Survey sites in the Columbia River basin and then validated
against an independent 10-year record of daily stream ﬂows
at 12 gauges and a 7-year record of daily stream temperatures
at 11 gauges. Model validation indicated strong correlation
between measured and simulated stream ﬂow and temperature
at the majority of gauges (Figure 4). Although there was no
consistent positive or negative bias in either stream ﬂow
or temperature, some deviations between modelled and
measured stream ﬂow or temperature are likely due to differences between the way VIC models runoff, groundwater,
and stream ﬂow compared with local physical processes. For
example, in the Willamette basin, groundwater discharge

Figure 3. Variation in global climate model (GCM) predictions

of precipitation change (%) and temperature increase (  C) in the
Paciﬁc Northwest. Circles indicate ensemble model averages for
each of three emissions scenarios, and lines indicate range of
predictions from 20 different GCMs for each emissions scenario.
Based on data from Mote and Salathé (2010)

runoff using the macroscale variable inﬁltration capacity
(VIC) model and (2) dynamic runoff routing, stream ﬂow
and stream temperature simulations based on VIC that
estimate water balance, energy balance, and runoff outputs.
The VIC model produces daily runoff and soil moisture, as
well as associated forcing variables including incoming shortwave and long wave radiation that are used later in the stream
temperature model. The coupled stream ﬂow and temperature
Copyright © 2012 John Wiley & Sons, Ltd.

Figure 4. Summary of dominant river tracing-based stream ﬂow and

temperature model validation results, illustrating distribution of
correlation coefﬁcients (R) for modelled versus measured stream
ﬂow and temperature. Heavy horizontal line indicates median R,
box indicates 25th and 75th percentiles, lines indicate 10th and
90th percentiles, and circles indicate range of values. (Wu and
Kimball, unpublished data)
River Res. Applic. 29: 939–960 (2013)
DOI: 10.1002/rra

 944

T. BEECHIE ET AL.

from the fractured basalts of the Cascade Mountains leads to
higher summer stream ﬂow and lower summer temperature
than predicted by the dominant river tracing-based stream
ﬂow and temperature model, primarily because the VIC
model does not effectively model deep groundwater storage
and its inﬂuence on ﬂow. Therefore, the VIC model
underestimates summer stream ﬂow, which in turn leads
to an overestimate of summer stream temperature. Other
studies have found regional biases in mean annual runoff
associated with arid regions or snowmelt systems and
showed overall underestimation of mean annual stream
ﬂow in the PNW from VIC modelling (Gangopadhyay
and Pruitt, 2011). However, validation of stream ﬂows for
this study did not suggest a clear spatial pattern of positive
or negative biases.
We focused our ﬂow analysis on mean monthly ﬂows
for four periods: 1970–1999, 2000–2029, 2030–2069, and
2070–2099. We calculated change in magnitude of the
maximum and minimum monthly ﬂows between periods for
each stream cell, as well as the change in timing of maximum
and minimum monthly ﬂows between periods. We focused on
predicted change in ﬂow relative to the modelled historical
baseline rather than absolute stream ﬂows to minimize
impacts of errors in the ﬂow model on our results. That is,
we assume that biases in the stream ﬂow model will be in
the same direction for all periods within a grid cell and that
using the change in ﬂow as our primary metric will reduce
the impact of those biases on our analysis. Finally, we used
cluster analysis to map three ﬂow regimes (rainfall-dominated,

snowmelt-dominated, and transitional) during each period
(Beechie et al., 2006).
We chose to map temperature predictions directly to
allow the greatest ﬂexibility in biological interpretations.
That is, we avoided selection of speciﬁc thermal limits
because thermal tolerances vary considerably among species
for each life stage. To aid in biological interpretation of
these temperature maps, we provide both species-speciﬁc
and generalized salmonid thermal limits (Table I), as well
as published temperature criteria that are recommended for
protection of Paciﬁc salmon from negative temperature
effects (Table II). Most upper lethal limits are between 20
and 24  C and recommended temperature thresholds for the
7-day average daily maximum range from 13 to 18  C. In
the absence of local data on thermal tolerances of salmonids
(which vary among species and environments), these data
can be used to gauge the likelihood that stream temperature
changes will be signiﬁcant for local species.
Results: stream ﬂow change scenario
Areas of the PNW with a snowmelt-dominated hydrologic
regime (in which the maximum monthly ﬂows are during
the spring snowmelt) shrink considerably under the ensemble
climate change scenario as snow level rises across the region
(Figure 5). By 2070–2099, the snowmelt hydrologic regime
no longer exists in the north Cascades and upper Snake River
basin, and the only remaining snowmelt-dominated area is in
the Canadian Rockies. The transitional regime, which has

Table I. Temperature thresholds (  C) for critical parts of the salmonid life cycle, including general and species speciﬁc information
Life stage

Chinook
(O. tshawytscha)

Adult migration
Optimal threshold
Lethal threshold
22
Thermal blockage
Adult holding and spawning
Optimal threshold
14.5
Detrimental to internally
held gametes
Incubation and early fry development
Upper threshold
14.5
Juvenile rearing
Optimal threshold
14.8a
Lethal threshold
24
UZNGb
Smoltiﬁcation
Impairment threshold
12–17

Chum
(O. keta)

Coho
(O. kisutch)

21

15.6
21

Pink
(O. gorbuscha)

Sockeye
(O. nerka)

Steelhead
(O. mykiss)

22
12.8

15.6
20

10

12

15
21
19.8

17
23
23.4
15

12.8

12

12.5

12
19

20
21

24
13

Temperatures cited are for constant exposure, unless otherwise noted. Data compiled from Bjornn and Reiser (1991), Eaton and Scheller (1996), McCullough
et al. (2001), and Richter and Kolmes (2005).
a
Natural rations level.
b
Upper zero net growth (UZNG) temperature: maximum weekly temperature at which ﬁsh can live for several days but at which they do not ingest enough food
to gain weight.
Copyright © 2012 John Wiley & Sons, Ltd.

River Res. Applic. 29: 939–960 (2013)
DOI: 10.1002/rra

 945

RESTORING SALMON HABITAT FOR A CHANGING CLIMATE

Table II. Recommended temperature criteria (upper thresholds;   C) for Paciﬁc salmon and steelhead
Life stage

Reference

Adult migration Spawning Incubation Juvenile rearing

Smoltiﬁcation
Steelhead
(non-steelhead) smoltiﬁcation

7-DADM

18

13

13

16

16

14

Weekly mean

16

10

10

15

15

12

Richter and Kolmes, 2005;
US EPA, 2003
Richter and Kolmes, 2005

Figure 5. Modelled hydrologic regime through time, based on cluster analyses of mean monthly ﬂows for each period

both spring snowmelt and fall–winter high ﬂows, shifts inland
and northwards, leaving only small areas of the transitional
regime in the north Cascades and the Rocky Mountains. The
rainfall-dominated regime, which has the highest ﬂows during
fall–winter ﬂoods, has historically been limited to the maritime
climate west of the Cascades and small low-elevation portion
Copyright © 2012 John Wiley & Sons, Ltd.

of the interior Columbia basin. By 2070–2099 however, the
rainfall-dominated regime expands to nearly the entire interior
Columbia basin in this climate change scenario.
The ensemble climate scenario suggests the largest decreases
in summer low ﬂows will be west of the Cascade Mountains,
where minimum monthly ﬂows decrease by 10–70% over the
River Res. Applic. 29: 939–960 (2013)
DOI: 10.1002/rra

 946

T. BEECHIE ET AL.

course of the 21st century (Figure 6). The largest changes are in
basins that currently have a transitional hydrologic regime (with
both a fall/winter storm peak and a late spring snowmelt peak),
and large decreases in minimum monthly ﬂows result from a
nearly complete loss of the spring snowmelt peak and a concomitant decline in late summer ﬂows. More modest decreases
in low ﬂow (10–35%) are predicted in the Rocky Mountains
south of the US–Canada border and in the Blue Mountains of
northeastern Oregon. In these areas, the hydrologic regime
shifts from snowmelt to transitional, and the decline in snowpack results in earlier spring melt and a decrease in late summer
stream ﬂows (Figure 6). Minimum monthly ﬂows in the Canadian Rockies increase by 10% or more, largely as a result of a
predicted increase in precipitation and snowpack. Increased
minimum ﬂows are also predicted in the upper Snake River
basin, caused by a shift in minimum monthly ﬂows from the
cold January–February period to minimum ﬂows in summer
following spring snowmelt. In this region, there is little predicted change in summer low ﬂows, but the minimum winter
ﬂows increase signiﬁcantly.
Simulated maximum monthly ﬂows increase by 10–50%
across most of the region as a result of an increasing fraction
of precipitation falling as rain rather than snow (Figure 7).
The few areas where maximum monthly ﬂows are expected
to increase by more than 50% are located in the Cascade
Mountains and in the middle and lower Snake River basin.
The large increases in the Cascade Mountains are predominantly a result of a shift from transitional to rainfalldominated hydrographs, with future ﬂood ﬂows in the fall
and winter being considerably larger than at present. Large
predicted increases in the monthly average peak stream ﬂow
in the Snake River tributaries result mainly from large
increases in spring precipitation. Overall, the ensemble
scenario suggests increasing volume of winter runoff and
increased ﬂooding in transitional basins and increasing spring
ﬂows in snow-dominant basins. Reductions in summer low
ﬂows are projected to be largest in the transitional basins in
the Cascade and Olympic Mountain ranges.
Results: stream temperature change scenario
Increased air temperatures will lead to increased water
temperatures on both the west and east sides of the
Cascade Mountains, and the scenario indicates a 1–4  C
increase in stream temperatures (maximum weekly mean
temperature) across the region by the 2030–2069 period
and a 2–6  C increase by the 2070–2099 period (Figure 8).
Highest mean weekly water temperatures vary signiﬁcantly
across the region in all periods, with highest temperatures
in reaches of the Snake and Willamette River basins
(Figure 9). Because these areas are close to or exceed
published thermal tolerances of most salmon species even
during the historical period (1970–1999), they are most
Copyright © 2012 John Wiley & Sons, Ltd.

likely to shift to stressful or lethal thermal conditions in
the future. Notably, many rivers within the current salmon
range have modelled temperatures above published lethal
or protective thresholds (Tables I and II), yet salmon
currently occupy the majority of these rivers. Stream
temperatures in the northern part of the Columbia basin
are currently within thermal tolerances of most rearing
juvenile salmonids, and under this climate change
scenario, projected temperature increases remain the lowest
in the region and within thermal tolerances. This area is
outside the current salmon range because of blocked migration
by dams, but within their historical range. Most coastal river
systems and rivers originating on the west slope of the
Cascade Mountains are likely to remain within published
thermal tolerances even in the 2070–2099 period. These river
systems have the smallest projected temperature increases,
whereas the largest increases are expected in the main stem
Columbia River and tributaries in the middle and lower
Columbia basin.

REVIEW OF RESTORATION ACTIONS AND
CLIMATE CHANGE
We grouped restoration actions on the basis of the watershed
processes or functions they attempt to restore (Beechie et al.,
2010) and then classiﬁed them as either likely or not likely to
ameliorate a climate change effect on high stream ﬂows, low
stream ﬂows, and stream temperatures (Table III). We classiﬁed actions on the basis of a literature review of restoration
action effectiveness and watershed processes to develop a
comprehensive summary of each action’s likelihood of
ameliorating climate change effects. Our basic rules were to
(1) classify an action as likely to ameliorate an effect if we
could ﬁnd literature support for that response and (2) avoid
including effects that were theoretically possible but not
supported by data. In a few cases, the literature was sparse
and suggested mixed effects depending on the context. In
those cases, we classiﬁed the action as having a contextdependent effect on stream ﬂow or temperature to indicate that
the ability of the action to ameliorate a climate effect depends
on the situation in which the action is employed. Although
these rules may omit a few effects, we felt that it was more
important to provide clear guidance on the dominant effects
and avoid including actions that only rarely would ameliorate
the climate change effect. Nevertheless, this review is not
intended to imply that less robust actions should be avoided
in all circumstances. For instance, where summer rearing
habitats constrain population recovery and summer stream
ﬂow and temperature are not expected to change signiﬁcantly,
any action that addresses causal factors for habitat and population declines should be implemented even if it does not
ameliorate a climate change effect. Only in cases where
River Res. Applic. 29: 939–960 (2013)
DOI: 10.1002/rra

 RESTORING SALMON HABITAT FOR A CHANGING CLIMATE

947

Figure 6. Modelled change in the minimum mean monthly ﬂow and shift in timing of minimum mean monthly ﬂow. Ratio is minimum mean

monthly ﬂow from the future period divided by the minimum mean monthly ﬂow from the period 1970–1999. For shift in low ﬂow timing,
areas mapped in red indicate a shift from minimum ﬂow in winter (usually February) to minimum ﬂow in August. Areas in yellow predominantly indicate a shift from minimum monthly ﬂow in September to August

Copyright © 2012 John Wiley & Sons, Ltd.

River Res. Applic. 29: 939–960 (2013)
DOI: 10.1002/rra

 948

T. BEECHIE ET AL.

Figure 7. Modelled change in the maximum mean monthly ﬂow and shift in timing of maximum monthly ﬂows. Ratio is maximum mean

monthly ﬂow from the future period 2070–2099 divided by the maximum mean monthly ﬂow from the period 1970–1999. For timing of
maximum monthly ﬂows, areas mapped in red indicate a shift from maximum monthly ﬂow during spring snowmelt (usually April or
May) to maximum ﬂows during December and January winter storms. Areas in yellow indicate areas with maximum monthly ﬂow remaining
in late winter or spring, but shifted 1–2 months earlier (generally April–May to February–March)

Copyright © 2012 John Wiley & Sons, Ltd.

River Res. Applic. 29: 939–960 (2013)
DOI: 10.1002/rra

 RESTORING SALMON HABITAT FOR A CHANGING CLIMATE

949

climate effects are expected to impact a project’s or plan’s
effectiveness within a few decades do we suggest adjustments
to project priorities or designs.
We also review restoration actions in the context of their
ability to maintain or increase resilience of river ecosystems
and salmon populations (Waples et al., 2009). We deﬁne
resilience as the ability of a system to absorb change and
still maintain its basic ecosystem functions and relationships, even though the balance of habitat types or species
may shift slowly through time (Holling, 1973; Waples
et al., 2009). Paciﬁc salmon are adapted to wide array of
natural disturbance regimes by virtue of their life history
diversity, and restoration actions designed to reduce constraints
on life history diversity allow Paciﬁc salmon a broader range of
options by which to respond to climate change (Waples et al.,
2008, 2009)—conferring resilience to both populations and
meta-populations (Greene et al., 2010; Schindler et al., 2010).
That is, restoration actions that increase habitat diversity to
the point that salmon have the ability to express alternative life
history strategies are considered to potentially increase population resilience. For example, restoring diverse ﬂoodplain
habitats or reconnecting cold-water tributaries to main stem
habitats by barrier removals offers salmon a variety of physical
and thermal conditions, allowing multiple species to persist and
to express varied life history strategies within species (Poole
et al., 2008; Waples et al., 2009). In contrast, creation of pools
by adding wood to a small stream generates a small increase in
habitat diversity but does not offer an array of habitats that
allow expression of alternative life histories. Hence, we
consider the former action to potentially increase resilience
but not the latter.
Restoring connectivity
Restoring connectivity (longitudinal, lateral, and vertical)
typically improves both physical and biological functions
of river systems. Restoring longitudinal connectivity for
salmon is primarily intended to reestablish salmon migration
to diverse habitats that have been lost through construction
of artiﬁcial barriers such as dams or culverts, but it often also
restores downstream transport of essential ﬂows, sediment,
and wood or organic matter. Restoring lateral connectivity
generally refers to reconnection of rivers to their ﬂoodplains
by removal of levees or bank armouring. These actions restore
the ability of the river system to create and sustain diverse
habitats and to allow migration of salmon into those habitats.
Actions that aim to restore vertical connectivity seek to
aggrade incised or scoured channels, which increases the
connection between surface and subsurface ﬂows and
increases ﬂoodplain connectivity over time.
Figure 8. Modelled increase in maximum average weekly temperature

through time in the Columbia River basin

Copyright © 2012 John Wiley & Sons, Ltd.

Longitudinal connectivity (barrier removal). The primary aims
of restoring longitudinal connectivity by removal of dams or
other blocking structures are to (1) reestablish upstream and
River Res. Applic. 29: 939–960 (2013)
DOI: 10.1002/rra

 950

T. BEECHIE ET AL.

Figure 9. Modelled maximum weekly temperature through time in the Columbia River basin

downstream ﬁsh migration pathways and (2) restore natural
stream ﬂow, sediment, and organic matter transport (Pess
et al., 2005). Removal of dams or providing ﬁsh passage on
dams that cannot be removed will allow ﬁsh to access
important upstream cool water habitats when downstream
areas become too warm (McClure et al., 2008; Pess et al.,
2008), thereby increasing habitat and life history diversity at
the population and meta-population scales (Beechie et al.,
2006; Waples et al., 2009). Where dams or other structures
contribute to reduced low ﬂows or increased stream
temperature, dam removal can also ameliorate low base ﬂow
and high temperature problems by restoring downstream
movement of sediment and water (Burroughs et al., 2009).
Lateral connectivity (ﬂoodplain reconnection). The aims of
re-establishing lateral connectivity between river channels
Copyright © 2012 John Wiley & Sons, Ltd.

and ﬂoodplains are often twofold: to restore river ﬂoodplain
dynamics that create diverse habitats and to restore ﬁsh
access to ﬂoodplain habitats (Pess et al., 2005; Waples et al.,
2009). These actions, which typically include reconnection
or creation of side channels and sloughs, removal or set
back of levees and dikes, and re-meandering of dredged or
straightened channels, can ameliorate peak ﬂow increases
by storing ﬂood water and reducing ﬂood peaks (Sparks
et al., 1998; McAlister et al., 2000) or by increasing the
availability of velocity and thermal refugia (Sommer et al.,
2001; Morley et al., 2005; Jeffres et al., 2008; Poole et al.,
2008). Similarly, removing levees or re-meandering
channels can ameliorate temperature increases by increasing
length of hyporheic ﬂow paths beneath the ﬂoodplain, which
can cool water during the summer (Arrigoni et al., 2008;
Konrad et al., 2008; Poole et al., 2008; Opperman et al.,
River Res. Applic. 29: 939–960 (2013)
DOI: 10.1002/rra

 951

RESTORING SALMON HABITAT FOR A CHANGING CLIMATE

Table III. Summary of restoration action types and their ability to ameliorate climate change effects on peak ﬂow, low ﬂow, stream
temperature, or to increase salmon population resilience

Category

Common
techniques

Longitudinal connectivity (barrier removal)
Removal or breaching of dam
Barrier or culvert replacement/removal
Lateral connectivity (ﬂoodplain reconnection)
Levee removal
Reconnection of ﬂoodplain features
(e.g. channels, ponds)
Creation of new ﬂoodplain habitats
Vertical connectivity (incised channel restoration)
Reintroduce beaver (dams increase
sediment storage)
Remove cattle (restored vegetation stores
sediment)
Install grade controls
Stream ﬂow regimes
Restoration of natural ﬂood regime
Reduce water withdrawals, restore
summer baseﬂow
Reduce upland grazing
Disconnect road drainage from streams
Natural drainage systems, retention ponds,
other urban stormwater techniques
Erosion and sediment delivery
Road resurfacing
Landslide hazard reduction (sidecast removal,
ﬁll removal)
Reduced cropland erosion (e.g. no-till seeding)
Reduced grazing (e.g. fencing livestock
away from streams)
Riparian functions
Grazing removal, fencing, controlled grazing
Planting (trees, other vegetation)
Thinning or removal of understory
Remove non-native plants
Instream rehabilitation
Re-meandering of straightened stream,
channel realignment
Addition of log structures, log jams
Boulder weirs and boulders
Brush bundles, cover structures
Gravel addition
Nutrient enrichment
Addition of organic and inorganic nutrients

Ameliorates
temperature
increase

●

Ameliorates
base ﬂow
decrease

Ameliorates
peak ﬂow
increase

Increases
salmon
resilience

●
●
●
●
●
●

○

○

●

○
○

●
●
●
●
●
●
●
●

○
○

●
●
●
●
●
●

○

●
●
●
●
●

○
○

○
○
◒

○

○
○
○

○

◒

◒

●
●

○
○
○

○
○

○
○

○
○

○
○

○

○
○

○
○

○
○

●
●

○

○
○
○
◒

○
○
○
○

○
○
○
○

◒

○

○

◒

◒
◒

○

○

○

○
○

○
○

○
○

○
○

○

○

○

○

◒

◒

◒

Actions are grouped by major processes or functions they attempt to restore: connectivity (longitudinal, lateral and vertical), watershed-scale processes
(stream ﬂow and erosion regimes), riparian processes, instream rehabilitation, and nutrient enrichment. Filled circles indicate positive effect, empty circles
indicate no effect, and partially ﬁlled circles indicate context-dependent effects. See text for supporting citations.

2010). Increasing habitat diversity by restoring lateral
connectivity generally allows for increased life history
diversity within a population (Waples et al., 2009), which
has been linked to increased population resilience (Greene
et al., 2010; Schindler et al., 2010). Floodplain reconnection
actions generally do not ameliorate base ﬂow decreases.
Copyright © 2012 John Wiley & Sons, Ltd.

Vertical connectivity (restoring incised channels). Channel
incision (or downcutting) has degraded stream and riparian
habitats in many rivers of the PNW by lowering water
tables, reducing exchange between surface and subsurface
ﬂows, and decreasing late summer stream ﬂows. Associated
losses in riparian vegetation lead to reduced shading
River Res. Applic. 29: 939–960 (2013)
DOI: 10.1002/rra

 952

T. BEECHIE ET AL.

and organic matter inputs to streams and increased
stream temperatures (Elmore and Beschta, 1987; Poole and
Berman, 2001). Moreover, incised streams rarely access
their ﬂoodplains, high ﬂows are concentrated within the
incised channel, and ﬁsh have no access to slow-water
refugia during ﬂoods (Harvey and Watson, 1986; Elmore
and Beschta, 1987; Shields et al., 1995). Efforts to restore
incised streams by increasing sediment storage and aggrading
the channel aim to restore ﬂoodplain aquifer storage, increase
summer base ﬂow and decrease summer stream temperature,
and increase availability of ﬂood refugia (Pollock et al.,
2007; Beechie et al., 2008b). Some restoration techniques,
such as use of beaver dams to increase sediment storage,
have the added effects of increasing summer base ﬂows,
locally decreasing or buffering stream temperature and
increasing habitat diversity and productivity (Ponce and
Lindquist, 1990; McRae and Edwards, 1994; Pollock et al.,
2003; Pollock et al., 2007). Hence, restoration of incised
channels has the potential to ameliorate climate-induced
increases in stream temperature, effects on peak ﬂows and
low ﬂows, and also to increase life history diversity through
creation of off-channel and pond habitats. We emphasize that
ameliorating climate change effects through aggradation
techniques is speciﬁc to incised channels and that the same
techniques may have different responses in other settings. For
example, water may be warmer in a beaver pond within an
otherwise closed canopy system, although stream cooling
may still occur downstream of the pond (Robison et al., 1999).

Restoring stream ﬂow regimes
Flood ﬂows are increased to a moderate degree by logging
and forest roads (Jones and Grant, 1996), grazing effects
(Belsky et al., 1999), and to a much greater degree by
impervious surfaces in urban areas (Booth et al., 2002).
The primary mechanism by which logging roads increase
peak ﬂows is interception of subsurface ﬂow through soils
(which moves relatively slowly) and rapid routing of water
to streams through ditches (Furniss et al., 1991; Jones and
Grant, 1996). Road rehabilitation actions to decrease peak
ﬂow effects generally focus on addition of cross-drains to
reduce routing of water directly from road ditches to the
stream (Furniss et al., 1991). In an urban environment, the
primary focus is to reduce the impacts of impervious surfaces
by creating additional stormwater retention structures or
modifying impervious surface areas so that rapid runoff is
routed into groundwater storage rather than storm drains
(e.g. Booth and Leavitt, 1999). In many cases, increased
runoff and ﬂood ﬂows cause summer baseﬂows to decrease
due to loss of inﬁltration and water storage in soils (e.g. Belsky
et al., 1999). Hence, reductions of grazing or logging effects
on ﬂood ﬂows may also increase low ﬂows in summer.
Copyright © 2012 John Wiley & Sons, Ltd.

Low stream ﬂows are often reduced by withdrawal of
water from streams for irrigation or consumptive uses (Poff
et al., 1997; Myers et al., 1998), and both peak and low
ﬂows may be dramatically reduced by water storage behind
dams (Stanford et al., 1996; Poff et al., 1997). Restoring
some or all of abstracted water to streams through purchase
of water rights or increased irrigation efﬁciency can dramatically increase low ﬂows to streams (Poff et al., 2010) and
directly ameliorate climate-induced decreases in low stream
ﬂow or increased stream temperature. In some cases, ﬂow
regulation has decreased peak ﬂows to the point that many
geomorphological and ecological functions of streams are
lost (Olden and Poff, 2004). Moreover, low ﬂows may be
reduced in summer, which can also lead to increased stream
temperature. In such cases, restoring ‘environmental ﬂow
regimes’ can ameliorate not only low stream ﬂows, but can
also increase habitat diversity by restoring channel-forming
ﬂows that maintain habitat diversity and other ecological functions (Stanford et al., 1996; Poff et al., 2010). Hence, where
water storage or withdrawal has decreased low ﬂows, purchase of water rights or use of water conservation measures
that leave more water in the stream can ameliorate predicted
decreases in low ﬂows. Some dams can release cool water
from deep in the reservoir, allowing dam operations to
ameliorate stream temperature increases. Where water storage
has decreased peak ﬂows, restoration of channel forming
ﬂows can increase habitat diversity through restoration of
physical functions that create diverse habitat features in
streams and across the ﬂoodplain and also maintain riparian
functions (Poff et al., 1997)—thereby increasing resilience
of river ecosystems to climate change.
Reducing erosion and sediment delivery
In forested environments of the PNW, sediment supply to
stream channels is typically increased through surface
erosion on unpaved roads or by increased landsliding from
roads or clearcuts (Reid and Dunne, 1984; Sidle et al.,
1985). Therefore, sediment reduction efforts in forest
environments commonly focus on road rehabilitation to
decrease landslide hazards and surface erosion (Beechie
et al., 2005; Roni et al., 2008). Landslide hazard reduction
is typically achieved by removing or abandoning roads, or
rebuilding stream crossings to avoid ﬁll failures when
culverts become blocked (Madej, 2001; McCaffery et al.,
2007). Despite these efforts, future increases in storm intensity
and a shift from snow to rainfall may drive more frequent
mass wasting in forest environments, especially where road
management has not yet achieved reductions in landslide
hazard. Effects of increased surface erosion on roads can be
abated by resurfacing the road or adding cross drains or water
bars to prevent delivery of eroded sediments to streams
(Furniss et al., 1991).
River Res. Applic. 29: 939–960 (2013)
DOI: 10.1002/rra

 RESTORING SALMON HABITAT FOR A CHANGING CLIMATE

In croplands, surface erosion is often increased by erosion
of exposed soil in fallow ﬁelds (Wendt and Burwell, 1985;
Ebbert and Roe, 1998). An increasingly common strategy
to manage surface erosion in agricultural lands is no-till
seeding, which preserves vegetative cover on croplands
and dramatically reduces erosion and sediment delivery to
streams (Wendt and Burwell, 1985; Ebbert and Roe,
1998). Grazing effects on sediment supply include removal
of hillslope vegetation and erosion of exposed soils, as well
as trampling of banks and increased bank erosion (Medina
et al., 2005). Grazing impacts can be controlled either by
removal of livestock from key areas (especially stream banks
and riparian areas) or by grazing rotations that retain sufﬁcient
vegetative cover to reduce surface erosion (Medina et al.,
2005). Although each of these actions can improve stream
habitat by decreasing ﬁnes in the stream bed, increasing pool
depth, or narrowing widened channels—none of these actions
ameliorate decreased low ﬂows, increased ﬂood magnitude, or
increased stream temperature (although increased pool depth
may create thermal refugia in rare cases). Moreover, these
actions do little to increase habitat or life history diversity
except in cases where extremely high sediment supply has
ﬁlled pools and reduced the diversity of habitat types
(see Beechie et al., 2005 for examples).
Restoring riparian functions
Riparian rehabilitation actions aim to restore riparian functions
such as stream shading, root reinforcement of banks, supply
of large wood and organic matter, and trapping sediment or
ﬁltering nutrients (Kauffman et al., 1997; Pollock et al.,
2005). In forested environments, restoration of riparian
functions commonly focuses on thinning or replanting of
riparian forests to restore wood recruitment and shade
functions and, secondarily, to restore other functions
(Beechie et al., 2000; Welty et al., 2002; Meleason et al.,
2003). Restored riparian functions do not directly ameliorate the stream ﬂow changes predicted by climate change
models, but may mitigate stream temperature increases via
increased shading (Johnson, 2004), or via increased wood
recruitment and sediment storage in headwater channels
that have been scoured to bedrock (Pollock et al., 2009).
However, removal of certain non-native species that use
more water than native species and provide less shade
can ameliorate increased stream temperatures or decreased
ﬂows. In non-forest environments, replanting of denuded
or managed riparian zones and removal or reduction of
livestock grazing typically results in regrowth of riparian
vegetation and should also ameliorate increases in stream
temperatures through increased shade, bank stability, and
narrowing of stream channels (Medina et al., 2005).
Riparian restoration can lead to modest increases in habitat
diversity over the long term via formation of pools or
Copyright © 2012 John Wiley & Sons, Ltd.

953

hiding cover (Beechie et al., 2000), whereas actions that
seek to thin riparian zones are unlikely to affect either
stream ﬂow or temperature (Pollock et al., 2009). Finally,
restoration of normative ﬂow regimes on regulated rivers
should help recovery of riparian areas on larger rivers, as
seedling establishment for key riparian species is often
dependent on ﬂood magnitudes and duration (Stanford
et al., 1996; Mahoney and Rood, 1998). Riparian restoration can be expected to increase ecosystem resilience
in the sense that rivers with intact riparian buffers can
buffer ecological functions against changes in stream ﬂow,
but it is unlikely to increase life history diversity and
salmon resilience beyond the buffering of temperature
effects.

Instream rehabilitation
Instream rehabilitation includes restoration actions that
seek to improve habitat conditions by actively altering
channel habitat structure (e.g. adding wood debris, spawning
gravel), reconstructing channel characteristics (re-meandering),
or by providing cover for ﬁsh (Roni et al., 2008). Such ﬁxed
structures are susceptible to failure or require maintenance,
especially in the face of increased magnitude and frequency
of peak ﬂow events as predicted by climate change models.
Although instream rehabilitation actions such as wood and
boulder placement have been documented to provide quick
improvements in both physical habitat and ﬁsh production
(Cederholm et al., 1997; Solazzi et al., 2000; Roni and
Quinn, 2001), they do not restore the underlying disrupted
process (typically large wood delivery) and are unlikely to
last more than one or two decades without additional intervention or maintenance (Roni et al., 2002). Moreover,
instream rehabilitation actions generally do not ameliorate
changes in temperature, base ﬂow, or peak ﬂows. For
example, some studies have shown that creation of poolrifﬂe sequences can lead to increased hyporheic exchange
and increased temperature variability, but none has shown
a signiﬁcant net decrease in stream temperature (Crispell
and Endreny, 2009; Hester et al., 2009). This is most likely
because the subsurface ﬂow path is too short to signiﬁcantly
affect stream temperature (Poole et al., 2008). By contrast,
restoring sediment storage to channels that are incised to
bedrock may reduce stream temperatures if the loss of sediment has completely eliminated hyporheic exchange and
increased stream temperature (Pollock et al., 2009). In such
cases, use of wood or boulder structures to store sediment
may decrease stream temperature. Finally, instream habitat
actions can increase local habitat complexity (particularly if
a large portion of the stream is treated), but such actions are
unlikely to increase life history diversity or resilience of
salmon populations.
River Res. Applic. 29: 939–960 (2013)
DOI: 10.1002/rra

 954

T. BEECHIE ET AL.

Nutrient enrichment
The purpose of nutrient enrichment is to compensate for
lack of marine-derived nutrients from reduced salmon
returns by adding nutrients and carbon to boost stream or
lake productivity, and ultimately ﬁsh production (Bilby
et al., 1998; Gresh et al., 2000; Kiffney et al., 2005). These
exogenous sources of nutrients are important to the productivity of naturally oligotrophic rivers of the PNW where
many salmonid populations are food-limited (e.g. Boss and
Richardson, 2002). As with instream rehabilitation actions,
nutrient additions do not address the ultimate cause of low
nutrient levels as a result of reduced salmon runs and, in
the absence of increased salmon returns, are dependent on
continually adding nutrients to maintain any beneﬁts (Roni
et al., 2008). Nutrient additions do not ameliorate climate
change effects on stream ﬂow, stream temperature, or habitat
diversity. However, an important secondary effect of
increased stream temperature is increased metabolism in
juvenile ﬁshes, which increases food requirements to maintain
positive growth (McCullough et al., 2001; Boughton et al.,
2007). Where reduced nutrients and food resources have
already compromised growth of juvenile salmonids, rehabilitation actions to increase nutrient supply—thereby increasing
invertebrate abundance and prey availability for juvenile
salmonids—may indirectly ameliorate temperature effects
on salmonid growth rates (Wipﬂi and Baxter, 2010).
However, this would require a consistent, long-term
nutrient supplementation programme and would not
lead to self-sustaining nutrient levels without continual
intervention.

DECISION SUPPORT STRUCTURE TO EVALUATE
CLIMATE CHANGE EFFECTS ON SALMON
RESTORATION
We proposed four guiding questions to help determine
whether restoration plans or actions should be altered to
accommodate climate change (ﬁrst column of Figure 10).
We have also provided key maps and information to answer
these questions (Figures 3–7, Tables 1–3). However, translation of this information into adaptation of restoration plans
or actions can be ambiguous, so we offer two simple decision
support tools to assist in evaluating restoration plans or actions
in the context of climate change. Both are simple ﬂow charts
that illustrate how answers to the guiding questions might lead
to logical adaptations of restoration plans or actions. We do
not intend these to be rigid protocols with predeﬁned
outcomes because there are many possible combinations of
future climates, restoration strategies, and species responses,
and it is difﬁcult to arrive at a set of rules that will apply to
all possible cases. Rather, we intend these tools to illustrate
how answers to the guiding questions can be integrated to
Copyright © 2012 John Wiley & Sons, Ltd.

arrive at management decisions in the context of local goals
and objectives, as well as in the context of local climate
change scenarios.
Evaluating a salmon recovery plan
In most cases, the ﬁrst of the guiding questions has been
answered in the process of developing local salmon recovery plans (e.g. Shared Strategy Development Committee,
2007). That is, an important component of salmon recovery
plans is the identiﬁcation of habitat impairments that
constrain salmon population growth, and which, if addressed,
will increase abundance or productivity (population growth
rate) of the population (McElhany et al., 2000; Beechie
et al., 2003). From this analysis, a list of important habitat
recovery actions can be developed and prioritized (Beechie
et al., 2008a). However, these lists are commonly developed
without consideration of future climate change effects on habitats and therefore have not considered how climate change
might alter the suite of restoration actions identiﬁed as necessary to achieve salmon recovery.
Evaluating whether potential climate change effects on
stream ﬂow or temperature will change the list of restoration
actions necessary for salmon recovery (question 2) begins a
decision tree that helps restoration planners determine
whether a salmon recovery plan should be revised to accommodate future climate change effects (Figure 10). Answering
this question requires examination of potential climate change
effects on stream ﬂow and temperature (Figures 3–6) and a
qualitative assessment of whether future stream ﬂows or
temperatures are likely to alter conclusions about which
habitat restoration actions are necessary for salmon recovery.
If climate change effects on stream ﬂow or temperature are
not expected to change the types or priority of restoration
actions, then restoration actions may proceed according to
the current plan. An important caution is that any assessment
using mapped scenarios of ﬂow or temperature changes
should recognize that the resolution of the maps is quite low
(in our analysis, the smallest grid cell representing a stream is
6 Â 6 km) and that there is considerable variation in predicted
stream ﬂow and temperature changes among emissions scenarios
and GCMs (Elsner et al., 2010). Therefore, any evaluation of
these stream ﬂow and temperature scenarios should acknowledge
that uncertainty in the climate scenarios is high.
If the evaluation of climate change scenarios indicates a
change in the types of actions needed for recovery, then
planned actions should be evaluated to determine whether
they ameliorate the local climate change effects (question
3). For example, a coho salmon population may currently
be constrained by winter rearing habitat availability, but
potential increases in summer stream temperatures and
decreases in summer stream ﬂows (Figure 9) may reduce
summer rearing habitat availability to the point that it
River Res. Applic. 29: 939–960 (2013)
DOI: 10.1002/rra

 RESTORING SALMON HABITAT FOR A CHANGING CLIMATE

955

Figure 10. Schematic diagram of the decision-making process for adapting salmon recovery plans or individual restoration actions to climate

change, and relationship to the four guiding questions

becomes the primary constraint on a population’s recovery.
If the planned restoration actions are likely to ameliorate
climate change effects on stream ﬂow or temperature (from
Table III) and preserve the anticipated effectiveness of the
restoration plan, then restoration can also proceed without
modiﬁcation. However, if actions will not ameliorate the
climate change effects, revisions to the restoration plan
should be considered. We acknowledge that determining
whether the plan will ameliorate a temperature effect enough
to prevent stream temperatures from exceeding critical
thresholds is extremely difﬁcult. In most cases, available
data and models are not sufﬁcient to answer this question
quantitatively and with conﬁdence. However, most restoration actions are not expected to reduce stream temperatures
by more than 1–2  C (e.g. Medina et al., 2005; Arrigoni
et al., 2008), so we suggest as a rule of thumb that restoration
actions in areas that are less than 1–2  C below a critical
threshold be considered unlikely to ameliorate a climate
change effect. Nevertheless, it is possible that some combinations of actions might reduce stream temperature by more
than 2  C, and local experience with restoration actions and
changes in stream ﬂow or temperature should be considered
in management decisions. Moreover, salmon are often
adapted to higher temperatures than are typically reported in
Copyright © 2012 John Wiley & Sons, Ltd.

the literature, and data on local thermal tolerances of salmon
should be considered.
When revision of a plan is warranted, the degree to which
the plan should be revised depends on whether the proposed
actions contribute to increasing resilience of the population
(question 4). Where the main habitat restoration actions in
a plan contribute signiﬁcantly to increasing resilience of
populations (e.g. by increasing habitat diversity; Waples
et al., 2009), the plan may be followed with the understanding that climate change may reduce effectiveness of the
habitat restoration plan over the long term. There will
always be considerable uncertainty about whether the
actions can increase resilience enough to allow population
recovery. But in any case, restoring diverse habitats will
increase resilience of the riverine ecosystem—thereby
increasing the likelihood that a salmon population can
recover under a warming climate. Where the main habitat
restoration actions do not contribute signiﬁcantly to increasing
habitat diversity or resilience of a population, then the restoration plan should be revised to increase the likelihood
that actions either ameliorate climate change effects or
increase habitat diversity and ecosystem resilience. The
re-evaluation should focus on identifying actions that will
help the population recover under both existing and future
River Res. Applic. 29: 939–960 (2013)
DOI: 10.1002/rra

 956

T. BEECHIE ET AL.

limiting habitats, so that climate change effects on habitat
conditions and population performance do not hinder
recovery of the population. Restoration of key physical
and biological processes will allow a river ecosystem to
adjust naturally to changes in key ecosystem drivers such
as stream ﬂow and temperature and will be more robust to
variation in future climate patterns than actions that attempt
to control river behaviour or build speciﬁc habitat features
(Waples et al., 2009; Beechie et al., 2010). We stress that
even where detailed models are used to assess how climate
change will affect habitats that limit population recovery
(e.g. Battin et al., 2007), there will always be considerable
uncertainty in both model structure and climate change
scenarios. Therefore, we encourage adjustments to recovery
plans that broaden the portfolio of actions to accommodate a
wide range of potential future climate scenarios.
Although we did not model future stream ﬂow and
temperature for multiple emissions scenarios and climate
models, previous studies give us some indication of the
range of potential outcomes for air temperature and precipitation (e.g. Elsner et al., 2010, Mote and Salathé, 2010). On
the basis of those studies, it seems prudent to consider a
range of potential stream temperature increases at least
2  C higher than those predicted from our A1B ensemble
climate scenario (although we recognize that there is not
always a strong correlation between air temperature and
stream temperature). We cannot suggest a similar range of
values for stream ﬂows because no studies have modelled
variation in stream ﬂow among GCMs in the PNW
(although Elsner et al., 2010 modelled stream ﬂow differences
among two emissions scenarios). Nevertheless, uncertainty in
precipitation predictions is very high, suggesting that a conservative approach might anticipate that changes in stream ﬂow
(either high or low) might be considerably larger than our
map illustrates.
Finally, population status may also inﬂuence management
choices. Populations at very low abundance may require an
emphasis on near-term habitat recovery actions to stabilize
abundance (i.e. longevity of actions may be only 10–20 years),
whereas more stable populations may beneﬁt more from
restoration of processes that persist for much longer periods.
For near-term actions, climate change will likely produce relatively small effects on habitat conditions, and plans that
emphasize near-term actions may need little revision regardless
of climate change threats. However, this is a relatively rare case
(most populations are threatened rather than endangered), and
emphasizing restoration of habitat-forming processes is more
likely to succeed over the long term (Beechie et al., 2010).
Evaluating an individual habitat restoration action
Individual restoration actions are perhaps simpler to evaluate
because their future effectiveness is primarily a function of
Copyright © 2012 John Wiley & Sons, Ltd.

their intended purpose, which can be very narrowly deﬁned.
Answering the four guiding questions leads to one of three
potential decisions: proceed with the action as planned,
redesign the project to accommodate climate change, or
abandon the project in favour of other projects that maintain
their effectiveness in the face of climate change. The ﬁrst
outcome—proceeding with the action as planned—is appropriate when climate change does not alter effectiveness or
longevity of the action, or when the action ameliorates the
climate change effect sufﬁciently to maintain project effectiveness in the future (Figure 10). Redesigning a project to accommodate climate change is most appropriate where the action
effectiveness is not reduced but longevity may be reduced as
a function of a climate change effect (e.g. if peak ﬂows will
increase then the project should be designed to accommodate
larger peak ﬂows than observed at present). Finally, it is
prudent to consider abandoning a speciﬁc project if climate
change will likely negate its intended purpose, the action does
not ameliorate the effect, and the action does not increase
diversity and resilience. In these cases, the action may be
abandoned in favour of other actions that will maintain their
effectiveness in the face of climate change.
In evaluating the potential effects of climate change on
individual restoration projects, it is ﬁrst necessary to know
which species and life stage the restoration action targets
in order to evaluate whether climate change alters the effectiveness of the action. For example, if an action is intended to
restore a winter rearing habitat, then changes to winter
stream ﬂows will be an important evaluation criterion
whereas summer stream temperatures may not. Once the
purpose of an action is identiﬁed, one can ask whether
climate change will alter action effectiveness. If the effectiveness is not altered, then the action can proceed pending evaluation of climate change effects on project longevity
(Figure 10). If the effectiveness is likely to be reduced, then
one should consider whether the action type signiﬁcantly
ameliorates climate-related changes in stream ﬂow and
temperature. Evaluating the ecological impacts of different
temperature change scenarios is relatively straightforward, as
each species and life stage has a relatively speciﬁc range of
thermal tolerances (Tables I and II), and temperature change
magnitudes (Figures 2 and 3) can be compared with those
tolerances to judge whether climate change will likely reduce
project effectiveness. Finally, it is also important to consider
how long the restoration action will last when determining
whether it will ameliorate the impacts of climate change.
Actions such as restoring ﬂoodplain connectivity or removing
migration barriers restore underlying watershed processes, can
last many decades, and will likely be the most effective longterm strategies for climate change because they both ameliorate climate change effects and increase habitat diversity and
resilience. Other actions that may last only two or three
decades without continued intervention will only provide
River Res. Applic. 29: 939–960 (2013)
DOI: 10.1002/rra

 957

RESTORING SALMON HABITAT FOR A CHANGING CLIMATE

short-term amelioration for climate change impacts and are
unlikely to appreciably increase resilience over the long term.
SUMMARY AND CONCLUSIONS
We developed a set of guiding questions and data to inform
adaptation of habitat restoration plans for salmonids in
northwestern USA. These same questions are applicable to
any salmon restoration effort, and—moreover—generally
applicable to restoration of many species or ecosystems.
Key elements of adapting any restoration strategy to climate
change include (1) understanding the current recovery
needs, (2) evaluating whether climate change effects will
likely alter those needs, (3) determining whether restoration
actions can ameliorate climate change effects, and (4) determining whether restoration actions can increase ecosystem
resilience. These components are not speciﬁc to salmon, nor
are they speciﬁc to aquatic species. These same questions
can be used for any ecosystem in which restoration actions
might need to be adapted to accommodate environmental
effects of climate change. The key questions that must be
answered for any adaptation strategy are as follows: Does
climate change alter restoration needs in the future? And can
restoration actions increase ecosystem resilience by reducing
climate change effects or increasing habitat diversity? When
these questions are applied to other species or environments,
data needs include an understanding of current restoration plans,
an assessment of how climate change might alter restoration
needs, data on likely environmental effects of climate change,
and a review of potential restoration actions to evaluate their
likely effectiveness under future climate scenarios.
Although habitat restoration can contribute to increasing
resilience of salmon populations to climate change, restoration
of freshwater habitats alone may not be enough for their
recovery. Climate change effects are imposed on top of a long
history of insults, including harvest and hatchery effects on
population status, a broad array of habitat losses that have
dramatically reduced salmon abundance in the western
USA, and continuing changes in ocean conditions that are at
least partly a result of climate change. These combined
constraints have reduced wild salmon populations to the point
that many have been listed under the US Endangered Species
Act. Hence, recovery of these populations may also require
adjustments to hatchery production and harvest levels that
impact wild populations, which we did not address here. In
combination, such actions will likely increase abundance and
diversity in wild populations, allowing them to adapt to a
changing climate.
ACKNOWLEDGEMENTS

We thank the Moore Foundation and National Marine Fisheries
Service for providing funding for this research. We also thank
Sarah Morley for her insightful review of the manuscript.
Copyright © 2012 John Wiley & Sons, Ltd.

REFERENCES
Arnell NW. 1999. Climate change and global water resources. Global
Environmental Change 9: S31–S49.
Arrigoni AS, Poole GC, Mertes LAK, O’Daniel SJ, Woessner WW,
Thomas SA. 2008. Buffered, lagged, or cooled? Disentangling hyporheic
inﬂuences on temperature cycles in stream channels. Water Resources
Research 44: W09418. doi: 10.1029/2007WR006480
Bartz KL, Lagueux K, Scheuerell MD, Beechie TJ, Haas A, Ruckelshaus
MH. 2006. Translating restoration scenarios into habitat conditions: an initial
step in evaluating recovery strategies for Chinook salmon (Oncorhynchus
tshawytscha). Canadian Journal of Fisheries and Aquatic Sciences
63: 1578–1595.
Battin J, Wiley MW, Ruckelshaus MH, Palmer RN, Korb E, Bartz KK, Imaki
H. 2007. Projected impacts of climate change on salmon habitat restoration.
Proceedings of the National Academy of Sciences 104: 6720–6725.
Beechie T, Pess G, Roni P, Giannico G. 2008a. Setting river restoration
priorities: a review of approaches and a general protocol for identifying
and prioritizing actions. North American Journal of Fisheries Management
28: 891–905.
Beechie TJ, Pollock MM, Baker S. 2008b. Channel incision, evolution and
potential recovery in the Walla Walla and Tucannon River basins, northwestern
USA. Earth Surface Processes and Landforms 33: 784–800.
Beechie TJ, Pess G, Kennard P, Bilby RE, Bolton S. 2000. Modeling recovery
rates and pathways for woody debris recruitment in northwestern Washington
streams. North American Journal of Fisheries Management 20: 436–452.
Beechie TJ, Ruckelshaus M, Buhle E, Fullerton A, Holsinger L. 2006.
Hydrologic regime and the conservation of salmon life history diversity.
Biological Conservation 130: 560–572.
Beechie TJ, Sear D, Olden J, Pess GR, Bufﬁngton J, Moir H, Roni P,
Pollock MM. 2010. Process-based principles for restoring river ecosystems. BioScience 60: 209–222.
Beechie TJ, Steel EA, Roni PR, Quimby E (eds). 2003. Ecosystem Recovery
Planning for Listed Salmon: An Integrated Assessment Approach for
Salmon Habitat. U.S. Dept. Commerce, NOAA Tech. Memo. NMFSNWFSC-58: Seattle, WA; 183.
Beechie TJ, Veldhuisen CN, Schuett-Hames DE, DeVries P, Conrad RH,
Beamer EM. 2005. Monitoring treatments to reduce sediment and
hydrologic effects from roads. In Monitoring Stream and Watershed
Restoration, Roni P (ed). American Fisheries Society: Bethesda; 35–65.
Belsky AJ, Matske A, Uselman S. 1999. Survey of livestock inﬂuences on
stream and riparian ecosystems in the western United States. Journal of
Soil and Water Conservation 54: 419–431.
Bilby RE, Fransen BR, Bisson PA, Walter JK. 1998. Response of juvenile coho
salmon (Oncorhynchus kisutch) and steelhead (O. mykiss) to the addition
of salmon carcasses to two streams in southwestern Washington, U.S.A.
Canadian Journal of Fisheries and Aquatic Sciences 55: 1909–1918.
Bjornn TC, Reiser DW. 1991. Habitat requirements of salmonids in
streams. In Inﬂuences of Forest and Rangeland Management on
Salmonid Fishes and Their Habitats, Meehan WR (ed.). American
Fisheries Society Special Publication 19: Bethesda; 83–138.
Booth DB, Leavitt JL. 1999. Field evaluation of permeable pavement
systems for improved stormwater management. Journal of the American
Planning Association 65: 314–325.
Booth DB, Hartley D, Jackson R. 2002. Forest cover, impervious-surface
area, and the mitigation of stormwater impacts. Journal of the American
Water Resources Association 38: 835–845.
Boss SM, Richardson JS. 2002. Effects of food and cover on the growth, survival,
and movement of cutthroat trout (Oncorhynchus clarki) in coastal streams.
Canadian Journal of Fisheries and Aquatic Sciences 59: 1044–1053.
Boughton DA, Gibson M, Yedor R, Kelley E. 2007. Stream temperature
and the potential growth and survival of juvenile Oncorhynchus mykiss
in a southern California Creek. Freshwater Biology 52: 1353–1364.
River Res. Applic. 29: 939–960 (2013)
DOI: 10.1002/rra

 958

T. BEECHIE ET AL.

Bryant MD. 2009. Global climate change and potential effects on Paciﬁc
salmonids in freshwater ecosystems of southeast Alaska. Climatic
Change 95: 169–193.
Burroughs BA, Hayes DB, Klomp KD, Hansen JF, Mistak J. 2009. Effects
of Stronach Dam removal on ﬂuvial geomorphology in the Pine River,
Michigan, United States. Geomorphology 110: 96–107.
Cederholm CJ, Bilby RE, Bisson PA, Bumstead, TW, Fransen, BR, Scarlett
WJ, Ward JW. 1997. Response of juvenile coho salmon and steelhead to
placement of large woody debris in a coastal Washington Stream. North
American Journal of Fisheries Management 17: 947–963.
Crispell JK, Endreny TA. 2009. Hyporheic exchange ﬂow around
constructed in-channel structures and implications for restoration design.
Hydrological Processes 23: 1158–1168.
Crozier LG, Hendry AP, Lawson PW, Quinn TP, Mantua NJ, Battin J,
Shaw RG, Huey RB. 2008. Potential responses to climate in organisms
with complex life histories: evolution and plasticity in Paciﬁc salmon.
Evolutionary Applications 1: 252–270.
Daly C, Gibson WP, Taylor GH, Johnson GL, Pasteris P. 2002. A knowledgebased approach to the statistical mapping of climate. Climate Research
22: 99–113.
Deser C, Phillips A, Bourdette V, Teng H. 2010. Uncertainty in climate
change projections: the role of internal variability. Climate Dynamics.
DOI: 10.1007/s00382–010–0977–x
Eaton JG, Scheller RM. 1996. Effects of climate warming on thermal
habitat in streams of the United States. Limnology and Oceanography
41: 1109–1115.
Ebbert JC, Roe RD. 1998. Soil erosion in the Palouse River basin: indications
of improvement. US Geological Survey, USGS Fact Sheet FS-069–98.
Elmore W, Beschta RL. 1987. Riparian areas: perceptions in management.
Rangelands 9: 260–265.
Elsner MM, Cuo L, Voisin N, Deems J, Hamlet AF, Vano JA, Mickelson
KEB, Lee SY, Lettenmaier DP. 2010. Implications of 21st century
climate change for the hydrology of Washington State. Climatic Change.
DOI: 10.1007/s10584–010–9855–0
Furniss MJ, Roelofs TD, Yee CS. 1991. Road construction and maintenance.
In Inﬂuences of Forest and Rangeland Management on Salmonid Fishes
and Their Habitats, Meehan WR (ed.). American Fisheries Society Special
Publication 19: Bethesda; 297–324.
Gangopadhyay S, Pruitt T. 2011. West-wide climate risk assessments: biascorrected and spatially downscaled surface water projections. Bureau of
Reclamation Technical Memorandum No. 86-68210-2011-01. U.S. Department
of the Interior, Bureau of Reclamation, Denver, Colorado.
Greene CM, Hall JE, Guilbault KR, Quinn TP. 2010. Improved viability of
populations with diverse life history portfolios. Biology Letters 6: 382–386.
Gresh T, Lichatowich J, Schoonmaker P. 2000. An estimation of historic
and current levels of salmon production in the northeast Paciﬁc ecosystem:
evidence of a nutrient deﬁcit in the freshwater systems of the Paciﬁc
Northwest. Fisheries 25: 15–21.
Groot C, Margolis L (eds). 1991. Paciﬁc Salmon Life Histories. University
of British Columbia Press: Vancouver, Canada.
Hamlet AF, Lettenmaier DP. 1999. Effects of climate change on hydrology
and water resources in the Columbia River basin. Journal of the
American Water Resources Association 35: 1597–1623.
Harvey M, Watson C. 1986. Fluvial processes and morphological thresholds
in incised channel restoration. Water Resources Bulletin 22: 359–36.
Hawkins E, Sutton R. 2011. The potential to narrow uncertainty in projections
of regional precipitation change. Climate Dynamics 37: 407–418. doi:
10.1007/s00382–010–0810–6
Hester ET, Doyle MW, Poole GC. 2009. The inﬂuence of in-stream structures
on summer water temperatures via induced hyporheic exchange. Limnology
and Oceanography 54: 355–367.
Holling CS. 1973. Resilience and stability of ecological systems. Annual
Review of Ecology and Systematics 4: 1–23.
Copyright © 2012 John Wiley & Sons, Ltd.

Jeffres CA, Opperman JJ, Moyle PB. 2008. Ephemeral ﬂoodplain habitats
provide best growth conditions for juvenile Chinook salmon in a California
river. Environmental Biology of Fishes 83: 449–458.
Johnson SL. 2004. Factors inﬂuencing stream temperature in small streams:
substrate effects and a shading experiment. Canadian Journal of Fisheries
and Aquatic Sciences 61: 913–923.
Jones JA, Grant GE. 1996. Peakﬂow responses to clear-cutting and roads in
small and large basins, western Cascades, Oregon. Water Resources
Research 32: 959–974.
Kauffman JB, Beschta RL, Otting N, Lytjen D. 1997. An ecological
perspective of riparian and stream restoration in the western United
States. Fisheries 22: 12–24.
Kiffney PM, Bilby RE, Sanderson BL. 2005. Monitoring the effects of
nutrient enrichment on freshwater ecosystems. In Monitoring Stream
and Watershed Restoration, Roni P (ed.). American Fisheries Society:
Bethesda; 237–266.
Konrad CP, Black RW, Voss F, Neale CMU. 2008. Integrating remotely
acquired and ﬁeld data to assess effects of setback levees on riparian
and aquatic habitats in glacial-meltwater rivers. River Research and
Applications 24: 355–72.
Lackey RT. 2003. Paciﬁc Northwest salmon: forecasting their status in
2100. Reviews in Fisheries Science 11: 35–88.
Lackey RT, Lach D, Duncan S. 2006. Salmon 2100: The Future of Wild
Paciﬁc Salmon. American Fisheries Society: Bethesda MD.
Madej MA. 2001. Erosion and sediment delivery following removal of
forest roads. Earth Surface Processes and Landforms 26: 175–190.
Mahoney JM, Rood SB. 1998. Streamﬂow requirements for cottonwood
seedling recruitment—an integrative model. Wetlands 18: 634–645.
Mantua N, Tolver I, Hamlet A. 2010. Climate change impacts on streamﬂow extremes and summertime stream temperature and their possible
consequences for freshwater salmon habitat in Washington State.
Climatic Change, [online], 1–37. DOI: 10.1007/s10584–010–9845–2
McAlister LS, Peniston BE, Leibowitz SG, Abbruzzese B, Hyman JB.
2000. A synoptic assessment for prioritizing wetland restoration efforts
to optimize ﬂood attenuation. Wetlands 20: 70–83.
McCaffery M, Switalski TA, Eby L. 2007. Effects of road decommissioning on stream habitat characteristics in the South Fork Flathead
River, Montana. Transactions of the American Fisheries Society
136: 553–561.
McClure MM, Carlson SM, Beechie TJ, Pess GR, Jorgensen JC, Sogard SM,
Sultan SE, Holzer DM, Travis J, Sanderson BL, Power ME, Carmichael
RW. 2008. Evolutionary consequences of habitat loss for Paciﬁc anadromous Salmonids. Evolutionary Applications 1: 300–318.
McCullough DA, Spalding S, Sturdevant D, Hicks M. 2001. Issue Paper 5.
Summary of technical literature examining the physiological effects of
temperature on salmonids. EPA-910-D-01-005, U.S. EPA Region 10;
Portland OR.
McElhany P, Ruckelshaus M, Ford M, Wainwright T, Bjorkstedt E. 2000.
Viable salmonid populations and the recovery of evolutionarily signiﬁcant
units. U.S. Dept. Commerce, NOAA Tech, Memo. NMFS-NWFSC-42,
Seattle.
McRae G, Edwards CJ. 1994. Thermal characteristics of Wisconsin streams
occupied by beaver: implications for brook trout habitat. Transactions of
the American Fisheries Society 123: 641–656.
Medina AL, Rinne JN, Roni PR. 2005. Riparian restoration through grazing
management: considerations for monitoring project effectiveness. In
Monitoring Stream and Watershed Restoration, Roni P (ed.). American
Fisheries Society: Bethesda; 97–126.
Meleason MA, Gregory SV, Bolte JP. 2003. Implications of riparian management strategies on wood in streams of the Paciﬁc Northwest. Ecological
Applications 13: 1212–1221. doi: 10.1890/02–5004
Morley SA, Garcia PS, Bennett TR, Roni P. 2005. Juvenile salmonid
(Oncorhynchus spp.) use of constructed and natural side channels in
River Res. Applic. 29: 939–960 (2013)
DOI: 10.1002/rra

 RESTORING SALMON HABITAT FOR A CHANGING CLIMATE

Paciﬁc Northwest rivers. Canadian Journal of Fisheries and Aquatic
Sciences 62: 2811–2821.
Mote PW, Salathé EP. 2010. Future climate in the Paciﬁc Northwest.
Climatic Change. DOI: 10.1007/s10584–010–9848–z
Mote PW, Parson EA, Hamlet AF, Keeton WS, Lettenmaier D, Mantua N,
Miles EL, Peterson DW, Peterson DL, Slaughter R, Snover AK. 2003.
Preparing for climatic change: the water, salmon, and forests of the
Paciﬁc Northwest. Climatic Change 61: 45–88.
Myers JM, Kope RG, Bryant GJ, Teel D, Lierheimer LJ, Wainwright TC,
Grant WS, Waknitz FW, Neely K, Lindley ST, Waples RS. 1998. Status
review of Chinook salmon from Washington, Idaho, Oregon, and
California. U.S. Department of Commerce, NOAA Technical Memorandum
NMFS-NWFSC-35.
Nelitz M, Wieckowski K, Prckard D, Pawley K, Marmorek DR. 2007. Helping
Paciﬁc Salmon Survive the Impact of Climate Change on Freshwater
Habitats. Paciﬁc Fisheries Resource Conservation Council: Vancouver, B.C.
Olden JD, Poff NL. 2004. Ecological processes driving biotic
homogenization: testing a mechanistic model using ﬁsh faunas. Ecology
85: 1867–1875.
Omernik JM, Bailey RG. 1997. Distinguishing between watersheds and
ecoregions, Journal of the American Water Resources Association 33:
935–949.
Opperman JJ, Luster R, McKenney BA, Roberts M, Meadows AW. 2010.
Ecologically functional ﬂoodplains: connectivity, ﬂow regime, and scale.
Journal of the American Water Resources Association 46: 211–226.
Pess GR, McHenry ML, Beechie TJ, Davies J. 2008. Biological impacts of
the Elwha River dams and potential salmonid responses to dam removal.
Northwest Science 82: 72–90.
Pess GR, Morley SA, Hall JL, Timm RK. 2005. Monitoring ﬂoodplain
restoration. In Monitoring Stream and Watershed Restoration, Roni P
(ed.). American Fisheries Society: Bethesda; 127–166.
Poff NL, Allen JD, Bain MB, Karr JR, Prestegaard KL, Richter BD, Sparks
RE, Stromberg JC. 1997. The natural ﬂow regime. BioScience 47: 769–784.
Poff NL, Richter BD, Arthington AH, Bunn SE, Naiman RJ, Kendy E,
Acrement M, Apse C, Bledsoe BP, Freeman MC, Henriksen J, Jacobsen
RB, Kennen JG, Merritt DM, O’Keeffe JH, Olden JD, Rovers K, Tharme
RE, Warner A. 2010. The ecological limits of hydrologic alteration
(ELOHA): a new framework for developing regional environmental ﬂow
standards. Freshwater Biology 55: 147–170.
Pollock MM, Beechie TJ, Chan S, Bigley R. 2005. Monitoring and evaluating riparian restoration efforts. In Monitoring Stream and Watershed
Restoration, Roni P (ed.). American Fisheries Society: Bethesda; 67–96.
Pollock MM, Beechie TJ, Jordan CE. 2007. Geomorphic changes upstream
of beaver dams in Bridge Creek, in incised stream channel in the interior
Columbia River basin, eastern Oregon. Earth Surface Processes and
Landforms 32: 1174–1185.
Pollock MM, Beechie TJ, Liermann M, Bigley RE. 2009. Stream
temperature relationships to forest harvest in the Olympic Peninsula,
Washington. Journal of the American Water Resources Association 45:
141–156.
Pollock MM, Heim M, Werner D. 2003. Hydrologic and geomorphic
effects of beaver dams and their inﬂuence on ﬁshes. In The Ecology
and Management of Wood in World Rivers. Gregory SV, Boyer K,
Gurnell A (eds). American Fisheries Society: Bethesda, MD; 213–233.
Ponce VM, Lindquist DS. 1990. Management of baseﬂow augmentation: a
review. Water Resources Bulletin 26: 259–268.
Poole GC, Berman CH. 2001. An ecological perspective on in-stream
temperature: natural heat dynamics and mechanisms of human-caused
thermal degradation. Environmental Management 27: 787–802.
Poole GC, O’Daniel SJ, Jones KL, Woessner WW, Bernhardt ES, Helton
AM, Stanford JA, Boer BR, Beechie TJ. 2008. Hydrologic spiraling:
the role of multiple interactive ﬂow paths in stream ecosystems. River
Research and Applications 24: 1018–1031.
Copyright © 2012 John Wiley & Sons, Ltd.

959

Quinn TP. 2005. The Behavior and Ecology of Paciﬁc Salmon and Trout.
University of Washington Press: Seattle.
Reid LM, Dunne T. 1984. Sediment production from forest road surfaces.
Water Resources Research 20: 1753–1761.
Richter A, Kolmes SA. 2005. Maximum temperature limits for Chinook,
coho, and chum salmon, and steelhead trout in the Paciﬁc Northwest.
Reviews in Fisheries Science 13: 23–49.
Rieman BE, Isaak D, Adams S, Horan D, Nagel D, Luce C, Myers D. 2007.
Anticipated climate warming effects on bull trout habitats and populations
across the interior Columbia River basin. Transactions of the American
Fisheries Society 136: 1552–1565. doi: 10.1577/T07–028.1
Robison, GE, Runyon J, Andrus C. 1999. Oregon Department of Forestry
cooperative stream temperature monitoring: project completion report
for 1994–1995. Forest Practices Technical Report 2, Oregon Department
of Forestry, Forest Practices, Salem, OR. 47.
Roni P, Quinn TP. 2001. Density and size of juvenile salmonids in
response to placement of large woody debris in western Oregon and
Washington streams. Canadian Journal of Fisheries and Aquatic
Sciences 58: 282–29.
Roni P, Beechie TJ, Bilby RE, Leonetti FE, Pollock MM, Pess GR. 2002. A
review of stream restoration techniques and a hierarchical strategy for
prioritizing restoration in Paciﬁc Northwest watersheds. North American
Journal of Fisheries Management 22: 1–20.
Roni P, Hanson K, Beechie T. 2008. Global review of the physical and
biological effectiveness of stream habitat rehabilitation techniques. North
American Journal of Fisheries Management 28: 856–890.
Scheuerell MD, Hilborn R, Ruckelshaus MH, Bartz KK, Lagueux KM,
Haas AD, Rawson K. 2006. The Shiraz model: a tool for incorporating
anthropogenic effects and ﬁsh–habitat relationships in conservation
planning. Canadian Journal of Fisheries and Aquatic Sciences 63:
1596–1607.
Schindler DE, Hilborn R, Chasco B, Boatright CP, Quinn TP, Rogers LA,
Webster MS. 2010. Population diversity and the portfolio effect in an
exploited species. Nature 465: 609–612.
Shared Strategy Development Committee. 2007. Puget Sound Salmon
Recovery Plan. National Marine Fisheries Service, Portland, Oregon.
(http://www.nwr.noaa.gov/Salmon-Recovery-Planning/Recovery-Domains/
Puget-Sound/PS-Recovery-Plan.cfm).
Shields FD Jr, Knight SS, Cooper CM. 1995. Rehabilitation of watersheds
with incising channels. Water Resources Bulletin 31:971–982.
Sidle RC, Pierce AJ, O’Loughlin CL. 1985. Hillslope Stability and
Land Use. Water Resources Monograph Series Volume 11. American
Geophysical Union: Washington, D.C.
Solazzi MF, Nickelson TE, Johnson SL, Rodgers JD. 2000. Effects of
increasing winter rearing habitat on abundance of salmonids in two
coastal Oregon streams. Canadian Journal of Fisheries and Aquatic
Sciences, 57: 906–91.
Sommer TR, Nobriga ML, Harrell WC, Batham W, Kimmerer WJ. 2001.
Floodplain rearing of juvenile Chinook salmon: evidence of enhanced
growth and survival. Canadian Journal of Fisheries and Aquatic
Sciences 58: 325–333.
Sparks RE, Nelson JC, Yin Y. 1998. Naturalization of the ﬂood regime in
regulated rivers. BioScience 48: 706–720.
Stanford JA, Ward JV, Liss WJ, Frissell CA, Williams RN, Lichatowich
JA, Coutant CC. 1996. A general protocol for restoration of regulated
rivers. Regulated Rivers: Research & Management, 12: 391–413.
U.S. Environmental Protection Agency. 2003. EPA Region 10 guidance for
Paciﬁc Northwest state and tribal temperature water quality standards.
EPA-910-B-03-002. Region 10 Ofﬁce of Water, Seattle, Washington.
Waples RS, Beechie TJ, Pess GR. 2009. Evolutionary history, habitat
disturbance regimes, and anthropogenic changes: what do these mean
for resilience of Paciﬁc salmon populations? Ecology and Society
14: 3. [online] http://www.ecologyandsociety.org/vol14/iss1/art3/.
River Res. Applic. 29: 939–960 (2013)
DOI: 10.1002/rra

 960

T. BEECHIE ET AL.

Waples RS, Pess GR, Beechie T. 2008. Evolutionary history of Paciﬁc salmon
in dynamic environments. Evolutionary Applications 1: 189–206.h
Welty JT, Beechie TJ, Sullivan K, Hyink DM, Bilby RE, Andrus C, Pess G.
2002. Riparian aquatic interaction simulator (RAIS): a model of riparian
forest dynamics for the generation of large woody debris and shade.
Forest Ecology and Management 162: 299–318.
Wendt RC, Burwell RE. 1985. Runoff and soil losses for conventional, reduced,
and no-till corn. Journal of Soil and Water Conservation 40: 450–454.

Copyright © 2012 John Wiley & Sons, Ltd.

Whited D, Kimball J, Lucotch J, Maumenee N, Wu H, Chilcote S,
Standford J. in press A riverscape analysis tool developed to assist wild
salmon conservation. Fisheries.
Wipﬂi MS, Baxter CV. 2010. Linking ecosystems, food webs, and ﬁsh
production: subsidies in salmonid watershed. Fisheries 35: 373–387.
Wu H, Kimball JS, Mantua N, Stanford JA. 2011. Automated upscaling of
river networks for macroscale hydrological modeling. Water Resources
Research 47: W03517. doi: 10.1029/WR008871

River Res. Applic. 29: 939–960 (2013)
DOI: 10.1002/rra