Comparing translocated beavers used as passive restoration tools to resident beavers in degraded desert rivers

Abstract

Wildlife translocation facilitates conservation efforts, including recovering imperiled species, reducing human–wildlife conflict, and restoring degraded ecosystems. Beaver (American, Castor canadensis; Eurasian, C. fiber) translocation may mitigate human–wildlife conflict and facilitate ecosystem restoration. However, few projects measure outcomes of translocations by monitoring beaver postrelease, and translocation to desert streams is relatively rare. We captured, tagged, and monitored 47 American beavers (hereafter, beavers) which we then translocated to two desert rivers in Utah, USA, to assist in passive river restoration. We compared translocated beaver site fidelity, survival, and dam‐building behavior to 24 resident beavers. We observed high apparent survival (i.e., survived and stayed in the study site) for eight weeks postrelease of resident adult beavers (0.88 ± 0.08; standard error) and lower but similar apparent survival rates between resident subadult (0.15 ± 0.15), translocated adult (0.26 ± 0.12), and translocated subadult beavers (0.09 ± 0.08). Neither the pre‐ nor the post‐translocation count of river reaches with beaver dams were predicted well by the Beaver Restoration Assessment Tool, which estimates maximum beaver dam capacity by river reach, suggesting beaver‐related restoration is not maximized in these rivers. Translocated beavers exhibited similar characteristics as resident subadult beavers during dispersal; they were more vulnerable to predation and many emigrated from the study sites. High mortality and low site fidelity should be anticipated when translocating beavers, but even so, translocation may have contributed to additional beaver dams in the restoration sites, which is the common goal of beaver‐assisted river restoration. Multiple releases at targeted restoration sites may eventually result in establishment and meet conservation objectives for desert rivers.

Full text

Comparing translocated beavers used as passive
restoration tools to resident beavers in degraded desert
rivers
E. Doden
1,2
, P. Budy
2,3,4
, M. Conner
1
& J. K. Young
1
1 Department of Wildland Resources, Utah State University, Logan, UT, USA
2 The Ecology Center, Utah State University, Logan, UT, USA
3 Utah Cooperative Fish and Wildlife Research Unit, U.S. Geological Survey, Utah State University, Logan, UT, USA
4 Department of Watershed Sciences, Utah State University, Logan, UT, USA
Keywords
beaver; Castor canadensis; translocation;
river restoration; survival; dam building.
Correspondence
Julie K. Young, Department of Wildland
Resources, Utah State University, 5230 Old
Main Hill, Logan, UT 84322-5230, USA.
Email: julie.young@usu.edu
Editor: John Ewen
Associate Editor: Pia Lentini
Received 06 August 2021; accepted 25
November 2022
doi:10.1111/acv.12846
Abstract
Wildlife translocation facilitates conservation efforts, including recovering imperiled
species, reducing human–wildlife conflict, and restoring degraded ecosystems. Bea-
ver (American, Castor canadensis; Eurasian, C. fiber) translocation may mitigate
human–wildlife conflict and facilitate ecosystem restoration. However, few projects
measure outcomes of translocations by monitoring beaver postrelease, and translo-
cation to desert streams is relatively rare. We captured, tagged, and monitored 47
American beavers (hereafter, beavers) which we then translocated to two desert riv-
ers in Utah, USA, to assist in passive river restoration. We compared translocated
beaver site fidelity, survival, and dam-building behavior to 24 resident beavers. We
observed high apparent survival (i.e., survived and stayed in the study site) for
eight weeks postrelease of resident adult beavers (0.88 0.08; standard error) and
lower but similar apparent survival rates between resident subadult (0.15 0.15),
translocated adult (0.26 0.12), and translocated subadult beavers (0.09 0.08).
Neither the pre- nor the post-translocation count of river reaches with beaver dams
were predicted well by the Beaver Restoration Assessment Tool, which estimates
maximum beaver dam capacity by river reach, suggesting beaver-related restoration
is not maximized in these rivers. Translocated beavers exhibited similar characteris-
tics as resident subadult beavers during dispersal; they were more vulnerable to
predation and many emigrated from the study sites. High mortality and low site
fidelity should be anticipated when translocating beavers, but even so, translocation
may have contributed to additional beaver dams in the restoration sites, which is
the common goal of beaver-assisted river restoration. Multiple releases at targeted
restoration sites may eventually result in establishment and meet conservation
objectives for desert rivers.
Introduction
Wildlife translocations are valuable conservation tools for
recovering imperiled species, reducing human–wildlife con-
flict, and restoring degraded ecosystems (Germano
et al., 2015; Mengak, 2018; Novak, Phelan, & Weber, 2021).
American and Eurasian beavers (Castor canadensis and
C. fiber) have been translocated for over 70 years after
extensive extirpation from much of their historical ranges
during the fur trade of the 1700s–1800s (Baker & Hill, 2003;
Halley, Saveljev, & Rosell, 2021). Translocations of Ameri-
can beavers (hereafter, beavers) in the United States often
focus on removing nuisance individuals from conflict situa-
tions where they would otherwise be lethally removed and
using them instead as ecosystem engineers for riparian
restoration; they may increase the large woody debris contri-
bution and the number of dams in the system to initiate
process-based restoration and improve degraded systems
(Naiman, Johnston, & Kelley, 1988; Pollock et al., 2014;
Pilliod et al., 2018). Through dam building, canal formation,
wood contributions, and foraging activity (Baker &
Hill, 2003), beavers provide numerous ecosystem services
such as adding heterogeneity to ecosystems (Wright, Jones,
& Flecker, 2002), reducing stream channel incision (Pollock
et al., 2014), and promoting drought, climate change, and
wildfire resistance (Hood & Bayley, 2008; Fairfax &
Small, 2018; Fairfax & Whittle, 2020), which also benefits
many other species (Rosell et al., 2005).
Animal Conservation  (2022) – ª2022 Zoological Society of London. This article has been contributed to by U.S. Government employees and their work is
in the public domain in the USA. 1
Animal Conservation. Print ISSN 1367-9430

Despite the history of using beaver translocations for spe-
cies and ecosystem conservation, best management practices
to ensure beaver establishment are lacking (Pilliod
et al., 2018; Nash et al., 2021). Understanding the life his-
tory characteristics of existing resident individuals could help
inform management practices, yet studies comparing translo-
cated individuals to residents are uncommon (but see Pinter-
Wollman, Isbell, & Hart, 2009; Baker et al., 2021; Muriel
et al., 2021). Successful beaver establishment for river
restoration is defined by long-term residency, survival, and
dam building at release sites; however, long-term residency
rarely exceeds 50% of individuals released (McKinstry &
Anderson, 2002; Petro, 2013; Dittbrenner, 2019; but see
Albert & Trimble, 2000), survival of translocated beavers
can be less than 50% (McKinstry & Anderson, 2002; Petro,
Taylor, & Sanchez, 2015), and there may be no apparent
link between dam-building behavior before and after translo-
cations (Petro et al., 2015). Thus, successful beaver translo-
cation remains challenging and requires further exploration.
Suitable habitats for dam building, foraging, and evading
predators are key components for the long-term residency
and survival of translocated beavers. Beavers build dams to
create pools as a refuge from predators, cache food, and
access bank dens and lodges, but beavers may not build
dams if these needs are already met (Baker & Hill, 2003;
Nash et al., 2021). Releasing translocated beavers at sites
where artificial or natural woody structures are already
placed, such as where partial or channel-spanning beaver
dam analogs (BDAs) have been constructed, may help them
establish at the release site and encourage dam-building
behavior (DeVries et al., 2012; Bouwes et al., 2016). Site-
specific factors can also affect success (e.g., predator density,
existing beaver densities, habitat availability), emphasizing
the importance of assessing the suitability of translocation
release sites on a case-by-case basis (Petro et al., 2018;
Touihri et al., 2018).
Translocation success can be more challenging in extreme
environments or low-quality habitats, such as in degraded
desert rivers (Armstrong & Seddon, 2008). Desert rivers are
integral for the survival of many desert species, from aquatic
macroinvertebrates to breeding birds (Knopf et al., 1988;
Kingsford & Thompson, 2006; Sada et al., 2021), but are
often jeopardized by altered flow regimes, over-allocation,
impoundment structures, invasive species (Stromberg, 2001;
Mott Lacroix, Tapia, & Springer, 2017), and mega-drought
(Udall & Overpeck, 2017; Williams, Cook, & Smer-
don, 2022). Beavers could mitigate the effects of these alter-
ations and naturally inhabit arid desert rivers, but little is
known about their ecology and effects on desert systems
(Gibson & Olden, 2014; Barela & Frey, 2016). Transloca-
tions could supplement existing populations but more infor-
mation is needed to ensure they are successful (Pilliod
et al., 2018).
In our study, we sought to identify whether beaver
translocation could serve as an effective means of restoration
in desert rivers by determining translocation success and
comparing the residency, survival, and dam building of
translocated beavers to residents. We defined translocation
success as beavers staying, surviving, and building dams
within the study sites for at least eight weeks after release.
This period was chosen for logistical (e.g., high rates of
transmitter loss) and biological reasons. While we had
intended to also obtain estimates over longer periods of time,
this time is biologically relevant because it was sufficient to
identify beavers that settled post-translocation from those that
dispersed outside of our study sites. We expected resident
adult beavers would remain in their territories, have high sur-
vival rates, and build dams, while resident subadult and
translocated beavers would have lower rates of release-site
fidelity, survival, and dam-building activity. We expected
some translocated beavers would successfully establish in the
study sites. Beavers in this system can serve as an effective
tool for creating and maintaining habitat for imperiled ende-
mic desert fish and many other wildlife species, so success-
ful establishment by translocated beavers could enhance the
effects of resident beavers already in the system (e.g.,
Remiszewski, 2022).
Although beavers create and maintain the dams, ulti-
mately, beaver dams and not the beavers themselves are the
common goal of restoration, and so we also sought to deter-
mine whether our study sites approached maximum dam
capacity by comparing the pre- and post-translocation river
reaches with ≥1 dams to the Utah Beaver Restoration
Assessment Tool (BRAT; Macfarlane, Wheaton, & Jensen,
2014; Macfarlane et al., 2017). The BRAT model assesses
the capacity of beaver dams in streams at the reach level
and has been widely applied to inform the selection of
restoration sites; we used BRAT in initial decisions to iden-
tify study sites for river restoration efforts involving beaver
translocations (MacFarlane et al., 2017). Beaver-related
restoration has the most potential in rivers at minimum dam
capacity.
Materials and methods
Study area
We conducted our study from May 2019 to March 2021 at
three sites along the lower stretches of the Price and San
Rafael Rivers, part of the greater Colorado River Basin in
east-central Utah, USA. River degradation is caused by sim-
plification, dewatering, and invasive species encroachment
especially in the lower river reaches. Several federally endan-
gered or state-sensitive fish use these rivers (Colorado pike-
minnow Ptychocheilus lucius, bonytail chub Gila elegans,
razorback sucker Xyrauchen texanus, bluehead sucker
Catostomus discobolus,flannelmouth sucker Catostomus
latipinnis, roundtail chub Gila robusta; Bottcher et al., 2013;
Budy et al., 2015). The first site was 20.5 km from the Price
River where a multi-faceted restoration project was underway
which included beaver translocations, tamarisk removal, and
BDA installations (Site 1). At the time of this study, only
beaver translocations were completed. The second site was
8.1 km of the San Rafael River at Moonshine Wash near the
confluence with the Green River, where relatively small-scale
tamarisk removal, gravel addition, native tree planting, and
2Animal Conservation  (2022) – ª2022 Zoological Society of London. This article has been contributed to by U.S. Government employees and their work
is in the public domain in the USA.
Beaver translocation in degraded desert rivers E. Doden et al.

low-density installation of BDAs were completed in 2015
(Site 2; Laub, 2015,2018). The third site was 1.5 km from
the San Rafael River near Cottonwood Wash, which experi-
enced a dramatic geomorphic change beginning in 2010 due
to a sediment plug, resulting in a rare stretch of in-stream
habitat complexity in the otherwise simplified lower San
Rafael River (Site 3; Lyster, 2018). We did not translocate
any beavers to Cottonwood Wash because a beaver colony
already existed and was actively building and maintaining
dams. Our intent was to use this site as a reference area.
The rivers flow through canyonlands and desert shrub-
lands, with temperatures ranging from 11°C in winter to
above 37°C in summer. Annual rainfall averages 21 cm per
year (NOAA, 2021a). Dominant riparian vegetation includes
a limited mix of native and nonnative species: willow (Salix
spp.), Fremont cottonwood Populus fremontii, common reed
Phragmites spp., as well as tamarisk (live and dead; Tamarix
ramosissima) and Russian olive Elaeagnus angustifolia. Cat-
tails Typha spp. were also present at Cottonwood Wash.
Capture, quarantine, and tagging
All capture, handling, and monitoring procedures were
approved by Utah State University’s Institute for Animal
Care and Use Committee (No. 10128). We responded to nui-
sance beaver calls in northern, central, and eastern Utah to
capture beavers for translocation (Fig. 1) and captured resi-
dent beavers at the Cottonwood Wash and Price River study
sites. We captured beavers from May to October of 2019
and 2020 using Hancock/Koro suitcase-style traps, Comstock
box traps, or nonlethal cable restraints. We held captured
beavers at the Utah State University Beaver Ecology and
Relocation Center in Logan, Utah, or the field site, providing
food and fresh water daily (Campbell-Palmer &
Rosell, 2015). We held translocated beavers for at least
three days, following state quarantine protocols (Utah Divi-
sion of Wildlife Resources, 2017; Pilliod et al., 2018).
We processed and fitted beavers with monitoring tags
before release. We first assigned age class using weight and
body size (kit <1 year and <6.0 kg, subadult =1–2 years
and 6.1–14.2 kg, adult >2 years and >14.3 kg; Patric &
Webb, 1960) and determined the sex via anal gland secretion
(Schulte, M€uller-Schwarze, & Sun, 1995; Woodruff & Pol-
lock, 2018). Next, we inserted passive integrated transponder
(PIT) tags (Biomark APT12 tags, Boise, ID, USA) in all
beaver tails and fit beavers >9.0 kg (i.e., subadults and
adults) with a remotely downloadable store on-board GPS
tag (Africa Wildlife Tracking; Rietondale, Pretoria, South
Africa) or a VHF modified ear-tag as tail-mounted transmit-
ters (Advanced Telemetry Systems, Isanti, MN, USA; Model
#M3530; Rothmeyer, McKinstry, & Anderson, 2002; Arjo
et al., 2008). In the first year, we used VHF tags before
GPS tags were available and in the second year, we used
VHF tags because of the high failure rates of GPS tags. Ini-
tially, we attached GPS and VHF tags with 19 mm neoprene
and steel washers, then increased the washer size (neoprene:
38.1 mm, steel: 31.8 mm) in September 2019 to improve
transmitter retention (Windels & Belant, 2016). We
chemically immobilized beavers with isoflurane and supple-
mented them with oxygen during GPS- or VHF-tag attach-
ment (Roug et al., 2018). We released resident beavers at
their capture sites and translocated beavers in areas of Moon-
shine Wash near BDAs and on the Price River where there
was no beaver activity but identified as suitable based on
historical beaver activity, local food and building resources,
and availability of burrows and cover. Moonshine Wash on
the San Rafael River was unsuitable for beaver translocation
in 2020 due to extremely low river flows.
Monitoring
We located beavers 2–7 times per week via VHF- and GPS-
tag locations from May through October 2019 and 2020.
Beavers were also passively detected from May 2019
through March 2021 via permanent and submersible passive
integrated antennae (PIA; Biomark; Fig. 2). When VHF sig-
nals indicated mortality, we recovered the transmitter. If we
found a dead beaver, we performed a necropsy to determine
the cause of death. If no beaver was present and there were
no signs of predation, we recorded the event as a transmitter
loss. We searched for beavers that likely left the study sites
(i.e., were not regularly detected) for at least two weeks fol-
lowing their last detection and sporadically throughout the
remaining field season. To increase detections, we conducted
monthly scans along the Green River, one aerial flight, and
several river floats on the Price and San Rafael Rivers. We
considered beavers detected during these occasions to have
temporarily (later detected back in the study sites) or perma-
nently (never detected back in the study sites) emigrated
from the study sites. At the end of the monitoring period
(31 March 2021), we classified beavers into different fate
categories: unknown, mortality, and alive (detected ≥5
months and within two weeks of the end of the monitoring
period).
Beaver activity surveys
We conducted beaver activity surveys in June 2019 at Cot-
tonwood and Moonshine Wash, and in August 2019 at the
Price River study site, which consisted of walking or floating
along the rivers and recording all dams, lodges, burrows,
fresh scats, runs, and chews on a GPS unit (Garmin, Chi-
cago, IL, USA; Model GPSMAP 78s or 66st). We observed
areas of resident beaver activity at Cottonwood Wash and
certain stretches of the Price River, but little activity at
Moonshine Wash. We censused all existing beaver dams
before translocations occurred using activity surveys and
satellite imagery (Google Earth Pro, 2019), and then com-
pleted a final beaver dam count in October 2020. We catego-
rized beaver dams into four types: resident old, resident new,
translocated new, and unknown new. We considered dams
new if built after the first beaver dam census at each study
site. We considered old and new dams built within a resident
adult beaver’s 100% minimum convex polygon home range
to be built by that individual. We designated new dams built
by translocated beavers when the construction date was
Animal Conservation  (2022) – ª2022 Zoological Society of London. This article has been contributed to by U.S. Government employees and their work is
in the public domain in the USA. 3
E. Doden et al. Beaver translocation in degraded desert rivers

known and within 100 m of at least four concurrent loca-
tions of a translocated beaver (Woodford, Macfarland, &
Worland, 2013; Touihri et al., 2018; Matykiewicz
et al., 2021). We assigned all other new dams encountered
as unknown. Finally, we assigned each observed dam based
on its location in the river to its appropriate river reach
delineated by the Utah BRAT to determine whether the riv-
ers attained maximum capacity of reaches with ≥1 beaver
dam before and after translocations occurred (Macfarlane
et al., 2014,2017).
Analysis
We estimated the probability of beavers surviving and
remaining in the study sites (φ; apparent survival) and
encounter probability (p) using Cormack-Jolly-Seber (CJS)
models in Program MARK (version 9.0) with logit-link func-
tions to produce maximum likelihood estimates (Cormack,
1964; Jolly, 1965; Seber, 1965; Lebreton et al., 1992; White
& Burnham, 1999). We estimated apparent survival in the
study sites for eight weeks postrelease for four groups of
beavers (g)–resident adult, resident subadult, translocated
adult, and translocated subadult –and four sampling occa-
sions (t)–weeks 1–2, 3–4, 5–6, and 7–8. We used CJS
models to estimate φbecause a known-fate model was
unsuitable with our high rate of unknown fate. We limited
the analysis to eight weeks postrelease because we only
monitored most individuals for eight weeks before unknown
fate occurred. We first assessed the fit of our global model
(φ
g+t
p
g+t
) using the total goodness-of-fit Test 2 +Test 3
results from program RELEASE retrieved through MARK.
We estimated the overdispersion factor, ^
c, using 1500
Figure 1 Capture and release locations of translocated beavers. Beavers were captured in three ecoregion types in Utah: Central Basin and
Range, Wasatch and Uinta Mountains, and Colorado Plateaus (Omernik, 1987). Release sites are on the Price River and the San Rafael River
at Moonshine Wash, Utah, USA.
4Animal Conservation  (2022) – ª2022 Zoological Society of London. This article has been contributed to by U.S. Government employees and their work
is in the public domain in the USA.
Beaver translocation in degraded desert rivers E. Doden et al.

bootstrapped iterations of the model and dividing the
observed model deviance by the mean of the bootstrapped
deviances (Burnham et al., 1987; Anderson, Burnham, &
White, 1994; White & Burnham, 1999). We built a model
set with the additive effects of individual covariates on φ
and p: including sex, year released, days held in quarantine
before release, average discharge, and air temperature across
the eight weeks (USGS, 2021; NOAA, 2021b), maximum
air temperature the day of release (USGS, 2021;
NOAA, 2021b), season released (spring: March 1–May 31,
summer: June 1–August 31, fall: September 1–November
30), and ecoregion of an individual’s origin (Colorado Pla-
teaus, Wasatch and Uinta Mountains, and Central Basin and
Range; Omernik, 1987). We did not include study site as a
covariate because of the small sample size and sites were
confounded by the year; we only released beavers on the
San Rafael River in 2019. We constrained models to include
≤7 parameters to balance between optimizing model likeli-
hood and avoiding model overparameterization (Burnham &
Anderson, 2002). We compared models using an information
theoretic approach with Akaike information criterion tables
adjusted for a small sample size (AICc; Anderson
et al., 1994; Burnham & Anderson, 2002). Using the model-
averaged estimates of apparent survival and encounter proba-
bility, we multiplied the apparent survival estimated for each
time interval to generate apparent survival probability for the
entire eight weeks and used the Delta Method to estimate
the variance (White & Burnham, 1999; Ver Hoef, 2012). We
Figure 2 The three study sites used to compare resident and translocated beaver ecology in east-central Utah, USA: one on the Price River
and two on the lower San Rafael River at Cottonwood Wash and Moonshine Wash. Permanent and submersible “wagon wheel” passive
integrated antennae (PIA) locations are included. All locations where submersible PIAs were deployed are shown but submersible PIAs were
moved several times throughout the study and none remained in one location for the entire study.
Animal Conservation  (2022) – ª2022 Zoological Society of London. This article has been contributed to by U.S. Government employees and their work is
in the public domain in the USA. 5
E. Doden et al. Beaver translocation in degraded desert rivers

considered parameters included in the top model with 95%
confidence intervals that overlapped zero to be insignificant
(Arnold, 2010).
We used logistic regression with a logit-link function to
determine whether the dam capacity per river reach from
BRAT was a significant predictor of the probability pre- and
post-translocation that ≥1 dam was observed in a given river
reach at a=0.05. We performed all statistical analyses in
Program R (version 4.0.3; R Core Team, 2020).
Results
We captured and PIT-tagged 24 residents (15 females, 9
males) and 47 translocated beavers (20 females, 27 males),
with a subset radio-tagged (Table 1). Three additional resi-
dent beavers died from capture- and processing-related
events, so we excluded them from analyses. We excluded
beaver kits from analyses due to low detection rates. We
released an average of 1.6 beavers per translocation at
Moonshine Wash (range 1–2, 5 release efforts in 2 locations)
and 2.4 beavers per translocation to the Price River (range
1–8, 16 release efforts in 3 locations).
We never detected resident adult beavers outside of the
study sites. One resident subadult beaver was not detected at
the Price River study site for 18 days before being detected
back on the site. We detected only 6.4% (n=3 adult bea-
vers) of all PIT-tagged translocated beavers exclusively
inside the study sites after eight weeks. We detected 40.4%
(n=19) of all PIT-tagged translocated beavers outside of the
study sites at least once, primarily with PIAs.
At the end of the monitoring period, there were three of
24 resident beavers known to be alive: one kit and one suba-
dult at the Price River site and one adult at the Cottonwood
Wash site. There were also four of 47 translocated subadult
beavers alive, but all had emigrated from the Price River
study site. Unknown fate made up the largest proportion of
translocated adult and subadult beavers (n=13, 81% of resi-
dent beavers; n=30, 73% of translocated beavers), caused
by GPS-transmitter failures, transmitter loss, or individuals
emigrating outside of the study sites. We recovered 11 trans-
mitters pulled out of tails (six resident and five translocated
beavers), with a 46% transmitter loss rate before washer
improvements, and a 17% transmitter loss rate afterward. We
continued to detect two individuals who lost their transmit-
ters with PIAs until the end of the study and therefore
included them in the “alive”category. We detected eight
mortalities for beavers fitted with radio transmitters, half
occurring within the first week postrelease. One resident dis-
persing subadult beaver was killed by a bobcat or mountain
lion. Seven translocated beavers died (four adults and three
subadults). One had poor body condition and may have died
from translocation stress, two likely died of the combined
stressors of sustaining cable-restraint injuries and being
released during an unanticipated drought on the Price River,
and up to four were killed by predators; one each by felid
and coyote, one kill was too old to distinguish the predator,
and one was either killed or scavenged by a black bear. Res-
ident adult beavers demonstrated significantly higher appar-
ent survival probability “φ”than all other groups (Table 2).
For the eight-week CJS analysis, we estimated ^
c=1.35.
To control for overdispersion, we used QAICc for model
selection and inflated variances of parameter estimates by ^
c
(Burnham & Anderson, 2002). We included no other param-
eters for pin the final model set due to the small sample
size, and therefore pwas estimated as constant across all
time intervals and group types (p=0.95 0.04). No other
parameters besides beaver group type were significant for φ
in any models, and no model carried the majority of the
weight so model-averaging was employed to derive estimates
of φ(Table 3).
Before beaver translocations began, we observed 23 exist-
ing dams in 17 river reaches built by resident beavers in the
study sites (six at Cottonwood Wash and 17 on the Price
River). We recorded 22 new dams built in 16 previously
undammed reaches and four reaches where dams already
existed. We observed evidence that one dam was built by a
translocated beaver and two by resident beavers on the Price
River. For the 19 other new dams detected (three at Moon-
shine Wash, 14 on the Price River), we were unable to iden-
tify which beavers were responsible for the construction
(Fig. 3). The translocated beaver who built a dam was later
killed by a predator.
We excluded one dam and river reach from the analysis
because we were unable to determine whether the dam was
Table 1 The number of resident and translocated beavers PIT-tagged and released in study sites on the Price River and at Cottonwood
Wash and Moonshine Wash on the San Rafael River, Utah, USA, from May–October 2019 and 2020
Study site
Year
released
Resident
adult
Resident
subadult
Resident
kit
Translocated
adult
Translocated
subadult
Translocated
kit
Price River (Site 1) 2019 3 (3) 2 (2) 3 2 (2) 1 (1) 1
2020 5 (5) 4 (2) 5 14 (14) 16 (10) 5
Moonshine Wash (Site 2) 2019 0 0 0 5 (5) 3 (3) 0
2020 0 0 0 0 0 0
Cottonwood Wash (Site 3) 2019 2 (2) 1 0 0 0 0
2020 0 0 0 0 0 0
Total 10 6 8 21 20 6
Numbers in parenthesis indicate the number of individuals radio-tagged (GPS tag or VHF tag) in addition to being passive integrated
transponder (PIT)-tagged.
6Animal Conservation  (2022) – ª2022 Zoological Society of London. This article has been contributed to by U.S. Government employees and their work
is in the public domain in the USA.
Beaver translocation in degraded desert rivers E. Doden et al.

built before or after beaver translocations. We included 210
total river reaches in analyses, with BRAT predicting 171.7
expected dams in those reaches. Results from the logistic
regression revealed that BRAT dam capacity was not a sig-
nificant predictor of the probability of observed reaches hav-
ing a dam or not for both pre- and post-translocation
observations (Table 4).
Discussion
Although there have been many successful translocation
efforts, challenges remain (Griffith et al., 1989; Wolf
et al., 1996). Animals can leave their targeted release site,
experience high rates of mortality, or behave unpredictably
(Mengak, 2018; Berger-Tal, Blumstein, & Swaisgood, 2020).
Identifying ways to mitigate these challenges improves the
success of future translocation efforts, and ultimately aids in
wildlife conservation (Fischer & Lindenmayer, 2000; Arm-
strong & Seddon, 2008). In our study, translocated beavers
had low success in surviving and building dams but may
have helped increase the number of beaver dams without
changing the behavior of resident beavers, indicating translo-
cations may result in an increased number of dams and
enhanced habitat (e.g., contribute large wood to the active
channel). However, our results could have occurred for other
ecological reasons associated with high water discharge in
the first year and more monitoring efforts to determine
whether the translocation of beavers can augment and
enhance natural populations is still needed since supplemen-
tal translocations could more quickly reach restoration goals.
We de fined success as translocated beavers staying, sur-
viving, and building dams within the study sites for at least
eight weeks postrelease and found limited evidence of suc-
cess based on that definition and issues with tag loss and
failure. One translocated beaver on the Price River was
detected in a 400-m stretch of river for 19 days and built a
dam, but it was killed by a predator before eight weeks.
However, it is likely a few other translocated beavers that
lost their tags established home ranges and built dams at the
study sites because activity was observed where no known
resident beaver colonies existed, but we were unable to con-
firm this. We translocated a high number of beavers to com-
pensate for the mobility, mortality, and unpredictable nature
of most translocated beavers. Translocated beavers initially
behaved and had mortality risks like dispersing resident
Table 2 Apparent survival probability estimates (the probability an
individual survived and stayed in the study site) from Cormack-
Jolly-Seber models of beavers monitored for eight weeks postre-
lease, May–October 2019 and 2020, in study sites on the Price
River and at Cottonwood Wash and Moonshine Wash on the San
Rafael River, Utah, USA
Group type
Sample
size
Apparent survival
probability (φ)
1 Standard
error
Resident adult 9 0.88 0.08
Resident subadult 3 0.15 0.15
Translocated adult 21 0.26 0.12
Translocated
subadult
14 0.09 0.08
Table 3 QAICc output adjusted for small sample size and overdispersion of Cormack-Jolly-Seber models estimating apparent survival proba-
bility for beaver detections within three study sites in a desert ecosystem of Utah, USA, for eight weeks postrelease, May–October 2019
and 2020
Model
a
Number of parameters QAICc DQAICc QAICc weight Model likelihood QDeviance
φ(g) p(.) 5 89.25 0.00 0.18 1.00 78.48
φ(g +Year) p(.) 6 89.68 0.43 0.14 0.81 76.59
φ(g +Av8WkDis) p(.) 6 90.35 1.11 0.10 0.57 77.26
φ(g +SeasSum) p(.) 6 91.00 1.76 0.07 0.42 77.91
φ(g +Sex) p(.) 6 91.06 1.81 0.07 0.40 77.97
φ(g +DaysHeld) p(.) 6 91.16 1.91 0.07 0.38 78.07
φ(g +MaxTmpRel) p(.) 6 91.24 1.99 0.06 0.37 78.15
φ(g +EcoB) p(.) 6 91.26 2.01 0.06 0.37 78.17
φ(g +Av8WkTmp) p(.) 6 91.37 2.12 0.06 0.35 78.28
φ(g +EcoM) p(.) 6 91.47 2.23 0.06 0.33 78.38
φ(g +SeasSpr) p(.) 6 91.55 2.30 0.06 0.32 78.45
φ(g +SeasSpr +SeasSum) p(.) 7 93.36 4.12 0.02 0.13 77.89
φ(g +t) p(.) 7 93.58 4.33 0.02 0.11 78.11
φ(g +EcoB +EcoM) p(.) 7 93.63 4.39 0.02 0.11 78.16
φ(.) p(.) 2 95.93 6.68 0.01 0.04 91.78
φ(t) p(.) 4 100.12 10.88 0.00 0.00 91.62
a
Key: g –group type (resident adult, resident subadult, translocated adult, translocated subadult); t –sampling occasion, Year –year
released, 2019 or 2020; Av8WkDis –mean 8-week discharge; Sex –female or male; SeasSpr –spring season; SeasSum –summer season;
DaysHeld –days held in quarantine; MaxTmpRel –maximum air temperature on day of release; Av8WkTmp –mean 8-week air tempera-
ture; EcoB –Central Basin and Range ecoregion; EcoM –Wasatch and Uinta Mountains ecoregion; “.” indicates parameter was constant,
null model. See Supporting Information Table S1 for more information.
Animal Conservation  (2022) – ª2022 Zoological Society of London. This article has been contributed to by U.S. Government employees and their work is
in the public domain in the USA. 7
E. Doden et al. Beaver translocation in degraded desert rivers

subadult beavers, spending time exploring their novel envi-
ronment before finding a place to establish, often outside of
the study sites. Resident adult beavers reliably stayed, sur-
vived, and built and maintained dams in the study sites.
These patterns suggest that adding translocated beavers may
not impact resident beavers or that impacts are delayed. For
example, translocated beavers may be more susceptible to
predation and indirectly increase the survival of resident bea-
vers (Gable & Windels, 2018). Indeed, several mortalities of
translocated beavers in our study were caused by predators.
Further, translocated beavers that survive to establish territo-
ries in the study sites could delay the dispersal of resident
beavers because there are fewer unoccupied spaces (Mayer,
Zedrosser, & Rosell, 2017). However, in our system, we
hypothesized that resident beavers may have prevented
translocated beavers from staying near release sites. Translo-
cation of beavers may be used as a potential tool for popula-
tion augmentation or restoration strategy but with
consideration of the locations of resident beavers. Notably,
all beavers translocated as part of this study would have
been lethally removed if not translocated.
Our apparent survival analysis predicted <25% of translo-
cated beavers survived and remained in the study sites for at
least eight weeks (56 days). This is not much longer than
the mean dispersal-settlement time for subadult beavers in
Montana (40.9 days; Ritter, 2018), where resources required
for beavers to settle are likely more abundant than in our
Figure 3 Beaver dams observed in relation to translocated beaver release sites at Cottonwood Wash (a) and Moonshine Wash (b) on the
San Rafael River, and at the Price River (c), in Utah, USA, May–October 2019 and 2020.
Table 4 Estimates of the expected number of dams in each river
reach from the Utah Beaver Restoration Assessment Tool (BRAT;
Macfarlane et al.,2014,2017) were used to predict observed river
reaches with ≥1 beaver dam before beaver translocations occurred
(pre-translocation), and after beaver translocations occurred (post-
translocation) in the Price and San Rafael Rivers, Utah, USA, May–
October 2019 and 2020
Parameters
Probability
≥1 dam
Lower
95% CI
Upper
95% CI
P
value
Pre-translocation ~BRAT
Estimated dam
capacity from BRAT
0.42 0.29 0.57 0.28
Post-translocation ~BRAT
Estimated dam
capacity from BRAT
0.50 0.41 0.60 0.93
Estimates and their 95% confidence intervals (CI) were back-
transformed from logistic regression models.
8Animal Conservation  (2022) – ª2022 Zoological Society of London. This article has been contributed to by U.S. Government employees and their work
is in the public domain in the USA.
Beaver translocation in degraded desert rivers E. Doden et al.

desert system. Because translocated beavers behaved like
subadult residents in our study system, mean dispersal-
settlement time could be a proxy for the expected time
translocated beavers need to establish residency. Thus, our
survival analysis likely captured all the translocated beavers
that did establish and could have contributed to building
new dams.
Our resident subadult apparent survival rate was substan-
tially lower than the survival rates reported by other studies
(ranging from 0.43 0.12 [standard error] to 0.84 0.04;
McNew & Woolf, 2005; DeStefano et al., 2006; Ritter,
2018). Similarly, our apparent survival of translocated
beavers was also lower than American beavers in Oregon
(0.47 0.12; Petro et al., 2015) and Eurasian beavers in the
Netherlands (64–67%, Nolet & Baveco, 1996). However,
unlike these studies, our study was exclusively in a desert
system where resources are likely to be limited and differ
from the resources available in these other study systems. In
fact, our results resemble establishment rates reported in
Wyoming, where only 19% of beavers translocated to
degraded streams survived >180 days and built dams in the
drainages where they were released but had a higher survival
rate than in our study (0.49 0.07; McKinstry & Ander-
son, 2002).
Due to the common limitations of CJS analyses, we were
unable to estimate mortality separately from emigration.
Thus, it is unlikely that our low apparent survival rates are
solely attributed to mortality since we only encountered eight
mortalities out of 38 radio-tagged resident subadults and
translocated beavers. Apparent survival was likely biased
low due to unknown fate and emigration from the study sites
(Schaub & Royle, 2014). Indeed, a concurrent study in the
same rivers demonstrated resident subadult and translocated
beavers moved on average between 11 and 21 river kilome-
ters between their most upstream and downstream locations,
indicating many beavers traveled outside the study sites
(Doden et al., 2022). In degraded desert rivers, resources are
scarce and dynamic (Gibson & Olden, 2014; Barela &
Frey, 2016), potentially leading to high movement rates if
release sites do not have adequate resources for survival or
mates nearby. In addition, beavers have an increased risk of
predation or starvation during dispersal or translocation while
in unfamiliar waters without known lodges, burrows, and
foraging resources (Letty, Marchandeau, & Aubineau, 2007;
M€uller-Schwarze, 2011; Bonte et al., 2012). Predation was
the largest cause of mortality for translocated beavers in our
study, similar to previous studies (McKinstry & Ander-
son, 2002; Petro et al., 2015). Translocation-related stressors
can also decrease beaver survival, which contributed to the
deaths of some of our beavers (Teixeira et al., 2007; Dick-
ens, Delehanty, & Romero, 2010). Further efforts are needed
to identify ways to enhance the survival of translocated bea-
vers because translocating beavers may still be an improve-
ment over lethal removal of nuisance beavers (Woodruff &
Pollock, 2018). For example, it may be possible to provide
refuge space from predators before beavers are released (e.g.,
soft releases or artificial lodge construction) .
We did not observe any resident adult beavers emigrating
from the study sites, and their apparent survival rates
appeared more comparable to other resident beavers (ranging
from 0.76 0.05 to 1.00; McNew Jr. & Woolf, 2005;
DeStefano et al., 2006; Bloomquist & Nielsen, 2010; Maen-
hout, 2013), and Eurasian beavers in Norway (0.87 0.02
for dominant adults; Campbell et al., 2012). Despite inhabit-
ing an arid system, our resident adult beavers may have
comparable survival to other studies in less harsh climates
because there may be low predator densities (Menge &
Sutherland, 1987) and there was no beaver harvest. Our
high resident adult apparent survival rate suggests the sites
are suitable for beavers, indicating that if the habitat is still
available, once translocated beavers establish a territory like
resident adult beavers, they will likely survive and provide
ecosystem services. Though resident beavers in a similar
arid system in Bridge Creek, Oregon had high survival rates
and overlapping territories indicating the system may have
been at biological carrying capacity, we believed our rivers
were not at carrying capacity (Maenhout, 2013). We
observed unoccupied gaps between resident beaver colonies
and all radio-tagged resident subadult beavers in our study
exhibited dispersal behavior, potentially indicating that suit-
able unoccupied habitats remained in the rivers (Doden
et al., 2022).
Translocation efforts may have increased restoration ser-
vices in the rivers despite being in dynamic, wood-limited
systems. In our regulated rivers, monsoons and flash floods
still occur and dams can get blown out (Andersen & Sha-
froth, 2010), with high stream power and limited availability
of woody materials inhibiting long-lasting dams (DeVries
et al., 2012; Persico & Meyer, 2013; Barela & Frey, 2016).
Thus, it is possible that the increase in dams we observed
represents seasonal differences. More information across mul-
tiple years would better determine the causes of changes in
dam density (Demmer & Beschta, 2008), but we did find
evidence of at least some new dam building by translocated
beavers that suggest translocated beavers provide this func-
tion. While we could not assign which beavers were respon-
sible for each new dam we detected, at least one dam was
built by a translocated beaver and eight other new dams
were built where we did not document resident beavers on
the Price River; all appeared likely to be able to withstand
high flows. We did not document any resident beaver pres-
ence at the start of our study in Moonshine Wash, yet three
new partial channel-spanning dams were built during the
drought in July 2020. These partial dams at only 0.5 m high
were unlikely to withstand high flows, but the positive
impacts of beaver dams on in-stream habitat complexity con-
tinue even when breaching occurs (Demmer & Beschta, 2008;
Pollock et al., 2014). We also anecdotally observed that par-
tially spanning BDAs provided habitat improvements,
although the effects were local. Two of the three dams were
within 30 m of BDAs, suggesting that beaver translocation
may be more effective when artificial structures are provided;
BDA installation similarly led to an overall increase in dams
in Oregon (Bouwes et al., 2016).
Animal Conservation  (2022) – ª2022 Zoological Society of London. This article has been contributed to by U.S. Government employees and their work is
in the public domain in the USA. 9
E. Doden et al. Beaver translocation in degraded desert rivers

We used BRAT as a tool to inform whether our study sites
were near dam capacity, and our analysis indicated that both
our observed pre- and post-translocation reaches with dams
were not well-predicted by BRAT, suggesting the rivers are
far below their maximum dam capacity. Only 1% of the esti-
mated BRAT dam capacity in the Price River watershed was
fulfilled by beavers in a survey conducted by Macfarlane
et al.(2017), suggesting these rivers are below their dam
capacity. Although BRAT predicts the maximum capacity of
dams and not beavers on the landscape, it uses the availability
of vegetation for foraging and dam building to inform the
model (Macfarlane et al., 2017). Even so, there could be other
factors limiting the construction and maintenance of beaver
dams. Translocated and resident beavers in our study sites
may be fulfilling their safety and food accessibility require-
ments without needing to construct dams (Nash et al., 2021).
We note this was especially likely in 2019 when discharges in
our rivers were above historical averages; water depths may
have been naturally sufficient for travel and cover and pre-
vented beavers from building dams during peak discharge.
The subsequent increase in dams may therefore have occurred
even if we did not translocate beavers to the sites because
water flows were relatively ideal for beavers. However, the
increase in the number of rivers reaches with dams post-
translocation compared with pre-translocation in areas with no
known resident beaver activity suggests that translocated bea-
vers built at least some of these new dams. Further, after a
large number of BDAs were built and at least partially blown
out during a large monsoon event in 2021, we observed bea-
vers using the extra wood to build many hybrid bank, partial
channel-spanning dams, which appeared to be enhancing in-
stream habitat complexity.
Establishment success can be related to the number of
individuals released (Morris et al., 2021). We translocated 39
individuals to the Price River and eight to Moonshine Wash
in 21 separate release efforts, potentially leading to establish-
ment success based on our observations of dams being built
in areas with no resident beaver activity, although we were
unable to confirm that translocated beavers constructed these
dams due to transmitter failure and tag loss. Previous beaver
studies have tied success to the translocation of entire colo-
nies or opposite-sex pairs, often requiring multiple release
efforts (McKinstry & Anderson, 2002; Petro et al., 2015;
Brick & Woodruff, 2019). Persistent release efforts have also
contributed to the successful translocation establishment of
Eurasian beaver (Dewas et al., 2012; Halley et al., 2021).
Our research is the first to compare the ecology of resi-
dent beavers to that of translocated beavers in the same sys-
tem. Though our study had limited inference from challenges
associated with long-term monitoring and small sample size,
it represents novel comparative research in an understudied
desert ecosystem. Further research with a well-balanced
design would strengthen the conclusions we present here,
but every beaver translocated requires a significant effort in
terms of staff, money, and time. Simultaneously monitoring
resident beavers alongside translocated beavers allowed us to
directly compare the site fidelity, survival, and dam-building
behavior of the two groups. In general, translocation projects
do not expect every individual to successfully establish at
targeted sites (Berger-Tal et al., 2020), so comparing translo-
cated and resident individuals can help set expectations,
ensure existing populations are not negatively impacted, and
increase the efficacy of future translocations. This strategy
could be useful for translocations of other species.
Beavers were historically widespread and abundant in the
northern hemisphere, impacting virtually every low-gradient,
small-order stream with their dams (Naiman et al., 1988;
Pollock, Heim, & Werner, 2003). Invasive species encroach-
ment, altered flow regimes, simplification, and climate
change limit the restoration of rivers on a landscape scale,
and long-term, watershed-level management can be challeng-
ing to implement (Bennett et al., 2016). However, a water-
shed approach may be necessary to restore structure and
function to degraded rivers and induce population-level
responses in imperiled species tied to these ecosystems
(Bernhardt & Palmer, 2011; Bennett et al., 2016; Bouwes
et al., 2016). Therefore, we propose to expand the definition
of translocation success to include most or all waterways in
a watershed. Provided that the potential for human–wildlife
conflict is addressed and minimized, this large-scale perspec-
tive of success permits higher tolerance for the movement of
translocated beavers away from release sites since even those
individuals who emigrate from release sites could still pro-
vide restoration services in degraded areas. For example,
both rivers in this study occupy a large, impaired watershed
where beavers were likely largely trapped out; if beavers
who emigrated and survived established elsewhere in the
watershed, they would still be contributing to river restora-
tion, just at a much larger scale (i.e., Green River Water-
shed). With a more flexible expectation of success and
further improvements to the trapping and processing of bea-
vers for their survival and health, beaver translocation can be
an effective restoration tool, while giving nuisance beavers a
second chance at life (Pilliod et al., 2018; Nash et al.,
2021). Methods to improve the survival and site fidelity of
translocated beavers at release sites are still needed so that
translocation can serve as a high-impact conservation strat-
egy for rivers when challenges are recognized and mitigated.
Acknowledgments
We thank the U.S. Bureau of Land Management for primary
funding as well as the Utah Division of Wildlife Resources
(UDWR), the U.S. Bureau of Reclamation, the Ecology Cen-
ter at Utah State University (USU), the Utah Chapter of the
Wildlife Society, the U.S. Department of Agriculture-
National Wildlife Research Center, and the U.S. Geological
Survey (in-kind) for additional funding and logistical sup-
port. Gary Thiede (USU, Fish Ecology Lab) provided logisti-
cal support; Peter MacKinnon (USU, Biomark Inc.) and
Daniel Keller (UDWR) provided equipment support. Thanks
to Annette Roug (UDWR), Nate Norman, and the USU Bea-
ver Ecology & Relocation Center for their collaboration. We
are grateful to Tal Avgar (USU) and Nick Bouwes (USU)
for their suggestions on previous manuscript drafts, numer-
ous people who assisted with fieldwork including Hunter
10Animal Conservation  (2022) – ª2022 Zoological Society of London. This article has been contributed to by U.S. Government employees and their work
is in the public domain in the USA.
Beaver translocation in degraded desert rivers E. Doden et al.

Warick, Kelsey Demeny, Christian Hernandez, Richard
Sloan, and Josh Perry, and the landowners who granted
access to their property. Any use of trade, firm, or product
names is for descriptive purposes only and does not imply
endorsement by the United States Government. This study
was performed under the auspices of Utah State University
IACUC protocol 10128 and permitting from the Utah Divi-
sion of Wildlife Resources Certificate of Registration Num-
ber 6COLL10395.
Authors’ contributions
ED, PB, and JKY conceived the ideas and designed the
methodology; ED collected the data; ED and MC analyzed
the data; ED and JKY led the writing of the manuscript. All
authors contributed critically to the drafts and gave final
approval for publication.
References
Albert, S. & Trimble, T. (2000). Beavers are partners in
riparian restoration on the Zuni Indian Reservation. Ecol.
Rest. 18,87–92.
Andersen, D.C. & Shafroth, P.B. (2010). Beaver dams,
hydrological thresholds, and controlled floods as a
management tool in a desert riverine ecosystem, Bill
Williams River, Arizona. Ecohydrology 3, 325–338.
Anderson, D.R., Burnham, K.P. & White, G.C. (1994). AIC
model selection in overdispersed capture-recapture data.
Ecology 75, 1780–1793.
Arjo, W.M., Joos, R.E., Kochanny, C.O., Harper, J.L., Nolte,
D.L. & Bergman, D.L. (2008). Assessment of transmitter
models to monitor beaver Castor canadensis and C. fiber
populations. Wildl. Biol. 14, 309–317.
Armstrong, D. & Seddon, P. (2008). Directions in
reintroduction biology. Trends Ecol. Evol. 23,20–25.
Arnold, T.W. (2010). Uninformative parameters and model
selection using Akaike’s information criterion. J. Wildl.
Mgmt. 74, 1175–1178.
Baker, B.W. & Hill, E.P. (2003). Beaver (Castor canadensis).
In Wild mammals of North America: biology, management,
and conservation. 2nd edn: 288–310. Feldhamer, G.A.,
Thompson, B.C. & Chapman, J.A. (Eds). Baltimore: Johns
Hopkins University Press.
Baker, J.D., Barbieri, M.M., Johanos, T.C., Littnan, C.L.,
Bohlander, J.L., Kaufman, A.C., Harting, A.L., Farry, S.C.
& Yoshinaga, C.H. (2021). Conservation translocations of
Hawaiian monk seals: accounting for variability in body
condition improves evaluation of translocation efficacy.
Anim. Conserv. 24, 206–216.
Barela,I.A.&Frey,J.K.(2016).Habitatandforageselectionby
the American beaver (Castor canadensis)onaregulatedriver
in the Chihuahuan Desert. Southwestern Nat. 61,286–293.
Bennett, S., Pess, G., Bouwes, N., Roni, P., Bilby, R.E.,
Gallagher, S., Ruzycki, J., Buehrens, T., Krueger, K., Ehinger,
W., Anderson, J., Jordan, C., Bowersox, B. & Greene, C.
(2016). Progress and challenges of testing the effectiveness of
stream restoration in the Pacific Northwest using intensively
monitored watersheds. Fisheries 41,92–103.
Berger-Tal, O., Blumstein, D.T. & Swaisgood, R.R. (2020).
Conservation translocations: a review of common difficulties
and promising directions. Anim. Conserv. 23, 121–131.
Bernhardt, E.S. & Palmer, M.A. (2011). River restoration: the
fuzzy logic of repairing reaches to reverse catchment scale
degradation. Ecol. Appl. 21, 1926–1931.
Bloomquist, C.K. & Nielsen, C.K. (2010). Demography of
unexploited beavers in Southern Illinois. J. Wildl. Mgmt. 74,
228–235.
Bonte, D., Van Dyck, H., Bullock, J.M., Coulon, A., Delgado,
M., Gibbs, M., Lehouck, V., Matthysen, E., Mustin, K.,
Saastamoinen, M., Schtickzelle, N., Stevens, V.M.,
Vandewoestijne, S., Baguette, M., Barton, K., Benton, T.G.,
Chaput-Bardy, A., Clobert, J., Dytham, C., Hovestadt, T.,
Meier, C.M., Palmer, S.C.F., Turlure, C. & Travis, J.M.J.
(2012). Costs of dispersal. Biol. Rev. 87, 290–312.
Bottcher, J.L., Walsworth, T.E., Thiede, G.P., Budy, P. &
Speas, D.W. (2013). Frequent usage of tributaries by the
endangered fishes of the Upper Colorado River Basin:
observations from the San Rafael River, Utah. N. Am. J.
Fish. Mgmt. 33, 585–594.
Bouwes, N., Weber, N., Jordan, C.E., Saunders, W.C., Tattam,
I.A., Volk, C., Wheaton, J.M. & Pollock, M.M. (2016).
Ecosystem experiment reveals benefits of natural and
simulated beaver dams to a threatened population of
steelhead (Oncorhynchus mykiss). Sci. Rep. 6,1–13.
Brick, P. & Woodruff, K. (2019). The Methow Beaver Project:
the challenges of an ecosystem services experiment. Case
Stud. Environ. 3,1–14.
Budy, P., Conner, M.M., Salant, N.L. & Macfarlane, W.W.
(2015). An occupancy-based quantification of the highly
imperiled status of desert fishes of the southwestern United
States. Conserv. Biol. 29, 1142–1152.
Burnham, K.P. & Anderson, D.R. (2002). Model selection and
multimodel inference: a practical information-theoretic
approach, 2nd edn. New York: Springer.
Burham, K.P., Anderson, D.R., White, G.C., Brownie, C. &
Pollock, K.H. (1987). Design and analysis of fish survival
experiments based on release-recapture data. Am. Fish. Soc.
Monograph 5.
Campbell, R.D., Nouvellet, P., Newman, C., Macdonald, D.W.
& Rosell, F. (2012). The influence of mean climate trends
and climate variance on beaver survival and recruitment
dynamics. Glob. Change Biol. 18, 2730–2742.
Campbell-Palmer, R. & Rosell, F. (2015). Captive care and
welfare considerations for beavers. Zoo Biol. 34, 101–
109.
Cormack, R.M. (1964). Estimates of survival from the sighting
of marked animals. Biometrika 51, 429–438.
Demmer, R. & Beschta, R.L. (2008). Recent history (1988–
2004) of beaver dams along Bridge Creek in Central
Oregon. Northwest Sci. 82, 309–318.
Animal Conservation  (2022) – ª2022 Zoological Society of London. This article has been contributed to by U.S. Government employees and their work is
in the public domain in the USA. 11
E. Doden et al. Beaver translocation in degraded desert rivers

DeStefano, S., Koenen, K.K.G., Henner, C.M. & Strules, J.
(2006). Transition to independence by subadult beavers
(Castor canadensis) in an unexploited, exponentially
growing population. J. Zool. 269, 434–441.
DeVries, P., Fetherston, K.L., Vitale, A. & Madsen, S. (2012).
Emulating riverine landscape controls of beaver in stream
restoration. Fisheries 37, 246–255.
Dewas, M., Herr, J., Schley, L., Angst, C., Manet, B., Landry,
P. & Catusse, M. (2012). Recovery and status of native and
introduced beavers Castor fiber and Castor canadensis in
France and neighbouring countries. Mam. Rev. 42, 144–165.
Dickens, M.J., Delehanty, D.J. & Romero, M.L. (2010).
Stress: an inevitable component of animal translocation.
Biol. Conserv. 143, 1329–1341.
Dittbrenner, B.J. (2019). Restoration potential of beaver for
hydrological resilience in a changing climate. Ph.D.
Dissertation, University of Washington.
Doden, E., Budy, P., Avgar, T. & Young, J.K. (2022).
Movement patterns of resident and translocated beavers at
multiple spatiotemporal scales in desert rivers. Front.
Conserv. Sci. 3, 777797.
Fairfax, E. & Small, E.E. (2018). Using remote sensing to
assess the impact of beaver damming on riparian
evapotranspiration in an arid landscape. Ecohydrology 11,
e1993.
Fairfax, E. & Whittle, A. (2020). Smokey the beaver: beaver-
dammed riparian corridors stay green during wildfire
throughout the western United States. Ecol. Appl. 30,
e02225.
Fischer, J. & Lindenmayer, D.B. (2000). An assessment of the
published results of animal relocations. Biol. Conserv. 96,
1–11.
Gable, T.D. & Windels, S.K. (2018). Kill rates and predation
rates of wolves on beavers. J. Wildl. Mgmt. 82, 466–472.
Germano, J.M., Field, K.J., Griffiths, R.A., Clulow, S., Foster,
J., Harding, G. & Swaisgood, R.R. (2015). Mitigation-
driven translocations: are we moving wildlife in the right
direction? Front. Ecol. Environ. 13, 100–105.
Gibson, P.P. & Olden, J.D. (2014). Ecology, management, and
conservation implications of north American beaver (Castor
canadensis) in dryland streams. Aquat. Conserv. Mar.
Freshw. Ecosyst. 24, 391–409.
Griffith, B., Scott, J.M., Carpenter, J.W. & Reed, C. (1989).
Translocation as a species conservation tool: status and
strategy. Science 245, 477–480.
Halley, D.J., Saveljev, A.P. & Rosell, F. (2021). Population
and distribution of beavers Castor fiber and Castor
canadensis in Eurasia. Mam. Rev. 51,1–24.
Hood, G.A. & Bayley, S.E. (2008). Beaver (Castor
canadensis) mitigate the effects of climate on the area of
open water in boreal wetlands in western Canada. Biol.
Conserv. 141, 556–567.
Jolly, G.M. (1965). Explicit estimates from capture-recapture
data with both death and immigration - stochasitic model.
Biometrika 52, 225–247.
Kingsford, R.T. & Thompson, J.R. (2006). Desert or dryland
rivers of the world: an introduction. In Ecology of desert
rivers:1–10. Kingsford, R. (Ed.). Cambridge: Cambridge
University Press.
Knopf, F.L., Johnson, R.R., Rich, T., Samson, F.B. & Szaro,
R.C. (1988). Conservation of riparian ecosystems in the
United States. Wilson Bull. 100, 272–284.
Laub, B.G. (2015). Lower San Rafael River and riparian
corridor restoration: BLM land phase 1. Utah Water
Restoration Initiative.
Laub, B.G. (2018). Lower San Rafael River and riparian
corridor habitat improvement: phase 1a. Utah Water
Restoration Initiative.
Lebreton, J.-D., Burnham, K.P., Clobert, J. & Anderson, D.R.
(1992). Modeling survival and testing biological hypotheses
using marked animals: a unified approach with case studies.
Ecol. Monogr. 62,67–118.
Letty, J., Marchandeau, S. & Aubineau, J. (2007). Problems
encountered by individuals in animal translocations: lessons
from field studies.

Ecoscience 14, 420–431.
Lyster, S. (2018). San Rafael river restoration project progress
report: gravel study. Utah State University.
Macfarlane, W.W., Wheaton, J.M., Bouwes, N., Jensen, M.L.,
Gilbert, J.T., Hough-Snee, N. & Shivik, J.A. (2017).
Modeling the capacity of riverscapes to support beaver
dams. Geomorphology 277,72–99.
Macfarlane, W.W., Wheaton, J.M. & Jensen, M.L. (2014). The
Utah Beaver Restoration Assessment Tool: a decision
support and planning tool. Logan: Ecogeomorphology and
Topographic Analysis Lab, Utah State University.
Maenhout, J. (2013). Beaver ecology in Bridge Creek, a
tributary to the John Day River. M.S. Thesis, Oregon State
University.
Matykiewicz, B.R., Windels, S.K., Olson, B.T., Plumb, R.T.,
Wolf, T.M. & Ahlers, A.A. (2021). Assessing translocation
effects on the spatial ecology and survival of muskrats
Ondatra zibethicus.Wildl. Biol. 2021, wlb-00823.
Mayer, M., Zedrosser, A. & Rosell, F. (2017). When to leave:
the timing of natal dispersal in a large, monogamous rodent,
the Eurasian beaver. Anim. Behav. 123, 375–382.
McKinstry, M.C. & Anderson, S.H. (2002). Survival, fates,
and success of transplanted beavers, Castor canadensis,in
Wyoming. Can. Field Nat. 116,60–68.
McNew, L.B., Jr. & Woolf, A. (2005). Dispersal and survival
of juvenile beavers (Castor canadensis) in Southern Illinois.
Am. Midland Nat. 154, 217–228.
Mengak, M.T. (2018). Wildlife translocation. In Wildlife
damage management technical series. Fort Collins: USDA,
APHIS, WS National Wildlife Research Center. 15p.
Menge, B.A. & Sutherland, J.P. (1987). Community
regulation: variation in disturbance, competition, and
predation in relation to environmental stress and recruitment.
Am. Nat. 130, 730–757.
Morris, S.D., Brook, B.W., Moseby, K.E. & Johnson, C.N.
(2021). Factors affecting success of conservation
12Animal Conservation  (2022) – ª2022 Zoological Society of London. This article has been contributed to by U.S. Government employees and their work
is in the public domain in the USA.
Beaver translocation in degraded desert rivers E. Doden et al.

translocations of terrestrial vertebrates: a global systematic
review. Global Ecol. Conserv. 28, e01630.
Mott Lacroix, K.E., Tapia, E. & Springer, A. (2017).
Environmental flows in the desert rivers of the United
States and Mexico: synthesis of available data and gap
analysis. J. Arid Environ. 140,67–78.
M€
uller-Schwarze, D. (2011). The beaver: its life and impact,
2nd edn. Ithaca: Cornell University Press, Comstock Pub.
Associates.
Muriel, R., Balbont
ın, J., Calabuig, C.P., Morlanes, V. &
Ferrer, M. (2021). Does translocation affect short-term
survival in a long-lived species, the Spanish imperial eagle?
Anim. Conserv. 24,38–50.
Naiman, R.J., Johnston, C.A. & Kelley, J.C. (1988). Alteration
of North American streams by beaver. Bioscience 38, 753–
762.
Nash, C.S., Grant, G.E., Charnley, S., Dunham, J.B., Gosnell,
H., Hausner, M.B., Pilliod, D.S. & Taylor, J.D. (2021).
Great expectations: deconstructing the process pathways
underlying beaver-related restoration. Bioscience 71, 249–
267.
National Oceanic & Atmospheric Administration. (2021a).
Data tools: 1981–2010 annual/seasonal normals. National
Centers for Environmental Information. https://www.ncdc.
noaa.gov/cdo-web/datatools/normals (Accessed 04 April
2021).
National Oceanic & Atmospheric Administration. (2021b).
Global historical climatology network –daily. https://www.
climate.gov/maps-data/dataset/past-weather-zip-code-data-
table (Accessed 14 March 2021).
Nolet, B.A. & Baveco, J.M. (1996). Development and
viability of a translocated beaver Castor fiber population in
The Netherlands. Biol. Conserv. 75, 125–137.
Novak, B.J., Phelan, R. & Weber, M. (2021). U.S.
conservation translocations: over a century of intended
consequences. Conserv. Sci. Pract. 3, e394.
Omernik, J.M. (1987). Ecoregions of the conterminous United
States. Annals Assoc. Am. Geogr. 77,118–125.
Patric, E.F. & Webb, W.L. (1960). An evaluation of three age
determination criteria in live beavers. J. Wildl. Mgmt. 24,
37–44.
Persico, L. & Meyer, G. (2013). Natural and historical
variability in fluvial processes, beaver activity, and climate
in the Greater Yellowstone Ecosystem. Earth Surf. Process.
Landforms 38, 728–750.
Petro, V. (2013). Evaluating “nuisance”beaver relocation as
a tool to increase Coho salmon habitat in the Alsea Basin
of the Central Oregon Coast Range. M.S. Thesis, Oregon
State University.
Petro, V.M., Taylor, J.D. & Sanchez, D.M. (2015). Evaluating
landowner-based beaver relocation as a tool to restore
salmon habitat. Global Ecol. Conserv. 3, 477–486.
Petro, V.M., Taylor, J.D., Sanchez, D.M. & Burnett, K.M.
(2018). Methods to predict beaver dam occurrence in
coastal Oregon. Northwest Sci. 92, 278–289.
Pilliod, D.S., Rohde, A.T., Charnley, S., Davee, R.R.,
Dunham, J.B., Gosnell, H., Grant, G.E., Hausner, M.B.,
Huntington, J.L. & Nash, C. (2018). Survey of beaver-
related restoration practices in rangeland streams of the
Western USA. Environ. Mgmt. 61,58–68.
Pinter-Wollman, N., Isbell, L.A. & Hart, L.A. (2009). Assessing
translocation outcome: comparing behavioral and physiological
aspects of translocated and resident African elephants
(Loxodonta africana). Biol. Conserv. 142,1116–1124.
Pollock, M.M., Beechie, T.J., Wheaton, J.M., Jordan, C.E.,
Bouwes, N., Weber, N. & Volk, C. (2014). Using beaver
dams to restore incised stream ecosystems. Bioscience 64,
279–290.
Pollock, M.M., Heim, M. & Werner, D. (2003). Hydrologic
and geomorphic effects of beaver dams and their influence
on fishes. Am. Fish. Soc. Symp. 37, 213–233.
R Development Core Team. (2020). R: a language and environment
for statistical computing. Vienna, Austria: R Foundation for
Statistical Computing. https://www.R-project.org/.
Remiszewski, T.T. (2022). Extreme, positive geomorphic
change in a historically degraded desert river: implications
for imperiled fishes. M.S. Thesis, Utah State University,
Logan, UT, 8640. https://digitalcommons.usu.edu/etd/8640.
Ritter, T.D. (2018). Ecosystem pioneers: beaver dispersal and
settlement site selection in the context of habitat restoration.
M.S. Thesis, Montana State University.
Rosell, F., Bozser, O., Collen, P. & Parker, H. (2005).
Ecological impact of beavers Castor fiber and Castor
canadensis and their ability to modify ecosystems. Mam.
Rev. 35, 248–276.
Rothmeyer, S.W., McKinstry, M.C. & Anderson, S.H. (2002).
Tail attachment of modified ear-tag radio transmitters on
beavers. Wildl. Soc. Bull. 30, 425–429.
Roug, A., Talley, H., Davis, T., Roueche, M. & DeBloois, D.
(2018). A mixture of butorphanol, azaperone, and
medetomidine for the immobilization of American beavers
(Castor canadensis). J. Wildl. Dis. 54, 617–621.
Sada, D.W. & Stevens, L.E. (2021). Conservation and ecological
rehabilitation of North American desert spring ecosystems. In
Standing between life and extinction: ethics and ecology of
conserving aquatic species in north American deserts. Propst,
D.L., Williams, J.E., Bestgen, K.R. & Hoagstrom, C.W. (Eds).
Chicago: University of Chicago Press.
Schaub, M. & Royle, A. (2014). Estimating true instead of
apparent survival using spatial Cormack-Jolly-Seber models.
Methods Ecol. Evol. 5, 1316–1326.
Schulte, B.A., M€
uller-Schwarze, D. & Sun, L. (1995). Using
anal gland secretion to determine sex in beaver. J. Wildl.
Mgmt. 59, 614–618.
Seber, G.A.F. (1965). A note on the multiple-recapture census.
Biometrika 52, 249–259.
Stromberg, J.C. (2001). Restoration of riparian vegetation in
the South-Western United States: importance of flow
regimes and fluvial dynamism. J. Arid Environ. 49,
17–34.
Animal Conservation  (2022) – ª2022 Zoological Society of London. This article has been contributed to by U.S. Government employees and their work is
in the public domain in the USA. 13
E. Doden et al. Beaver translocation in degraded desert rivers

Teixeira, C.P., de Azevedo, C.S., Mendl, M., Cipreste, C.F. &
Young, R.J. (2007). Revisiting translocation and
reintroduction programmes: the importance of considering
stress. Anim. Behav. 73,1–13.
Touihri, M., Labb
e, J., Imbeau, L. & Darveau, M. (2018).
North American beaver (Castor canadensis Kuhl) key
habitat characteristics: review of the relative effects of
geomorphology, food availability and anthropogenic
infrastructure.

Ecoscience 25,9–23.
Udall, B. & Overpeck, J. (2017). The twenty-first century
Colorado River hot drought and implications for the future.
Water Resour. Res. 53, 2404–2418.
United States Geological Survey. (2021). USGS current water
data for Utah. National Water Information System: Web
Interface. https://waterdata.usgs.gov/ut/nwis/rt (Accessed 12
March 2021).
Utah Division of Wildlife Resources. (2017). Protocol for live
trapping, holding and transplanting beaver. Utah Division
of Wildlife Resources.
Ver Hoef, J.M. (2012). Who invented the Delta method? Am.
Stat. 66, 124–127.
White, G.C. & Burnham, K.P. (1999). Program MARK:
survival estimation from populations of marked animals.
Bird Study 46, S120–S139.
Williams, A.P., Cook, B.I. & Smerdon, J.E. (2022). Rapid
intensification of the emerging southwestern north American
megadrought in 2020–2021. Nature Climate Change. 12,
232–234.
Windels, S.K. & Belant, J.L. (2016). Performance of
tail-mounted transmitters on American beavers Castor
canadensis in a northern climate. Wildl. Biol. 22,
124–129.
Wolf, C.M., Griffith, B., Reed, C. & Temple, S.A. (1996).
Avian and mammalian translocations: update and reanalysis
of 1987 survey data. Conserv. Biol. 10, 1142–1154.
Woodford, J.E., Macfarland, D.M. & Worland, M. (2013).
Movement, survival, and home range size of translocated
American martens (Martes Americana) in Wisconsin. Wildl.
Soc. Bull. 37, 616–622.
Woodruff, K. & Pollock, M.M. (2018). Relocating beaver. In
The beaver restoration guidebook: working with beaver to
restore streams, wetlands, and floodplains:64–84. Pollock,
M.M., Lewallen, G.M., Woodruff, K., Jordan, C.E. &
Castro, J.M. (Eds). Portland. United States Fish & Wildlife
Service.
Wright, J.P., Jones, C.G. & Flecker, A.S. (2002). An
ecosystem engineer, the beaver, increases species richness at
the landscape scale. Oecologia 132,96–101.
Supporting information
Additional supporting information may be found online in
the Supporting Information section at the end of the article.
Table S1. Key to parameter codes used for AICc table of
eight-week Cormack-Jolly-Seber models of apparent survival
probability for resident and translocated beavers in the desert
Price and San Rafael Rivers of Utah, USA, May–October
2019 and 2020.
14Animal Conservation  (2022) – ª2022 Zoological Society of London. This article has been contributed to by U.S. Government employees and their work
is in the public domain in the USA.
Beaver translocation in degraded desert rivers E. Doden et al.