BioControl (2012) 57:331–347
DOI 10.1007/s10526-011-9436-9
Tamarisk biocontrol, endangered species risk and resolution
of conflict through riparian restoration
Tom L. Dudley • Daniel W. Bean
Received: 1 April 2011 / Accepted: 18 December 2011 / Published online: 30 December 2011
International Organization for Biological Control (IOBC) 2011
Abstract A long-standing debate between wildlife
agencies and biological control researchers and practitioners concerns Diorhabda carinulata Desbrochers
(Coleoptera: Chrysomelidae) introduced to suppress
invasive Tamarix spp. (Tamaricaceae), and potential
impacts of Tamarix defoliation on endangered southwestern willow flycatchers using this non-native plant as
nesting habitat in some western riparian ecosystems.
The conflict and ensuing legal actions are currently
centered on the presence of D. carinulata within the
breeding range of the flycatcher in the Virgin River
watershed, which has led to APHIS termination of
permits supporting the biocontrol development program
and has also affected other programs to develop
biocontrol agents against environmental weeds. Central
to concerns over wildlife is the lack of rehabilitation of
native vegetation where biocontrol is expected, so there
are current and planned efforts to promote restoration of
native cottonwood-willow habitat to mitigate the anticipation decline in Tamarix cover. A strategic approach
Handling Editor: Mark Hoddle
T. L. Dudley (&)
Marine Science Institute, University of California,
Santa Barbara, CA 93106-6150, USA
e-mail: tdudley@msi.ucsb.edu
D. W. Bean
Colorado Department of Agriculture, Palisade Insectary,
750 37.8 Rd., Palisade, CO 81526, USA
e-mail: Dan.Bean@ag.state.co.us
to riparian restoration is outlined which could facilitate
sustainable, and scientifically documented recovery of
this iconic habitat type. While the results of these efforts
will not be known immediately, the process which is
leading to riparian restoration has brought specialists
from both sides of the debate together in search of
resolution via collaboration, and if successful, may
allow re-initiation of the Tamarix biocontrol program
attendant with habitat enhancement for wildlife species
of conservation concern.
Keywords Tamarisk Biological control Southwestern willow flycatcher Ecological
restoration Riparian ecosystem Tamarix Diorhabda Coleoptera Chrysomelidae
Introduction
The introduction of saltcedar leaf beetles, Diorhabda
spp. Weise, for biological control of invasive tamarisk
(Tamarix spp.; also known as saltcedar) has been one of
the most visually dramatic weed biocontrol programs
ever developed (DeLoach et al. 2004; Hudgeons et al.
2007; Bateman et al. 2010). Following one or two years
of initial establishment at several release sites in
western North America, insect dispersal and subsequent
feeding by larval and adult insects numbering in excess
of 1,000 per tree, near-complete defoliation of plants
over large areas has been observed (Tracy and Robbins
123
332
2009; Pattison et al. 2010). In particular the establishment of D. carinulata Desbrochers in Nevada, USA and
several other western states, now including dispersal
into Arizona, has led to thousands of hectares defoliated, often two or three times over the course of a single
season with multiple generations of beetles (Bean et al.
2007; Pattison et al. 2010; Dalin et al. 2010). This
success has met not with cheers of approval, but instead
has exacerbated a long-running conflict between weed
control measures and protection of endangered species
(Dudley and DeLoach 2004) that not only has greatly
hindered the expansion and evaluation of this biocontrol
program, but has repercussions for biocontrol development programs targeting many environmental weeds.
This paper presents key information related to the
tamarisk biocontrol program, and is also intended to
explore the opportunity to promote habitat restoration in
parallel with weed control as a practical approach to
moving beyond conflict into facilitating ecosystem
recovery and biodiversity enhancement in the context of
invasive plant suppression.
Background and statement of problem
Tamarix biocontrol program history
Tamarix species were introduced into North America
from Eurasia roughly two centuries ago, but expanded
dramatically in western riparian systems in the early
part of the twentieth century as river regulation, water
diversion and human occupation generally led to
conditions favoring its invasion (Horton 1977; Shafroth et al. 2005). This expansion, including into many
watersheds that were relatively undegraded and
retained largely natural hydrologic regimes (Cleverly
et al. 1997; Whiteman 2006; Mortenson and Weisberg
2010; Merritt and Poff 2010), was associated with a
decline of native riparian cottonwood-willow (Populus spp., Salix spp.) woodlands (Howe and Knopf
1991) as well as mesquite (Prosopis spp.) bosque and
other native vegetation complexes (Busch and Smith
1995; D’Antonio and Dudley 1997). Tamarix is now
the third most common woody plant in western US
riparian areas (Friedman et al. 2005), and is considered
a noxious weed in most states west of the Mississippi
River. Tamarisk dominance has led to alterations of
riparian ecosystems including soil salinization, local
reduction in water resources via excess transpiration
123
T. L. Dudley, D. W. Bean
(Sala et al. 1996; Dahm et al. 2002), changing erosion
and sedimentation regimes (Graf 1978; Vincent et al.
2009), and increasing the susceptibility of riparian
areas to wildfire (Busch and Smith 1993; Dudley
et al. 2011) although the environmental impact of
some of these factors has been questioned (Sogge
et al. 2008; Hultine et al. 2010). Tamarix spp. are
also well-known to provide poor habitat to wildlife
species, at least relative to the native riparian
vegetation that has often been replaced (Ellis 1995;
Hunter et al. 1988; Shafroth et al. 2005; Bateman and
Paxton 2010).
These apparent impacts have led to many programs
to reduce Tamarix abundance in riparian systems
(Shafroth et al. 2005), including the development of
biological control to suppress infestations (Dudley
et al. 2000; DeLoach et al. 2004). Overseas exploration for specialist herbivores suitable for biocontrol
development was initiated in the 1970s by Lloyd
Andres and Robert Pemberton, with the active support
of wildlife agencies, including the US Fish & Wildlife
Service (FWS; Stenquist 1999), and resulted in over
300 arthropods that were sufficiently specialized to
be considered as potential candidates for biocontrol
development (DeLoach et al. 1996). Of these, a
chrysomelid leaf beetle, Diorhabda elongata Brullé
was chosen along with several others for further
efficacy and host range testing. In 1996 D. elongata
carinulata (=D. elongata deserticola Chen), reclassified as D. carinulata (Tracy and Robbins 2009),
received approval from the USDA Animal and Plant
Health Inspection Service (APHIS) for field release,
along with a mealy bug (Trabutina mannipara) suited
to warm desert regions and later, a Mediterranean
weevil (Coniatus tamarisci) (Tracy and Robbins
2009).
About the same time, the southwestern sub-species
of willow flycatcher (Empidonax traillii extimus) was
listed in 1995 as endangered by the US Fish and
Wildlife Service, and was documented to use Tamarix
as nesting habitat in several locations in the lower
Colorado River system and New Mexico (Sogge et al.
2003). Although displacement of native riparian
vegetation by non-native plants was considered an
important factor leading to endangerment of this bird
(Federal Register 1995), Tamarix was a common
nesting substrate in the region of proposed Critical
Habitat (Fig. 1) and so it became necessary for the
USDA Agricultural Research Service (ARS) and
Tamarisk biocontrol, endangered species risk and resolution
333
Fig. 1 Southwestern
willow flycatcher
approximate breeding range
shown by shaded envelope,
with designated Critical
Habitat indicated with
darkened rivercourse lines,
including the Virgin River in
the northern extent of the
birds’s range. The four other
sub-species of willow
flycatcher that breed north or
east of this zone, while
declining, are not formally
protected
APHIS, and its co-operating researchers, to enter into
Section 7 Consultation with the FWS to assess
whether biocontrol would potentially lead to a ‘taking’
of this listed species (Malakoff 1999; Dudley and
DeLoach 2004). After extensive delays and negotiations between federal agencies, restrictions were
placed on the program, including the elimination of
proposed field release sites within an arbitrary
200-mile buffer zone around any known nesting of
the flycatcher in Tamarix, and in 1999 cage releases
went forward in the field at sites in Colorado, Nevada,
California, Utah and other states well outside the
flycatcher zone (Dudley et al. 2001).
Open field releases were subsequently done at sites
where cage trials had been successful, including sites
in northern Nevada (Humboldt and Walker River
Basins) and central Utah (Sevier River). At those sites
establishment and defoliation by D. carinulata was
dramatic (Pattison et al. 2010), leading to substantial
target mortality over the course of three or more years
(Fig. 2; Hudgeons et al. 2007). In addition, the
introduction of this herbivore has lead to an upwardly
cascading trophic effect, with increased abundance of
generalist invertebrate predators (Bateman et al. 2010)
and enhanced food availability for insectivorous
wildlife species (Longland and Dudley 2008). Such
resource enhancement has led to increased diversity
and abundance of insectivorous passerines as observed
at our northern Nevada study area (Fig. 3). The
biocontrol program transitioned to its implementation
stage for states above 38 North latitude (Kaufmann
2005), but the ‘no release’ policy was retained within
the southwestern willow flycatcher buffer zone.
However, that policy was informally modified without
discussion as referring to any known presence of the
bird, regardless of whether Tamarix was being used
for habitat [‘‘biocontrol agents should not be used
within the occupied range of the southwestern willow
flycatcher’’ (Federal Register 2005)]. Greater detail
regarding the Tamarix biocontrol program can be
found elsewhere (DeLoach et al. 2004; Tracy and
Robbins 2009; Dalin et al. 2010).
The primary concern of wildlife agencies was that
Tamarix defoliation during nesting could reduce
tamarisk canopy cover over the nest site, exposing
birds in the nest to excessive heating and desiccation, particularly because this species breeds relatively late in the season (June–July) (Sogge et al.
2003, 2008). Other worries were that defoliation also
increased exposure of birds to predators and nest
parasites, and even that the beetles themselves may
be toxic. In some regions it was projected that
stream conditions were too degraded to support reestablishment of native riparian plants, so Tamarix
suppression might remove the only vegetation that
tolerates anthropogenically damaged environments
123
334
Fig. 2 Herbivore-induced mortality of T. ramosissima trees
following D. carinulata establishment at the Humboldt River,
Nevada research release location. Bars indicate proportion of
trees surviving in subsequent years, with solid bars representing
the original release site which received intense repeat defoliation starting in 2002 following release in 2001. Cross-hatched
bars are for a site 2 km distant where biocontrol insects fully
established in 2004, and defoliation occurred repeatedly but
with lesser intensity. Data are for the same individual trees
monitored regularly at each site (n = 48 at release site, n = 30
at 2 km site), not a statistical sampling so no statistical error
estimates are shown. Monitoring was discontinued in 2009
when numerous trees were lost as a consequence of fire and
landowner management practices
(Stromberg et al. 2007; Sogge et al. 2008; Hultine
et al. 2010). It is likely correct that altered hydrology
and degraded conditions in such locations present
major barriers to re-establishment of native vegetation. However, the willow flycatcher no longer nests
in the systems in question, and with rare exception
occupies systems where some native riparian component is present, so recovery potential is realistic
(Dudley and DeLoach 2004).
Biocontrol establishment and controversy
in the flycatcher zone
In 2006 resource managers in southern Utah transported Diorhabda beetles from the Sevier River
research release site in central Utah to control Tamarix
along the Virgin River in the vicinity of St. George in
southern Utah (Bateman et al. 2010). This brought the
beetles into proximity to the willow flycatcher, which
had been documented to nest in both tamarisk and
native vegetation at several locations in the Virgin
watershed (McLeod and Koronkiewicz 2010). We
initiated monitoring of this population soon afterwards
as it dispersed incrementally downstream and across
the watershed, from its headwaters in Zion National
123
T. L. Dudley, D. W. Bean
Park through the NW corner of Arizona and the Virgin
Valley of southern Nevada (Bateman et al. 2010;
Dudley et al. 2011). The D. carinulata population in
the Virgin River has now converged with similar
dispersal down a tributary stream, the Muddy River, at
Lake Mead (Fig. 4).
Knowing the extent of host defoliation observed at
the original release sites, wildlife managers reacted
with alarm to this unauthorized transfer (not illegal, as
APHIS authorizes interstate transport, but does not
regulate movements of biocontrol agents within a
state’s boundaries). Federal regulators made the
unusual decision to promote a media campaign in an
attempt to halt and discredit the biocontrol program,
even to the point of distributing Wild West-themed
posters stating ‘‘not wanted in Arizona: Tamarisk leaf
beetles’’. The growing controversy led to a lawsuit
filed against APHIS and FWS by the Center for
Biological Diversity and the Maricopa County Audubon Society, demanding that these federal agencies
re-enter negotiations on the unsupported grounds that
USDA participated in this release and also that the
agency reneged on its assurance that no releases would
occur near willow flycatcher breeding areas. As an
example of the hyperbole that has unfortunately
surrounded this debate, the following quotation is
from a media release by the Center for Biological
Diversity regarding its lawsuit: ‘‘We face loss of the
flycatcher in the Southwest because APHIS has broken
its promises and refuses to take responsibility for its
actions. We now must appeal to the courts to help us
save this adorable little migratory songbird’’ (CDB
Center for Biological Diversity 2009). Because of this
pending lawsuit, APHIS terminated consideration of
PPQ 526 interstate transport permits and cancelled all
existing permits for Diorhabda spp., thereby leaving
current research and some monitoring programs
without regulatory and logistical support, with admonitions to cease actions related to these permits with
potential fines of US$250,000 (APHIS 2010). This
cancellation affected programs nationwide, even in
areas where releases had already been conducted
and which were far from the southwestern willow
flycatcher breeding range. The controversy caused
consternation amongst practitioners, who in places as
distant as Oregon, Colorado and Texas feared continuing existing Tamarix biocontrol programs and even
considered insecticide treatments for sites where
Diorhabda was already established.
Tamarisk biocontrol, endangered species risk and resolution
335
Fig. 3 Abundances of
passerine birds at the Walker
River, Nevada research
release site, shown as mean
number of observations per
transect over time within the
Diorhabda establishment
zone (black bars) and nearby
where beetles had not yet
colonized (grey bars); error
values only given for those
species observed on either
transect for at least six of the
eight census dates. Species
that were significantly more
common in the presence of
beetles are indicated by
asterisks (t-test P \ 0.05)
while none was more
common in their absence
(tested for the six species
observed on Csix survey
dates). When the two
blackbird species (Redwing, Brewer’s) were
removed from analysis
because these omnivores
forage aerially or on the
ground, rather than gleaning
insects from branches,
overall bird abundance was
also greater where
Diorahbda was present
(paired t-test: t = 2.90,
df = 18, P = 0.0096)
(Dudley et al. 2009;
Longland, Hitchcock and
Dudley, unpub data)
Adding to the controversy was the issue of how far
south Tamarix biocontrol agents might establish.
At a latitude of 37.1 North D. carinulata was near
its original physiological southern limit in the upper
Virgin River, restricted by day length-induced reproductive diapause as the Critical Day Length comes
increasingly early in the season at more southerly
latitudes (Bean et al. 2007). Population establishment
was, however, quite robust (Fig. 4) indicating an
improved capacity to delay response to this diapause
cue. The ability for Diorhabda to expand its population extent was a novel, but not unexpected, factor
resulting from natural selection for later diapause
induction in the southern extent of insect range (Dalin
et al. 2010; Bean et al. 2012) but the rate of continued
southward progress beyond this indistinct boundary
may be slow owing to the selection process requiring
multiple generations for effective expression. Nonetheless, wildlife managers feared that beetles would
rapidly overwhelm tamarisk-dominated riparian systems throughout the lower Colorado River region (as
well as the upper Rio Grande system), potentially
putting other southwestern willow flycatcher populations at risk (Sogge et al. 2008). In addition,
unregulated biocontrol agent transfers, or ‘guerrilla
introductions’, are suspected to have occurred elsewhere, and the potential exists for such transfers into
southwestern willow flycatchers (SWFL) habitat
using other Diorhabda species adapted to day length
regimes in southern regions and already established in
123
336
T. L. Dudley, D. W. Bean
Fig. 4 Distribution of
Diorhabda carinulata
(indicated by grey shading)
as of September 2011,
following dispersal from the
initial release sites in St.
George, Utah. Most
colonization occurred from
movement downstream, and
upstream, in the Virgin
River corridor, while some
establishment was from
disjunct dispersal into
tributary systems, such as
the Meadow Valley Wash/
Muddy River population
which converged in 2011
with the Virgin River
population at Lake Mead,
and also from St. George to
the mainstem Colorado
River via Paria Canyon and/
or Kanab Creek. A different
dispersal process not shown
is down the Colorado River
from Utah and Lake Powell
places such as Texas and California (Dalin et al. 2010;
Tracy and Robbins 2009).
The degree of controversy, legal dispute and
anxiety among all stakeholders increased the need to
seek resolution and move forward, and in particular to
address the long-recognized need to facilitate riparian
ecosystem recovery as Tamarix biocontrol proceeded
both in impact to target plants and geographic extent
of Diorhabda establishment (Dudley and DeLoach
2004). This has been the key complaint of wildlife
regulators since the initial findings of SWFL nesting in
Tamarix: in the absence of a restoration plan for many
Tamarix-infested watersheds, its suppression would
be detrimental to the bird’s sustained presence (Sogge
et al. 2008). The original sites for Tamarix biocontrol
research testing were nominated by local stakeholders
who anticipated that, if successful in suppressing
Tamarix infestations, biocontrol would be followed by
either direct restoration or anticipated natural recruitment of desirable species. In most locations, sufficient
native plant populations, particularly of the Salix and
Populus species (Salicaceae) earlier displaced by
Tamarix invasion, were present to provide propagules
(mainly seeds) to recruit under suitable hydrological
conditions. These are the willows and cottonwoods
123
that are well-known to produce pulses of establishment following flood events that create environmental
conditions favoring their germination (Stella et al.
2006a). However, in some of the more heavily
degraded riparian systems of the southwestern US
these iconic taxa are greatly reduced in abundance and
environmental conditions may not be suitable for
supporting native riparian recovery (Stromberg et al.
2007). Thus, wildlife managers assumed that either it
was not possible to replace Tamarix with native plants
that could support the willow flycatcher, or would take
extraordinary efforts well ahead of biocontrol implementation to generate sufficient habitat to support the
bird. Although these legitimate concerns are, in fact,
largely irrelevant in much of the southwestern desert
region where the SWFL no longer breeds, resistance
to Tamarix biocontrol has not diminished, with no
movement toward legal resolution of current disputes.
A recent development further complicates the
biocontrol dilemma, that is the presence of another
specialist herbivore of Tamarix, a tamarisk weevil,
possibly Coniatus splendidulus, recently documented
in the Virgin watershed (Eckberg and Foster 2011;
Dudley et al. 2011). While its taxonomic identity is
not yet clear, this 3-mm insect is very similar in
Tamarisk biocontrol, endangered species risk and resolution
appearance and behavior to C. tamarisci considered
previously for Tamarix biocontrol (Fornasari 1998).
There is no evidence, however, that the insect
maintained in quarantine several years was the source
of this population which is now widespread in the
desert southwest. Nonetheless, the eventual effect of
Coniatus herbivory on Tamarix status in the Virgin
River and elsewhere is unclear, but it does reinforce
the concept that we must move beyond the notion of
somehow stopping ‘biocontrol’, and developing ecosystem management approaches that incorporate
Tamarix herbivory into riparian ecosystem dynamics.
This is primarily a policy debate, not a scientific one,
yet resolution requires marshalling scientific evidence
and opinion through the political/regulatory gauntlet—
towards illustrating that conservation objectives can be
rationally met in the course of managing tamarisk
infestations. If resolution can be found, it has broad
implications as the legal and regulatory conflict arising
from the Tamarix-Diorhabda-willow flycatcher situation also interferes with the development of biocontrol
for other environmental weeds (e.g., Smith and Cristofaro 2010). Recent events have, thus, precipitated
both the ecological and the regulatory conditions that
make it important to overcome this impasse, bringing
concerned parties together to develop a path forward
for endangered species conservation and, in turn,
possible renewed activity for the Tamarix biocontrol
program. The central focus of the current discussion
involves co-operative actions on promoting passive
and active recovery of native riparian vegetation, and
the ecological context for this process follows.
Restoration for endangered species habitat
Restoration and the willow flycatcher
Restoration ecology is a subset of conservation
biology, a multi-disciplinary field that seeks to
promote biodiversity protection along with rehabilitation of ecosystem functions that sustain both environmental and economic values in human-altered
ecosystems. Restoration of riparian ecosystems has
been practiced for years with varying success (Palmer
et al. 2009), primarily for natural resource enhancement and to promote so-called ecosystem services
such as bank stabilization and erosion reduction, water
conservation, effluent sequestration, and aquatic and
337
terrestrial wildlife production (Ehrenfeld 2000). Less
often, the primary objective is to enhance populations
of protected species, but this is an important secondary
objective in promoting restoration programs. Achieving these goals is generally less successful than
anticipated, particularly because long-term, self-sustaining biotic assemblages depend on re-creating
complex, inter-linking ecosystems that are both resistant to natural and anthropogenic disturbances, and
resilient so that natural recovery processes such as
primary and secondary succession can proceed with
little intervention (Poff et al. 1997).
Restoration to specifically create habitat for a single
endangered species, particularly in the context of a
weed management program, is a narrow and atypical
goal for ecosystem rehabilitation efforts. In riparian
systems, wildlife species may have very specific
requirements for habitat selection that are hard to
re-create, and these tend to be functional ecosystem
parameters (e.g., vegetation density, patch size, moisture conditions) rather than simply the presence of a
target plant species growing at a site; you may build
it… but they won’t necessarily come. Avian nest site
selection is a particularly complex decision-making
process using indirect environmental cues to indicate
that future conditions will lead to breeding success
(sufficient food resources, low incidence of predators,
etc.), and often prior knowledge (site fidelity) that the
site has been successful before (Stamps and Swaisgood 2007). The general habitat associations of the
willow flycatcher are known (dense, multi-level
vegetation adjacent to open water or saturated soils),
but the precise cues it uses for nest site choice are
poorly understood (Sogge et al. 2003). Furthermore,
the type of vegetation it prefers is transitory, with
habitat suitability declining as vegetative succession
matures to larger trees and a less dense understory.
Thus, riparian restoration should be done on a
landscape scale, mimicking the natural successional
processes and promoting conditions in which there is a
shifting mosaic of vegetation patches with areas of
early successional stands consistently present (cf.
Cardinal and Paxton 2005). It is generally insufficient
to just get some green plants growing.
In the case of the willow flycatcher in tamariskinfested riparian areas, the basic goal is to ensure
sufficient habitat is available which meets nesting
requirements so that biocontrol-induced decline in
Tamarix suitability can be mitigated by presence of
123
338
nearby vegetation as a refuge or alternative habitat
element. The question is how to provide those requirements. Assuming that key environmental features (stand
extent, adjacent standing water or saturated substrate,
arthropod availability, etc.) are suitable, it should be
relatively easy to establish fast-growing natives such
as willows to supplement the suppressed tamarisk.
However, willow flycatcher specialists acknowledge
that such efforts to-date have not been successful
in facilitating occupation (E. Paxton, pers. comm.).
Large-scale native tree horticulture near the lower
Colorado River by the US Bureau of Reclamation has
attracted some sensitive native birds as the vegetation
matures, particularly yellow-billed cuckoo (T. Olson,
unpub. data) but willow flycatchers have not been
similarly attracted to this novel feature. Anderson and
Ohmart (1982) caution, however, that many years are
required before desert riparian restoration success or
failure can be reasonably judged.
On the other hand, there are at least two situations
where unassisted establishment of native plants, albeit
with Tamarix present in the mixed composition stands,
has promoted important increases in willow flycatcher
occupation. At the Salt River and Tonto Creek inflows
to Roosevelt Lake in Arizona, inundation followed by
receding lake level during a drier period in the 1990s
allowed establishment of willows, cottonwoods and
some tamarisk, which led to a four-fold increase in the
number of SWFL territories in the area (Newell et al.
2003). Likewise, at Elephant Butte Reservoir on the
Rio Grande in New Mexico, when a large stand of
tamarisk was inundated and subsequently replaced by
native willows as levels stabilized a consistent increase
in the willow flycatcher population was observed,
from near absence prior to 1995 to over 225 nesting
territories by 2008 (Fig. 5; Ahlers and Moore 2009).
Almost 90% of nests were in native vegetation, 10%
in mixed stands and only a trace were in tamarisk
dominated areas. Both these cases represent numerical
expansion of an extant flycatcher population rather
than new colonization, but they provide good evidence
that avian populations previously using tamarisk have
sufficient resiliency to respond positively to recovery
of native-dominated vegetation.
Scale of restoration
There are three levels of restoration or ecosystem
recovery to be considered in the context of mitigating
123
T. L. Dudley, D. W. Bean
Fig. 5 Data from Ahlers and Moore (2009) showing increase in
abundance of nesting southwestern willow flycatchers at
Elephant Butte Reservoir on the Rio Grande following reestablishment of primarily native riparian vegetation
potential negative effects of Tamarix biocontrol, each
involving a different approach to enhancing reproductive success. The first is to ensure that suitable
structural vegetation elements are available in the
immediate proximity of currently nesting birds, so that
a nest that experiences diminished Tamarix cover will
still have residual protection from direct sun physiological stress provided by other species (successful
nests exhibited C90% canopy cover—McLeod and
Koronkiewicz 2010). The second is to promote native
vegetation in the general area of known nesting,
facilitating the availability of suitable replacement
nesting habitat to sustain local viability of returning
birds. But the third approach is to restore native
vegetation along currently unoccupied (at least by
SWFL) river reaches to attract and support the
expansion of the population, re-connect meta-population structures, and ideally to enable future de-listing
of the species (the goal of endangered species policy).
Protecting currently nesting birds from exposure by
defoliation is probably not a realistic tactic, as it would
involve immediate installation of canopy plants at the
time of, or just prior to Diorhabda colonization and/or
defoliation, potentially disrupting behavior of the
nesting birds. Alternatively, managers could install
artificial shade structures over individual nests if
defoliation is imminent, but that could also attract
predators to the nest site. The use of insecticides or
chemicals that make the tamarisk foliage unappealing
to Diorhabda has been discussed and may bear further
consideration. Such manipulations, if not necessarily
detrimental, are still unlikely to be allowed out of
Tamarisk biocontrol, endangered species risk and resolution
concern for interfering with a listed species. If site
fidelity, the tendency to return to prior nesting
locations, is high then perhaps woody native plants
such as willows could be installed prior to nest
establishment, such as the autumn of the previous
year. While fidelity to a general location can be fairly
high, returning birds are actually fairly mobile and are
probably not likely to use the same tree or patch
(Newell et al. 2003). Vagility is not surprising for a
bird associated with highly dynamic fluvial systems
where sites can change dramatically between years.
The second tactic of promoting stands of native
vegetation in close proximity to known nesting is a
reasonable approach to restoration, taking advantage
of the same site fidelity behavior but with lesser
specificity by occupying alternative plants in the
vicinity if they meet the requirements for nest
placement. This may be an outcome if the first nesting
attempt fails owing to exposure, as the SWFL is
documented to regularly re-nest following nest failure
(e.g., 25% of females in lower Colorado River studies,
McLeod and Koronkiewicz 2010). The areal extent of
currently occupied vegetation gives a clue as to how
large a re-vegetated stand should be to serve this need,
and for regional SWFL sites can be fairly narrow and
less than a hectare in area (Ellis et al. 2009). Because
the dimensional area is not exceptionally large, the
expense of moderately intensive site management
(manual installation of native plants, irrigation, fencing to exclude herbivores, etc.) can still be a costeffective approach to protecting endangered species in
the short-term. Since SWFL prefer sites adjacent to
open water or saturated soils, choosing restoration
sites based on the availability of water close to the soil
surface also means that conditions for growing
phreatophytic plants are probably met and further,
that irrigation may be a temporary need or even
unnecessary. And in some inhabited locations in the
339
Virgin system, nesting stands are comprised of a
mixture of Tamarix and native plants, mostly Salix
spp. and some Populus fremontii, so restoration efforts
would consist of encouraging existing vegetation
rather than requiring large-scale installation of new
plants.
As noted above, such targeted re-vegetation has not
been shown previously to attract SWFL in other
locations. However, the first clear-cut case of restored
site occupation was recently documented in the Virgin
River system where the SWFL-tamarisk biocontrol
controversy is focused. In the Utah portion of the
watershed the Virgin River Program has facilitated
mechanical and chemical tamarisk control in proximity to the City of St. George, simultaneous with the
introduction and establishment of D. carinulata (Bateman et al. 2010). Re-vegetation was implemented by
agencies affiliated with the Virgin River Program to
restore native vegetation, and by 2009–2010 plants
had developed a dense canopy and multi-layer vegetative structure. The willow flycatcher is already
known to nest at several sites in this reach of the river,
mostly in Tamarix despite presence of native willows
in the system. Initially there was little change in nest
site choice, and in 2009 Utah Division of Wildlife
Resources biologists reported 15 nesting attempts by
ten females, almost entirely in tamarisk resulting in
40% nest failures and only two juvenile birds fledged
(Table 1). At least one of the nest trees was defoliated
by Diorhabda, but a causative role of biocontrol could
not be established as the mechanisms for nest failures
were not identified. Similarly high failure rates are
observed elsewhere in the watershed that beetles had
not yet colonized, and predation remains the major
source of mortality (McLeod and Koronkiewicz
2010).
In 2010, however, there was a major behavioral
shift to nesting in the restored native vegetation, with a
Table 1 Southwestern willow flycatcher nesting success at restoration sites on the Virgin River near St. George, Utah, when nesting
territories generally switched from tamarisk vegetation in 2009 to native willow vegetation in 2010
Nesting
season
No. of
females
% of Females
using Tamarix*
Nesting
attempts
Nests fledged
2009
10
90
15
2 (13%; 1 in Salix)
2010
9
22
20
6 (30%; all in Salix)
The percentage of nests fledged is of all nests, including those that were from re-nesting by the same females although re-nesting is
most commonly associated with prior nest failure in the same season. * Proportionate use of Tamarix versus Salix was not precisely
reported for all nesting attempts
123
340
threefold increase in the number of juveniles fledged
(Table 1). This is the best evidence to-date that active
restoration of native vegetation can facilitate utilization of such habitat by SWFL, even where tamarisk is
still present in the system. A note of caution: major
flooding of the Virgin River in December 2010
scoured away a substantial quantity of restored SWFL
habitat (S. Meismer, Virgin River Program, personnel
communication), so it is unclear how sustainable these
efforts will prove to be and illustrates why restoration
in flood-prone watersheds must include eco-hydrological evaluation to enhance the probability of longterm success (Stella et al. 2006b).
A more fundamental test is to verify whether
restoration of native vegetation where the SWFL is no
longer present can induce birds to re-occupy these
abandoned reaches and river systems. These birds will
move fairly large distances, as much as 140 km within
and between watersheds (Paxton et al. 2007). Documented movement was to locations where other
conspecifics form territories, but it is conceivable that
they would occupy new areas within the dispersal
range if conditions were appropriate. Thus the third,
and most robust approach to sustaining and enhancing
southwestern willow flycatcher populations is to
promote watershed-scale restoration of native riparian
assemblages. Restoration on the modest scale of sites
or stream reaches is capital-intensive (Shafroth et al.
2005, 2007), but can still be cost-effective because the
objectives are limited in spatial scope with relatively
high potential for being achieved. On a larger scale the
expense and labor involved in active re-vegetation of
many river miles, often where access is limited, would
be prohibitive. Our interdisciplinary team has proposed an innovative approach to riparian restoration
of the Virgin River ecosystem that could lead to
Fig. 6 Distribution of
P. fremontii adult trees and
seedlings over a 46-km
reach of the Virgin River in
Arizona and Nevada in
summer 2009, indicating
large spatial gaps in
cottonwood distribution,
and localized establishment
in proximity to adult trees;
courtesy M. Taylor
and J. Stella
123
T. L. Dudley, D. W. Bean
large-scale recovery of native vegetation to the benefit
of the SWFL as well as many other riparian-dependent
wildlife species, and at the same time may be quite
cost-effective.
Propagule Islands for watershed restoration
Riparian plants that develop taproots from germinating seeds tend to grow more vigorously than transplanted trees (Bell 1997), but natural recruitment
depends on availability of adult trees that can produce
sufficient seed for extensive dispersal across floodplains, where many if not most sites are not favorable
for germination (Mahoney and Rood 1998). Cottonwood trees that can act as seed sources for recruitment
are nearly absent from the Arizona/Nevada reaches of
the Virgin River (Fig. 6), save for limited patches near
the towns of Littlefield and Beaver Dam, Arizona and
Mesquite, Nevada and arborescent willows (e.g., Salix
gooddingii) are infrequent. Nonetheless, hydrological
conditions in the Virgin floodplain should still favor
the establishment of these taxa (Mortenson and
Weisberg 2010), which are evolved to disperse seeds
in synchrony with high flow events (snowmelt run-off
or direct precipitation, depending on region) that
create the scoured substrates and moisture availability
key to their germination and growth (Mahoney and
Rood 1998; Stella et al. 2006a). Thus, recruitment
seems to be limited by the low abundance of adult
plants capable of dispersing propagules to extensive
reaches, and so recruitment limitation (Lytle and
Merritt 2004) could lead to poor replacement by native
plants even as tamarisk declines in vigor and abundance. Unregulated livestock grazing also has an
important role in excluding the few native trees that do
germinate in the region and inhibiting maturation of
Tamarisk biocontrol, endangered species risk and resolution
surviving seedlings (Taylor and Dudley, unpub. data),
but this impact may be lessened if propagule pressure
were to be enhanced by expanding the number and
distribution seed-bearing adult trees.
A practical approach to promoting extensive seed
dispersal and germination is to create a series of
protected ‘propagule islands’, that is, native plant
patches that could mature and produce sufficient seed
‘rain’ or dispersal kernels (Clark et al. 1999) to
regenerate a native-dominated riparian woodlands
along whole stream reaches under suitable hydrological conditions. This would be particularly valuable,
and cost-effective, in the environmental context of
Tamarix declining in abundance and cover over a very
large landscape as biocontrol by Diorhabda progresses. Our team is currently evaluating the timing
and abundance of seed production and dispersal in
sites where mature P. fremontii (and S. gooddingii to
a lesser extent) still occur, with the objective of
determining the spatial distribution for propagule
island creation that would ensure extensive seed
dissemination and post-flood establishment. An ecohydrological assessment of the floodplain is also being
conducted to determine where soil texture and salinity,
moisture and seasonal depth to groundwater, and low
relative probability of flood scour during high flow
events will allow manually installed plants to thrive
and achieve maturity. The assessment will also
generate predictions of where and to what extent
subsequent natural recruitment of native plants is
expected to occur. This approach is being applied on a
similarly large scale in the San Joaquin River drainage
of California by co-operators in the Virgin River
project (Stella et al. 2006a, b) and is related to the
concept of defining the ‘recruitment box’ for favorable
riparian plant restoration (particularly cottonwoods) in
other western rivers (Rood et al. 2005; Cooper et al.
2003).
Post-biocontrol vegetation composition of native
plants and Tamarix
Under this restoration scenario, an acceptable outcome
would be composite stands of native willows and
cottonwoods interspersed with residual Tamarix. We
would recommend that soil disturbance be minimized
during the course of active re-vegetation, partly to
reduce the likelihood that secondary weeds will take
advantage of disturbance to dominate the sites
341
(D’Antonio and Meyerson 2002; Bay and Sher 2008;
Shafroth et al. 2007; Dudley et al. 2011). In addition,
leaving the standing biomass of declining Tamarix in
place could afford some degree of physical protection
to other plants even as its competitive impact is
relaxed, and continue to provide some structural
wildlife habitat in the interim. Based on experiences
in the upper Virgin watershed, it appears to take
roughly two–three years before re-vegetated sites are
sufficiently grown into be re-occupied by SWFL, and
this renewal could potentially be accelerated if newly
installed plants supplement existing non-native vegetation rather than removing the tamarisk. The goal of
weed biocontrol is to suppress, or reduce the dominance of, invasive plants and not to eradicate them,
thereby mitigating the negative effects of tamarisk
invasion without requiring its total removal. McLeod
and Koronkiewicz (2010) observed that roughly a
quarter to three-quarters of cover over SWFL nest sites
in this watershed is comprised of Tamarix foliage with
the rest made up of native vegetation, so best
management practices for restoring the habitat in the
short term may consist of increasing the proportional
composition of native plants (mostly Salix spp.) in
currently occupied locations. The continued presence
of Tamarix could even enhance the habitat quality of
these sites, as well as propagule island restoration
sites, because its architectural structure seems useful
for encouraging site choice by this and other avian
species. Furthermore, SWFL feed on a wide range of
arthropods in approximate proportion to their occurrence (Wiesenborn and Heydon 2007), so food
resources would also enhanced by availability of
larval and adult Diorhabda (Longland and Dudley
2008).
Over the larger landscape, we similarly see little
necessity for the massive reduction of Tamarix
biomass, unless the dead and dying material poses a
fire hazard or interferes with recovery. In that case
biomass removal, including by prescription fire (Dudley et al. 2011), would be recommended. Along the
Colorado River near Moab, Utah where agencies have
been following Tamarix biocontrol results for five
years, there has been an encouraging increase in
foliar cover of the shrubby willows, e.g., narrowleaf
willow (Salix exigua), as existing plants are released
from competition following Tamarix defoliation
(Bean, unpub. data). Unlike the arboreal willows
(S. gooddingii, S. laevigata, etc.) this clonally
123
342
spreading species, also known as coyote or sandbar
willow, remains common on the Virgin River amongst
the dominant Tamarix vegetation (Dudley et al. 2011)
where it could similarly rebound following Tamarix
suppression (as it did following mechanical control;
Busch and Smith 1995). Shrubby willows comprise a
valuable vegetative component in many western
locations of the SWFL, providing important mid-level
dense foliar cover for shade and refugia from predators
(Sogge et al. 2003).
Furthermore, S. exigua is more tolerant of fire than
are the arboreal taxa (Mount et al. 1996) so it can better
co-exist interspersed with flammable Tamarix vegetation. The risk of wildfire declines as the native
proportion of riparian vegetation increases (Dudley
et al. 2011), and the combination of re-growth of
existing plants, re-vegetation with native trees and
recruitment of native species into zones where
Tamarix biocontrol is effective will eventually make
these systems more resistant to fire. Absent biological
(or mechanical) control treatments, such systems
would otherwise likely proceed toward fire-prone
Tamarix monocultures (Mortenson and Weisberg
2010). That is why planting native trees is currently
a risky management strategy in areas prone to wildfire,
such as along the lower Colorado River, so it is
possible that Tamarix suppression will facilitate more
successful restoration efforts in these regions. Fire has
direct effects on endangered wildlife as well, with
cases of Tamarix-fueled wildfire destroying active
nests of SWFL (Paxton et al. 1996). This occurred
recently within the Virgin watershed, in which two
active SWFL nests were lost at the Moapa Valley/
Warm Springs Natural Area in July 2010 before
juvenile birds had fledged (R. Johnson, personnel
communication). In the absence of biocontrol to
constrain further domination by Tamarix, this will be
an increasingly common occurrence throughout the
Southwest as wildfire is linked with a feedback loop in
which greater tamarisk cover promotes more fire,
leading to greater loss of native vegetation, and so on
(Dudley et al. 2011).
Promoting a substantial portion of native vegetation
not only reduces wildfire risk, but is also key to
sustaining wildlife as avian abundance in mixed
vegetation stands remains relatively unchanged until
approximately 60–75% dominance by Tamarix (van
Riper et al. 2008). Food availability may be part
of this relationship, as native plants support more
123
T. L. Dudley, D. W. Bean
herbivorous arthropods than non-native Tamarix
(Shafroth et al. 2005), so having a substantial component of native vegetation can mean that more food
resources will be available to insectivores than in
monocultural systems (c.f. Herrera and Dudley 2003).
A native overstory also creates a moderated microenvironment, as S. gooddingii canopy maintained
cooler and more humid conditions in the SWFL
nesting zone (2–4 m height) than other vegetation
types (McLeod and Koronkiewicz 2010). In the Virgin
River ecosystem, the diversity and/or abundance of
several wildlife groups, including small mammals,
reptiles, birds and even bats, is lower in Tamarix
monocultures than in mixed species stands, even with
a substantial proportion of Tamarix present (Bateman
et al. 2010; Dudley et al. 2011; Bateman and Ostoja
2012).
In this context the gradual suppression of Tamarix
by biocontrol in association with active and passive
approaches to riparian restoration may lead to incremental yet sustainable enhancement of this habitat
along with associated wildlife species. We know from
this and other locations that the SWFL can respond
positively to expansion of native vegetation under the
appropriate conditions. The regional scope of SWFL
population censusing creates the basis for assessing
large-scale meta-population dynamics of the subspecies (Hatten et al. 2010). Likewise, a restoration
program that links local actions with regional restoration planning within and between watersheds may
allow rehabilitation of this and other spatially
restricted migratory species by reconnecting metapopulation structure and functional movement among
suitable patches throughout the Colorado and other
southwestern basins.
Restoration and resolution of conflicts
between weed biocontrol and endangered species
protection
There is improved potential for achieving ecosystem
recovery and special status species enhancement in the
Virgin River watershed based on a series of initiatives
and partnerships focused on restoration of riparian
resources to meet water and wildlife conservation
goals (Estergard 2008). These are loosely co-ordinated
through a multi-agency task force, the Virgin River
Conservation Partnership and regional Habitat Conservation Plans (Clark County MSHCP, Virgin River
Tamarisk biocontrol, endangered species risk and resolution
HCRP, Colorado River MSCP). A nascent program
will build on these initiatives to carry out an
integrated, watershed-based restoration program targeting SWFL habitat enhancement. The Tamarisk
Coalition, a non-governmental organization focused
on managing tamarisk invasion and promoting riparian restoration, is facilitating this effort that involves
federal natural resource managers (e.g., Fish and
Wildlife Service, Bureau of Reclamation, Geological
Survey—Biological Resources Division, Department
of Agriculture), state resource agencies in Utah,
Arizona and Nevada, University researchers and
private consultants to develop and implement a
restoration program for the Colorado River Basin to
promote native riparian vegetation recovery for SWFL
in the context of current and anticipated Tamarix
biocontrol. With support from the Walton Family
Foundation with its emphasis on Colorado River
conservation issues, the Virgin River was selected as
the first watershed in the Basin for applying habitat
enhancement efforts in support of this mission.
By itself, active restoration will not eliminate
conflicts over strongly held positions, but it can
provide a framework for bringing divergent viewpoints into the same arena. Bridging the conceptual
gap between the ‘opposing sides’ of the biocontrol
issue will be much easier if a co-operative program can
take form that maintains focus on the long-term goal of
restoring functioning river systems, not staking out
positions regarding the perceptions of risks associated
with one of the tools (biocontrol) for achieving that
goal. There has been too much speculation in the
absence of data regarding the benefits and risks of
biocontrol, and the conflict reflects poorly on both the
regulatory agencies that have not fully understood
the process and expectations of weed biocontrol, and
the biocontrol researchers and managers who have not
been effective in presenting the case, as well as the
cautious and rigorous environment in which modern
biocontrol development takes place. Likewise, weed
control practitioners and researchers need to broaden
their perspectives by recognizing that wildlife concerns, even if ultimately shown to be invalid, are
nonetheless real and deserve serious consideration.
We have now come to a point where we are literally
sitting at the same table, and developing the strategy
and tactics for integrating restoration with the ongoing Tamarix biocontrol program. In the course of
executing that process, it is anticipated that the points
343
of disagreement may be made more open to challenge,
and deliberated through the honest communication
that has been lacking in this contentious debate.
The multi-watershed program will involve strategic
planning to ensure that when executed, restoration
will have a high probability of achieving its objectives.
A science-based monitoring protocol is being implemented to assess the process of ecosystem recovery,
from soil and water dynamics to wildlife habitat use
and other key parameters involved in restoring better
function to the system. Information will be used in an
adaptive management framework to increase effectiveness during the course of restoration, and this
approach will be applied coherently across ‘replicate’
watersheds. By incrementally building an objective
database of the direct and indirect responses to
Tamarix biocontrol, participants anticipate that the
resulting data could provide the basis for re-entering
negotiations between US-FWS and USDA-APHIS,
ultimately resolving the regulatory dispute between
the agencies and circumventing the litigation that has
held up this and other weed biocontrol programs
nationally.
One possible outcome would be to bring agencies
tasked with the mandate to protect natural resources into
active participation in the research and development of
biocontrol technologies for reducing impacts of invasive
species. The US Geological Survey (USGS), which is
functionally the research arm of the Department of the
Interior, is directly involved in monitoring of the
ecosystem responses to Tamarix biocontrol in the
Virgin River (Bateman et al. 2010). It may be a logical
next step to engage that agency in the biocontrol
development process itself, as the Bureau of Reclamation has done for many years. Doing so would bring to
biocontrol programs a natural resources perspective that
is outside the central missions of agencies, such as the
US Department of Agriculture, currently charged with
biological control development and implementation.
A recent policy review by the USGS considered new
directions in invasive species science that could include
active participation in biocontrol research, and while
that review has been put on hold, there is value in
renewing the discussion. Given the interest in applying
biological control to wildland environments (Newman
et al. 1998), as indicated by the Biocontrol for Nature
proceedings, ‘conservation biocontrol’ can and should
play a key role in enhancing conservation values and
natural resource benefits.
123
344
Acknowledgments We greatly appreciate the continuing
assistance of a multi-disciplinary research team that is
evaluating tamarisk control and riparian restoration in the
Virgin River watershed and elsewhere, particularly Meghan
Taylor, Matthew Brooks, Steve Ostoja, Curt Deuser, Bill
Longland, Heather Bateman, Mike Kuehn, Pat Shafroth,
Kevin Hultine, Matt Johnson, Ben Conrad, Kumud Acharya,
Bruce Orr, Glen Leverich, Derek Hitchcock, Jack DeLoach,
Ken Lair and many others, and the guidance of the Clark County
Desert Conservation Program (DCP), including John Brekke,
Elizabeth Bickmore and Sue Wainscot. The on-going
biodiversity enhancement program is particularly indebted to
the support and facilitation from the Tamarisk Coalition (Stacy
Kolegas, Shannon Hatch) and the Walton Family Foundation
(Tim Carlson, Margaret Bowman), and benefits from productive
discussion with participating biologists and managers including
Mary Anne McLoed, Nora Caplette, Mark Sogge, Eben Paxton,
Greg Beatty, Jeri Krueger, Theresa Olson, John Willis, Steve
Meismer and many more. Work has been supported, in part, by
grants from Clark County DCP (2005-UCSB-552-P), Forest
Service Forest Health Protection (STDP R4-2004-01) and
USDA-NRI (2006-35302). We dedicate this report to Brian
Cardall, a colleague and friend who believed that tamarisk
control and protection of our natural biodiversity can go handin-hand, and by working together we could actually make the
progress that everyone involved is seeking.
References
Ahlers D, Moore D (2009) A review of vegetation and hydrologic
parameters associated with the southwestern willow flycatcher—2002 to 2008 Elephant Butte Reservoir Delta, NM.
USDI-BOR Technical Service Center, Denver, USA. http://
www.usbr.gov/pmts/fish/Reports/EBR%20SWFL%20veg%
20and%20hydro%20review.pdf
Anderson BW, Ohmart RD (1982) Revegetation for wildlife
enhancement along the lower Colorado River. Final report to
USDI-BOR, Lower Colorado Region, Boulder City, USA
APHIS (2010) USDA APHIS PPQ Moratorium for biological
control of saltcedar. http://www.usbr.gov/uc/albuq/library/
eaba/saltcedar/pdfs/2010/BeetleMemoUSDA.pdf
Bateman HL, Ostoja SM (2012) Invasive woody plants affect the
composition of native lizard and small mammal communities in riparian woodlands. Animal Conserv 15:(in press)
Bateman HL, Paxton EH (2010) Saltcedar and Russian olive
interactions with wildlife. In: Shafroth PB, Brown CA,
Merritt DM (eds) Saltcedar and Russian olive Control
Demonstration Act science assessment. US Geological Survey Scientific Investigations Report 2009-5247, pp 49–63
Bateman HL, Dudley TL, Bean DW, Ostoja SM, Hultine KR,
Kuehn MJ (2010) A river system to watch: documenting
the effects of saltcedar (Tamarix spp.) biocontrol in the
Virgin River Valley. Ecol Restor 28:405–410
Bay RF, Sher AA (2008) Success of active revegetation after
Tamarix removal in riparian ecosystems of the southwestern United States: a quantitative assessment of past
restoration projects. Restor Ecol 16:113–128
Bean DW, Dudley TL, Keller JC (2007) Seasonal timing of
diapause limits the effective range of Diorhabda elongata
123
T. L. Dudley, D. W. Bean
deserticola (Coleoptera: Chrysomelidae) as a biological
control agent for tamarisk (Tamarix spp.). Environ Entomol 36:15–25
Bean DW, Dalin P, Dudley TL, Eberts D, Kazmer K (2012)
Evolution of critical day length for diapause induction
enables range expansion of Diorhabda carinulata, a biological control agent against tamarisk (Tamarix spp). Evol
Appl 5:(in press)
Bell G (1997) Ecology and management of Arundo donax, and
approaches to riparian habitat restoration in Southern California. In: Brock JH, Wade M, Pysek P, Green D (eds) Plant
invasions: studies from North America and Europe. Blackhuys Publishers, Leiden, The Netherlands, pp 103–113
Busch DE, Smith SD (1993) Effects of fire on water and salinity
relations of riparian woody taxa. Oecologia 94:186–194
Busch DE, Smith SD (1995) Mechanisms associated with
decline of woody species in riparian ecosystems of the
southwestern U.S. Ecol Monogr 65:347–370
Cardinal SN, Paxton EH (2005) Home range, movement, and
habitat use of the southwestern willow flycatcher, Roosevelt Lake, AZ—2004. USGS report to the U.S. Bureau of
Reclamation, Phoenix, USA
Center for Biological Diversity (2009) Lawsuit filed to save
endangered songbird; southwestern willow flycatcher
threatened by release of imported beetle. http://www.biological
diversity.org/news/press_releases/2009/southwestern-willowflycatcher-03-27-2009.html
Clark JS, Silman M, Kern R, Macklin E, HilleRisLambers J
(1999) Seed dispersal near and far: patterns across temperate and tropical forests. Ecology 80:1475–1494
Cleverly JR, Smith SD, Sala A, Devitt DA (1997) Invasive
capacity of Tamarix ramosissima in a Mojave Desert
floodplain: the role of drought. Oecologia 111:12–18
Cooper DJ, Andersen DC, Chimner RA (2003) Multiple pathways for woody plant establishment on floodplains at local
to regional scales. J Ecol 91:182–196
D’Antonio C, Dudley T (1997) Saltcedar as an invasive component of the riparian vegetation of Coyote Creek, AnzaBorrego State Park. Final Report, California Department of
Parks and Recreation, San Diego, USA, 56 p
D’Antonio CM, Meyerson L (2002) Exotic species and restoration: synthesis and research needs. Restor Ecol 10:703–713
Dahm CN, Cleverly JR, Allred Coonrod JE, Thibault JR, Mcdonnell DE, Gilroy DJ (2002) Evapotranspiration at the
land/water interface in a semi-arid drainage basin. Freshw
Biol 47:831–843
Dalin P, Bean DW, Dudley TL, Carney VA, Eberts D, Gardner
KT, Hebertson E, Jones EN, Kazmer DJ, Michels GJ,
O’Meara SA, Thompson DC (2010) Seasonal adaptations
to day length in ecotypes of Diorhabda spp. (Coleoptera:
Chrysomelidae) inform selection of agents against saltcedars (Tamarix spp.). Environ Entomol 39:1666–1675
DeLoach CJ, Gerling D, Fornasari L, Sobhian R, Myartseva S,
Mityaev ID, Lu QG, Tracy JL, Wang R, Wang JF, Kirk A,
Pemberton RW, Chikatunov V, Jashenko RV, Johnson JE,
Zeng H, Jiang SL, Liu MT, Liu AP, Cisneros J (1996)
Biological control programme against saltcedar (Tamarix
spp.) in the US: progress and problems. In: Moran VC,
Hoffman JH (eds.) Proceedings of 9th international symposium on biological control of weeds, January 1996,
Stellenbosch, South Africa, pp 253–260
Tamarisk biocontrol, endangered species risk and resolution
DeLoach CJ, Carruthers RI, Dudley TL, Eberts D, Kazmer DJ,
Knutson AE, Bean DW, Knight J, Lewis PA, Milbrath LR,
Tracy JL, Tomic-Carruthers N, Herr JC, Abbott G, Prestwich S, Harruff G, Everitt JH, Thompson DC, Mityaev I,
Jashenko R, Li B, Sobhian R, Kirk A, Robbins TO, Delfosse ES (2004) First results for control of saltcedar
(Tamarix spp.) in the open field in the western United
States. In: Cullen JM, Briese DT, Kriticos DJ, Lonsdale
WM, Morin L, Scott JK (eds) Proceedings of XI international symposium on biological control on weeds, Canberra, Australia, pp 505–513
Dudley TL, DeLoach CJ (2004) Saltcedar (Tamarix spp.),
endangered species, and biological weed control—can they
mix? Weed Technol 18:1542–1551
Dudley TL, DeLoach CJ, Lovich J, Carruthers RI (2000)
Saltcedar invasion of western riparian areas: impacts and
new prospects for control. Transactions of the 65th north
American wildlife and natural resources conference,
March 2000, Chicago, USA, pp 345–381
Dudley TL, DeLoach CJ, Lewis PA, Carruthers RI (2001) Cage
tests and field studies indicate leaf-eating beetle may
control saltcedar. Ecol Restor 19:260–261
Dudley T, Bean D, Dalin P (2009) Failure of biocontrol, success
of bioregulation? Tamarisk Coalition: Tamarisk & Russian
olive research conference, Reno, USA. http://tri-river.mesa
county.us/Tamarisk2009/Dudley/Dudley.html, February
18–19, 2009
Dudley T, Brooks M, Acharya K, Bean D, Conrad B, Deuser C,
Drus G, Kuehn M, Kuczynska I, Hultine K, Matchett JR,
Ostoja SM, Roberts S (2011) Effectiveness monitoring of
springfed wetlands and riparian restoration treatments:
progressive management of invasive tamarisk in the
Southern Nevada region, Final Report. Project 2005USGS-552-P, Clark County, Nevada desert conservation
program. http://www.clarkcountynv.gov/Depts/dcp/Documents/
Library/dcp%20reports/2011/EffectivenessOfWetlandAnd
RiparianRestoration_UCSB-552_Rpt_Jun2011.pdf
Eckberg JR, Foster ME (2011) First account of the splendid
tamarisk weevil, Coniatus splendidulus Fabricius, 1781
(Coleoptera: Curculionidae) in Nevada. Pan-Pacific Entomol 87:51–53
Ehrenfeld JG (2000) Defining the limits for restoration: the need
for realistic goals. Restor Ecol 8:2–9
Ellis LM (1995) Bird use of saltcedar and cottonwood vegetation in the Middle Rio Grande Valley of New Mexico.
J Arid Environ 30:339–349
Ellis LA, Stump SD, Weddle DM (2009) Southwestern willow
flycatcher population and habitat response to reservoir
inundation. J Wildlife Manag 73:946–954
Estergard S (2008) Virgin River watershed—Utah, Arizona and
Nevada comprehensive watershed analysis. US Army Corp
of Engineers, Final Report. http://www.usace.army.mil/
CECW/PlanningCOP/Documents/news/wshed/virgin_river.
pdf
Federal Register (1995) Final rule determining endangered
status for the southwestern willow flycatcher. USDI-FWS,
50 CFR Part 17, RIN 1018 AB97 Federal Register, vol 60,
no 38, pp 10694–10715. http://ecos.fws.gov/docs/federal_
register/fr2790.pdf, February 27, 1995
345
Federal Register (2005) Designation of critical habitat for the
Southwestern Willow Flycatcher (Empidonax traillii extimus). USDI-FWS, 50 CFR Part 17, RIN 1018–AT88.
60886 Federal Register, vol. 70, no. 201. http://www.
gpo.gov/fdsys/pkg/FR-2005-10-19/pdf/05-20144.pdf#page=1,
October 19, 2005
Fornasari L (1998) Biology, ethology, and impact on the host by
Coniatus tamarisci (F.) (Coleoptera : Curculionidae), a
natural enemy of Tamarix spp. (Tamaricaceae) in France.
Biol Control 13:25–40
Friedman JM, Auble GT, Shafroth PB, Scott ML, Merigliano
MF, Freehling MD, Griffin ER (2005) Dominance of nonnative riparian trees in western USA. Biol Invas 7:747–751
Graf WL (1978) Fluvial adjustments to the spread of tamarisk in
the Colorado Plateau region. Geol Soc Am Bull 89:
1491–1501
Hatten JR, Paxton EH, Sogge MK (2010) Modeling the dynamic
habitat and breeding population of southwestern willow
flycatcher. Ecol Model 221:1674–1686
Herrera A, Dudley TL (2003) Invertebrate community reduction
in response to Arundo donax invasion at Sonoma Creek.
Biol Invas 5:167–177
Horton JS (1977) The development and perpetuation of the
permanent tamarisk type in the phreatophyte zone of the
Southwest. In: Johnson RR, Jones DA (tech coord), Symposium on importance, preservation, and management of
riparian habitat, 9 July 1977, Tucson, USA. USDA Forest
Service, GTR RM-43, Fort Collins, pp 123–127
Howe WH, Knopf FL (1991) On the imminent decline of Rio
Grande cottonwoods in central New Mexico. Southwest
Nat 36:218–224
Hudgeons JL, Knutson AE, Heinz KM, DeLoach CJ, Dudley
TL, Pattison RR, Kiniry JR (2007) Defoliation by introduced Diorhabda elongata leaf beetles (Coleoptera:
Chrysomelidae) reduces carbohydrate reserves and
regrowth of Tamarix (Tamaricaceae). Biol Control 43:
213–221
Hultine KR, Belnap J, van Riper C, Ehleringer JR, Dennison PE,
Lee ME, Nagler PL, Snyder KA, Uselman SM, West JB
(2010) Tamarisk biocontrol in the western United States:
ecological and societal implications. Front Ecol Environ
8:467–474
Hunter WC, Anderson BW, Ohmart RD (1988) Use of exotic
saltcedar (Tamarix chinensis) by birds in arid riparian
systems. Condor 90:113–123
Kaufmann W (2005) Program for biological control of saltcedar
(Tamarix spp.) in thirteen states: environmental assessment. USDA-APHIS Western Region, Ft. Collins, USA.
http://www.aphis.usda.gov/plant_health/ea/downloads/sal
teafonsi.pdf
Longland WS, Dudley TL (2008) Effects of a biological control
agent on the use of saltcedar habitat by passerine birds. Gt
Basin Birds 10:21–26
Lytle DA, Merritt DM (2004) Hydrologic regimes and riparian
forests: a structured population model for cottonwood.
Ecology 85:2493–2503
Mahoney JM, Rood SB (1998) Streamflow requirements for
cottonwood seedling recruitment—an integrative model.
Wetlands 18:634–645
123
346
Malakoff D (1999) Plan to import exotic beetle drives some
scientists wild. Science 284:1255
McLeod MA, Koronkiewicz TJ (2010) Southwestern willow
flycatcher surveys, demography, and ecology along the
lower Colorado River and tributaries, 2009. Annual report
to Bureau of Reclamation, Boulder City, SWCA Environ
Conserv, Flagstaff, USA, 165 p
Merritt DM, Poff NL (2010) Shifting dominance of riparian
Populus and Tamarix along gradients of flow alteration in
western North American rivers. Ecol Appl 20:135–152
Mortenson SG, Weisberg PJ (2010) Does river regulation
increase the dominance of invasive woody species in
riparian landscapes? Glob Ecol Biogeogr 19:562–574
Mount J, Krausman W, Finch DM (1996) Riparian habitat
change along the Isleta-Belen reach of the Rio Grande. In:
Shaw DW, Finch DM (eds) Desired future conditions for
southwestern riparian ecosystems: bringing interests and
concerns together. USFS General Technical Report RMGTR-272, pp 58–61
Newell PJ, Paxton EH, Sogge MK (2003) Survivorship and
movements of southwestern willow flycatchers at Roosevelt Lake, AZ. USGS report to US Bureau of Reclamation,
Phoenix, USA
Newman RM, Thompson DC, Richman DB (1998) Conservation strategies for the biological control of weeds. In:
Barbosa P (ed) Conservation biological control. Academic
Press, London, UK, pp 371–396
Palmer MA, Menninger H, Benhardt ES (2009) River restoration, habitat heterogeneity, and biodiversity: a failure of
theory or practice? Freshw Biol 55:205–222
Pattison RR, D’Antonio CM, Dudley TL, Allander KK, Rice B
(2010) Early impacts of biological control on canopy cover
and water use of the invasive saltcedar tree (Tamarix spp.)
in western Nevada, USA. Oecologia 165:605–616
Paxton E, Owen J, Sogge MK (1996) Southwestern willow
flycatcher response to catastrophic habitat loss. USGS
Colorado Plateau Research Station Report, Flagstaff,
USA. http://sbsc.wr.usgs.gov/cprs/research/projects/swwf/
Reports/1996_paxton_et_al.pdf
Paxton EH, Sogge MK, Durst SL, Theimer TC, Hatten JR
(2007) The ecology of the southwestern willow flycatcher
in Central Arizona—a 10-year synthesis report. USGS
Open-File Report 2007-1381
Poff NL, Allan JD, Bain MB, Karr JR, Prestegaard KL, Richter
BD, Sparks RE, Stromberg JC (1997) The natural flow
regime. Bioscience 47:769–784
Rood SB, Samuelson GM, Braatne JH, Gourley CR, Hughes
FMR, Mahoney JM (2005) Managing river flows to restore
floodplain forests. Front Ecol Environ 3:193–201
Sala A, Smith SD, Devitt DA (1996) Water use by Tamarix
ramosissima and associated phreatophytes in a Mojave
Desert floodplain. Ecol Appl 6:888–898
Shafroth PB, Cleverly J, Dudley TL, Stuart J, van Riper C,
Weeks EP (2005) Saltcedar removal, water salvage and
wildlife habitat restoration along rivers in the southwestern
US. Environ Manag 35:231–246
Shafroth PB, Beauchamp VB, Briggs MK, Lair K, Scott ML,
Sher AA (2007) Planning riparian restoration in the context
123
T. L. Dudley, D. W. Bean
of Tamarix control in western North America. Restor Ecol
16:97–112
Smith L, Cristofaro M (2010) Utility of field experiments in land
of origin to measure host plant specificity and potential
efficacy of prospective arthropod biological control agents
of weeds. Biocontrol for Natural Conference, Northampton, UK. http://biocontrolfornature.ucr.edu/pdf/smith,lincolnutility_field_experiments_in_land_of_origin.pdf, October
3–7, 2010
Sogge MK, Kus BE, Sferra SJ, Whitfield MJ (2003) Ecology
and conservation of the willow flycatcher. Studies Avian
Biol 26. Cooper Ornithol Soc, Camarillo, USA, 210 p
Sogge MK, Sferra SJ, Paxton EH (2008) Tamarix as habitat for
birds: implications for riparian restoration in the southwestern United States. Restor Ecol 16:146–154
Stamps J, Swaisgood RR (2007) Someplace like home: experience, habitat selection and conservation biology. Appl
Anim Behav Sci 102:392–409
Stella JC, Battles JJ, Orr BK, McBride JR (2006a) Synchrony of
seed dispersal, hydrology and local climate in a semi-arid
river reach in California. Ecosystems 9:1200–1214
Stella JC, Battles JJ, McBride JR (2006b) Restoring recruitment
processes for riparian cottonwoods and willows: a fieldcalibrated predictive model for the lower San Joaquin Basin.
Stillwater Sciences/UC Berkeley Report to CALFED BayDelta Ecosystem Restoration Program, Sacramento, USA.
http://www.stillwatersci.com/resources/2006riprecruitbro
chure.pdf
Stenquist S (1999) Saltcedar biological control and the Saltcedar
Consortium. Aquat Nuis Species Digest 3:20–23
Stromberg JC, Lite SJ, Marler R, Paradzick C, Shafroth PB,
Shorrock D, White JM, White MS (2007) Altered streamflow regimes and invasive plant species: the Tamarix case.
Glob Ecol Biogeogr 16:381–393
Tracy JL, Robbins TO (2009) Taxonomic revision and biogeography of the Tamarix-feeding Diorhabda elongata
(Brullé, 1832) species group (Coleoptera: Chrysomelidae:
Galerucinae: Galerucini) and analysis of their potential in
biological control of Tamarisk. Zootaxa 2101:1–152
van Riper C III, Paxton KL, O’Brien C, Shafroth PB, McGrath
LJ (2008) Rethinking avian response to Tamarix on the
lower Colorado River: a threshold hypothesis. Restor Ecol
16:155–167
Vincent KR, Friedman JM, Griffin ER (2009) Erosional consequence of saltcedar control. Environ Mang 44:218–227
Whiteman KE (2006) Distribution of salt cedar (Tamarix spp. L)
along an unregulated river in south-western New Mexico.
J Arid Environ 64:364–368
Wiesenborn WD, Heydon SL (2007) Diet of southwestern
willow flycatcher compared among breeding populations
in different habitats. Wilson J Ornithol 119:547–557
Author Biographies
Tom L. Dudley is a riparian ecologist and research associate at
the University of California, Santa Barbara, USA. He directs
the Riparian Invasive Research Lab (RIVRLab; http://rivrlab.
msi.ucsb.edu/) which focuses on the ecology, impacts and
Tamarisk biocontrol, endangered species risk and resolution
management of invasive riparian and aquatic organisms,
including biological control of weeds and non-native animals
in river ecosystems. He has been involved since 1997 in the
development and implementation of the Tamarix Biocontrol
Program in North America. The research group is also involved
in the design and implementation of riparian restoration,
and in evaluating the recovery of biodiversity and ecosystem
processes.
347
Daniel W. Bean is an insect physiologist and molecular
biologist, and is Manager of the Colorado Department of
Agriculture Biological Pest Control Program in Palisade,
Colorado, USA. Its mission is to develop and oversee
biocontrol of agricultural and environmental pests in Colorado
and the western US.
123