Content uploaded by Marta Lizarralde
Author content
All content in this area was uploaded by Marta Lizarralde on Feb 18, 2016
Content may be subject to copyright.
Chapter
THE INVASIVE BEAVER CASTOR CANADENSIS
IN THE TIERRA DEL FUEGO ARCHIPELAGO: A
MITOCHONDRIAL DNA AND SPATIAL GENETIC
STRUCTURE ANALYSIS FOR CONTROLLING
POPULATION EXPANSION
Mariana Fasanella and Marta Susana Lizarralde
Centro Regional de Estudios Genómicos, UNLP.
Buenos Aires, Argentina
ABSTRACT
The Tierra del Fuego Archipelago (TDFA) contains numerous invasive species, of
which Castor canadensis is the most abundant and important invasion. Beaver are
responsible for the most drastic landscape alteration in Tierra del Fuego since the last
glacial age, affecting not only the hydrology and composition of the southern beech
forest, but more importantly allowing other exotic species to invade the ecosystem. From
25 pairs intentionally released in 1946, beavers have increased their numbers to a current
population size of aprox.100,000 individuals. Here, we present the genetic variability and
population structure of 222 mitochondrial DNA beaver samples. We detected 7 D-loop
haplotypes, 3 of them resulted the most abundant and distributed all along the
archipelago. To analyze the data, TDFA population was subdivided into five
subpopulations and a slight structure was found (ϕst =0.169). With these data we also
identify management units (MUs) of the invasive beaver in the TDFA. Although, sea
water and mountains could be considered barriers for beaver’s dispersion, we found no
barriers to gene flow within the Isla Grande, which may be due to the complex stream
watershed network that allows beavers to easily disperse. These results indicate that the
Isla Grande is a single management unit (MU) and every small island in the archipelago
is a separate MU. It is important to consider that successful eradication programs of
mammals always were conducted in small islands while invasive populations on larger
islands or those that display no distinct structure are more problematic, so the eradication
mlizarralde@fibertel.com.ar.
Mariana Fasanella and Marta Susana Lizarralde
2
of Castor canadensis in the TDFA would be logistically very difficult or perhaps
impossible. Therefore, we propose to (1) control the expansion from IG population to
others areas through trapping beavers and (2) eradicate beaver populations of the adjacent
small islands (small area comparative to Isla Grande) and especially from the continent.
If the species is not early removed in the mainland, beavers will begin to invade the
Patagonian forests very quickly and its future eradication in these continental areas will
be almost impossible. To carry out the control of the species in Tierra del Fuego, is
necessary to promote beaver extraction through the commercial trapping in a way to
control this species in the short term. To achieve this important point, the classic control
scheme must be change it stimulating new business development by incorporating new
materials and products to market and also coordinate the participation of public and
private actions to exploit species under control. In this work, we state three options for
economic use of the beaver: 1) the government could pay a price for killing beavers, 2) to
sell/export the beaver fur and 3) to sell beaver´s meat for human consumption. With these
three options of beaver exploitation, we believe that the Isla Grande control may be
feasible and logistically possible.
INTRODUCTION
Historical Aspects: Introduction and Colonization of
Tierra Del Fuego Archipelago
Species introductions is one of the most controversial topics in Ecology, since Elton’s
(1958) pioneering work, with implication in biological conservation, mostly due to the
biodiversity loss it enthralls (Sala et al. 2000; Vázquez 2002), and the great economic losses
it generates (Pimentel et al. 2000; Curtis and Jensen 2004). Exotic invasive species are
currently the second cause of threats to and extinction of native species, only surpassed by
habitat loss (UICN). Among the many exotic species introduced to new habitats, an estimated
10% have the ability to increase their populations, becoming successful invaders (Kolar and
Lodge 2001). The North American beaver (Castor canadensis) is one of them.
In November 1946, twenty-five mating pairs of C. canadensis (Kuhl 1820) were
introduced from Canada and released into Río Claro, located on the Isla Grande de Tierra del
Fuego (TDF). Such an introduction was made with the purpose of creating a fur industry, and
was backed at the time by Argentina’s Marine Ministry. The absence of most natural
predators, no laws regulating hunting, and the rich abundance of forage and habitats, led to a
rapid increase in the North American beaver populations (Lizarralde 1993). Besides, it is
worth mentioning that island ecosystems (such as Tierra del Fuego Archipelago, TDFA) have
few plants, herbivores, carnivores and decomposers playing active parts in ecological
processes, and therefore are more vulnerable to species invasions (Lever 1994). The
geographical progression of the species was noted on numerous sites throughout the years.
For instance, in 1964, beavers were present on the Chilean side of Lago Fagnano (western
extreme), and by the mid-1960s, beavers had crossed the Beagle Channel, invading Isla
Navarino (Anderson et al. 2011). The colonization of the section of Parque Nacional Tierra
del Fuego located south of Lago Fagnano, probably took place during the 1970s, given the
absence of colonies or impact signs in aerial photographs of the area dating from 1970
(Lizarralde 1989, 1993). In 1979, beavers were found in San Sebastián and later on, by 1986
The Invasive Beaver Castor Canadensis in the Tierra Del Fuego …
3
they were found in the area of China Creek, Puesto Calafate and Río Marazzi (Chile). The
original population progressively invaded the neighboring islands of Navarino, Hoste, Picton,
Nueva and Lennox (Anderson et al. 2009), reaching continental areas by mid-1990s, most
notably on Brunswick Peninsula, in Chile (Wallem et al. 2007). To date, the presence of
beavers has not been reported for the Wollaston islands (Parque Nacional Cabo de Hornos,
Chile), nor for the western sections of the Archipelago (Parque Nacional D´Agostini, Chile)
(Anderson et al. 2006; Moorman et al. 2006). Even though it has been proposed that the
optimum habitat for the species on the Isla Grande is the subantarctic Nothofagus forest
(Lizarralde 1993; Lizarralde et al. 1996, 2004, 2008; Martínez Pastur et al. 2006; Wallem et
al. 2007; Anderson et al. 2009), beavers have been able to colonize open spaces such as the
Fuegian ecotone (located in the center of Isla Grande), and the Magellanic steppe (northern
Isla Grande). This is in agreement with studies performed in the Northern hemisphere, where
the North American beaver showed a preference for pasture-type environments, with no
woody species (Barnes and Malik 1997; Susuki and McComb 1998).
Skewes et al. (1999) estimated the population advance rate of the North American
beaver, considering the year of introduction of beavers into TDF and calculating the lineal
distance from the original source point (Río Claro basin) to those locations where populations
had been reported. For instance, the population advance rate to the rural outpost in Fagnano
was 3.8 km yr-1, while in the northern part of the island, beavers presented an advance rate of
6.3 km yr-1, a much higher rate compared to other environments. Even if these population
advance rates are just approximate, it can be noted that the rate has not been uniform among
different geographic areas, with higher rates in northern (6.3 km yr-1) and southern (5.7 km yr-
1) areas. These geographical variations in advance rate can be explained as a function of the
main niche dimensions of the species, such as hydric (offering refuge and means of transport)
and forest richness (food source). The higher rate in northern TDF, where the conditions for
these two parameters would be most unfavorable, due to a scarcity of forests and its steppe
character, is at first hand contradictory. However the explanation could lie on the relatively
high abundance of water streams, totaling 2,042 km, in contrast to 1,711 km in the central
zone, and 3,630 km in the southwestern forested zone. On the other hand, the rapid advance
in the northern zone could be attributed to the scarcity of food for foraging, and the
difficulties in the construction of beaver lodges imposed by the landscape (e.g. this region
only presents small shrubs); what ultimately may have forced beavers to look for better
environments at a faster displacement rate. This situation would not necessarily take place in
the southern zone where the North American beaver has abundant trees for foraging. The low
population density in the northern zone, even if it presents plenty of water streams, indicates
that the presence of water streams by itself is not enough for the establishment of
numerically-important populations of the North American beaver (Skewes et al. 1999). The
absence of forests is a limiting factor of its advancement (although not absolutely). Briones et
al. (2001) established that the association between terrain slope, marginal vegetation, stream
width and forest community composition in the riparian zone is important for the
establishment of beaver populations. In Sweden, the rate and dispersal pattern of the
European beaver is mainly associated to hydrographic basins (Hartman 1995), while in TDF
the absence of streams is no obstacle to its dispersal (Skewes et al. 1999).
The North American beaver seems to have attained its northern distribution limit in TDF;
with the scarcity of water stream-associated plant biomass being the main limiting factor. The
altitude distribution limit is determined by the availability of vegetation, especially trees; on
Mariana Fasanella and Marta Susana Lizarralde
4
the other hand, the North American beaver reaches down to river mouths, with beaver lodges
and dams found only 50 m from the ocean. Its presence in brackish water has been
corroborated and explains the colonization of other islands within the Archipelago (Ramírez
Silva 2006). It is probable that those individuals that established the expansive south-west
route, gave origin to the populations currently present at Isla Dawson. The highest risk of
advancement of the invasive beaver population from this island towards the continent, has
been predicted on the basis of the short distance separating both locations, and the high
population density at Isla Dawson. In support of the latter, sporadic but continuous sightings
of individuals on Lago Parrillar, located in Brunswick Peninsula (continental Chile), have
taken place since 1994. The absence of predators in the zone might induce behavioral changes
(such as an increment in foraging time), which might in turn increase the beavers’ home
range, leading to the exploitation of larger areas of forest by a single colony (Wallem et al.
2007). However, the presence of a great predator such as the puma (Felis concolor) in
continental areas, might prevent its advancement and lower its chances of displacement for
foraging and lodge construction.
Even though the provincial government of Tierra del Fuego started regulating
commercial hunting of beavers in 1983, and diverse efforts were made to control populations,
the species rapidly expanded its range throughout the TDFA nonetheless, currently inhabiting
98% of streams at Isla Grande, with an estimated abundance of 100,000 individuals
(Lizarralde et al. 2008). Regardless of the lack of precise information on recent years
colonization, it can be mentioned that the invasive dynamics of the North American beaver
currently includes the reoccupation of abandoned sites (due to dam overflows), the
colonization of elevated sections of streams on mountain sides (not used during the initial
stages of colonization), and the dispersal towards the steppe.
The invasion of the North American beaver and its impacts throughout the whole
Archipelago, has generated interests on the part of Argentinean and Chilean government
agencies, conservation organizations and scientific institutions, agreeing on the need for
implementation and evaluation of control measures and its possible eradication from southern
South America. In this scenario, designing an effective management plan in order to prevent
the invasion of new sites, especially in the continent, becomes paramount.
Biological Aspects of Castor canadensis
The North American beaver is the second largest rodent, after the capybara
(Hydrochoerus hydrochaeris); it is taxonomically placed in the Castoridae family, within the
order Rodentia. There are currently two living species with allopatric distribution: C.
canadensis in North America and, C. fiber in Europe and Asia. C. canadensis ranges in its
original distribution from Alaska to the Gulf of Mexico (Müller-Schwarze and Sun 2003),
although it has been introduced to other regions such as the Scandinavian Peninsula, Finland,
and Argentina (TDF) (Figure 1). As a consequence of the introduction of C. canadensis, C.
fiber is currently displaced in certain areas of Europe and Asia (Durka et al. 2005).
Beavers are characterized by the construction of dams in association with flowing water
and the formation of “beaver marshes” and “burrows” where they inhabit and store food. Due
to the structural changes that it makes on the ecosystem, it is considered a “keystone species”.
The Invasive Beaver Castor Canadensis in the Tierra Del Fuego …
5
Specimens from TDF, range in weight between 14 and 30 kg, with an average of 17.88 ±
7.51 kg. Total length is 120 cm, with a mean length of 101.85 ± 7.48 cm. The most
remarkable characteristic of the species is the flattened, paddle-like tail, which may reach a
length of 23-33 cm and a width of 11-18 cm (Lizarralde and Escobar 1999).
Beavers do not present sexual dimorphism, with internal sexual organs; therefore, the
easiest way of sexing organisms is by palpation (for males), and during lactation (for
females). In the Fuegian Archipelago, the reproductive period starts in June and extends to
September, with a gestation period lasting 90-100 days. Kits (young-of-the-year) show up in
early spring; the mean number of kits in each cohort is 3.37 animals, ranging between 3-7
kits. There are no records of beavers having more than a single litter per year (Lizarralde and
Escobar 1999). Beavers have a social organization based on family groups (termed a “colony”
here) that inhabit a single marsh or a succession of marshes along a stream and use a common
feeding ground (termed “food catch”). The number of animals per colony is about 5 (range 4-
6) (Lizarralde 1993, Lizarralde et al. 1996). Beavers are monogamous and a colony is
comprised of a parental couple, the yearlings younger than 2 years, and the young-of-the-
year. The number of individuals and composition of the colony can vary according to habitat
quality, with some bigger colonies in forested areas and smaller ones in the steppe. Lizarralde
(1993) and Briones et al. (2001) reported a colony density in forested areas ranging between
0.7 - 1 colonies km-1 for Isla Grande, while in Isla Navarino the density was of 1.1 colonies
km-1 (Skewes et al. 2006). The diet is exclusively vegetarian and includes bark, leaves and
twigs of woody species; beavers are generalist herbivores that consume at least 80 woody
species and 149 herbaceous plants in their native range (Collen and Gibson 2001). Plants
serve mainly three functions: 1) a direct source of food, 2) food reserves for the winter, and 3)
construction materials for beaver lodges and dams (Busher 1996). In Tierra del Fuego, the
summer diet of beavers is mainly composed of herbaceous species, while in winter they turn
to woody species (Castillo 2006).
Source: Wikipedia.
Figure 1. Castor canadensis distribution. In dark green: native distribution. In light green: exotic
population.
Mariana Fasanella and Marta Susana Lizarralde
6
The ñire (Nothofagus antarctica), lenga (N. pumilio), guindo (N. betuloides) and shrubs
of the genera Pernettya, Berberis, Chiliotrichum, and Juncus constitute the main plant species
on which beavers forage in Tierra del Fuego. Tree-cutting activities in TDF are more intense
during the fall, when the material is chipped and stored underwater, in the proximity of
beaver lodges, as a food reserve for the winter (i.e., the so-called “food catches”) (Lizarralde
and Escobar 1999).
Beavers’ teeth are characterized by long, sharp and strong incisives which enable them to
chop down trees > 1 m in diameter. Additionally, the lips can be sealed behind the incisives,
allowing them to chip wood underwater (Massoia and Chébez 1993; Müller-Schwarze and
Sun 2003).
Effects of Alteration on the Fuegian Ecosystem
The ability to build dams and expand the marshy area, increases the habitat and food
availability for beavers, offering protection from predators such as wolves (Canis lupus) in
the northern Hemisphere, and potentially foxes (Pseudalopex culpaeus) in the southern
Hemisphere; the invasive beaver has not been found in the diet of foxes (Jaksic et al. 1983),
but scars have been found in specimens from Isla Grande (Andrade 2005). The dams built by
beavers modify the annual discharge regime of a river, they reduce current velocity
contributing to the formation of a series of graded “steps”and they expand the area of flooded
soils and increase sediment and organic matter retention (Naiman et al. 1988). The
introduction of the North American beaver has undoubtedly had several ecosystem effects
(some positive, some negative) in Tierra del Fuego (Lizarralde 1993, Iriarte 2000, APN
2006).
Some of the most important negative effects are: 1) the destruction of the riparian forest,
which produces soil destabilization and marked erosion effects on the rest of the forest; 2) as
a consequence of the opening of new clearings, the light regime gets disrupted with high solar
radiation, open terrain gets exposed to the erosive action of wind and rain, altogether
modifying the microclimate of unexposed surfaces; 3) modifications of habitat structure and
aquatic biota; 4) expansion of marsh areas, due to the changes in the drainage pattern and the
height of aquifers; 5) changes in temperature, nutrients and water fluxes, affecting negatively
native fish species; and 6) accumulation of sediment and organic matter, with the ultimate
modification in nutrient cycling within water basins and adjacent areas (Lizarralde et al.
1996, 2004). Without any controls, these alterations could remain as part of the landscape for
decades or centuries, maybe even indefinitely.
Plant litter resulting from chipping down trees, together with erosion of unforested areas
result in an accumulation of organic and inorganic products that acidify the water and bottom
of beaver dams and its surrounding area (Lizarralde et al. 1996). Moreover, the natural
dynamics of the forest are seriously impacted, given that the recovery rate in newly flooded
areas is very slow. Other negative impacts are evidenced by a loss of trees and modification
of soil properties, either through a direct (trees as a food source, or trees fallen to build dams),
or indirect actions (flooding). Nothofagus forests are the most affected ecosystem, especially
the highly productive forests of N. pumilio, since the lenga is the main food source for the
North American beaver in TDF. In all Patagonia, forests of N. pumilio and N. betuloides can
The Invasive Beaver Castor Canadensis in the Tierra Del Fuego …
7
adapt to large-scale alterations such as fires, wind- or landslide-induced fall of trees, but they
seem to be extremely vulnerable to beaver effects (Silva and Saavedra 2008).
Another common disturbance is observed in roads, bridges, hedges and trenches, which
are regularly weakened, interrupted or blocked by the action of the North American beaver.
The dynamics generated by the use, exhaustion and abandonment of those sites occupied by
beavers is associated with the later colonization by exotic plants. On the other hand, the fauna
associated with these altered environments can be directly or indirectly impacted, as is the
case of the torrent duck (Mergannetta armata) which has seen its range displaced to elevated
portions of rivers and water streams (Massoia and Chébez 1993).
Among the positive impacts, the generation of useful environments for some bird species,
especially within the order Anseriformes, is worth mentioning. These make use of the
exposed parts of beaver lodges for nesting, and they profit from the clearing of parts of the
dense forest which favor the growth of plantings and provide nesting areas for migrating
birds. Wright et al. (2002) found that habitat modifications produced by the North American
beaver, increased the number of herbaceous plant species growing on stream banks by 33%.
Lastly, it was estimated that in little more than half a century, the North American beaver has
modified 25,000 ha of Fuegian forests, or roughly the equivalent forest biomass modified for
agriculture practices (Lizarralde 1993).
Population Genetic Structure
Knowing the genetic structure of a certain species, makes it possible to identify the
evolutionary unit and the Management Unit (MU) for it (Moritz 1994). “Management unit” is
a category that makes reference to a group of individuals among which the degree of
ecological and genetic connectivity is low enough to justify a separate monitoring and
management for each group (subpopulation) separately (Palsbøll et al. 2006). Management
units are useful to recognize demographically-distinct populations which need to be managed
in order to ensure the viability of Evolutionarily Significant Unit (ESU), which are bigger. In
island populations one of the biggest risks of failure of a management program (control -
eradication) is the ability of species to recolonize from adjacent islands or even from the
continent. Interconnected or geographically close groups of islands enabling migrations of
specimens have been termed “Eradication Units” (EU) (Robertson and Gemmell 2004). These
EU can be defined as genetically-isolated units with groups (or clusters) of populations which
have to be eradicated at the same time in order to maximize the success of such strategy in the
long run. Identifying EU is no easy task, given that migration patterns depend upon a number
of biological, geographical and human factors. Therefore, the analysis of the population
structure of a species between island clusters and interpreting this structure in terms of gene
flow, is the adequate work frame for the identification of EU. Determining the inter-island
migrating ability of an invasive species and delimiting EU makes eradication efforts more
viable in the long run, since, in that way, a recolonization from neighboring islands is
prevented.
Invasive species eradication is an important tool in ecological conservation and
restoration. It provides the means to alleviate and/or remove the adverse effects that invasive
exotic species typically generate in their new ecosystems. For an eradication program to be
successful, the EU (i.e. the target population) must be clearly defined. Populations in small
Mariana Fasanella and Marta Susana Lizarralde
8
islands or those that are “isolated”, are intrinsically well defined as EU (e.g. the eradication of
mammals from coastal islands in New Zealand; Towns and Broome 2003), while populations
in bigger islands, or those that do not present a structure, are more problematic when it comes
to define EU and therefore for the eradication of species. Even though large scale eradications
are possible (Taylor et al. 2000; Towns and Broome 2003), they are logistically difficult to
carry out and very cost-expensive.
A successful eradication strategy requires considerable planning beyond the mere
identification of population units that are big enough and have low risk of recolonization. For
example, trying to eradicate a fraction of the population or a “sink” population within an
unidentified source-sink population dynamic, would inevitably lead to a rapid recolonization
and therefore an economic loss. It is worth mentioning that just a few migrants per cohort
have the capacity to reduce the genetic differences between populations. Because of this, the
successful eradication of a species requires the elimination of all individuals, or at least
preventing the flux of migrants between populations. In that sense, natural dispersal routes are
very important when defining eradication strategies. Therefore, the objectives of this chapter
are:
1. to quantify levels of intra- and interpopulation genetic variation of C. canadensis in
TDFA,
2. to characterize the population and genetic structure of the invasive species in TDFA,
and
3. to propose potential management or eradication strategies for TDFA.
MATERIALS AND METHODS
Study Area and Sampling
The study area comprised the whole Tierra del Fuego Archipelago (TDFA). Fresh-or
alcohol-fixed tissue samples from 255 beavers, were collected by hunters or donated by
different Argentinean and Chilean organizations (see Acknowledgements section). Samples
were georeferenced and deposited in the Sample Bank of Laboratorio de Ecología Molecular
del Centro Regional de Estudios Genómicos (CREG), Universidad Nacional de La Plata,
UNLP (Argentina).
DNA Extraction
DNA was extracted from fresh tissue samples (muscle, liver, spleen), and dry tail
samples. For most tissues, DNA was extracted following protocols by Sambrook et al.
(1989), with minor modifications in incubation times and concentrations, according to the
type of tissue. For those samples of fresher tissue, we followed the protocol by Aljanabi and
Martínez (1997). All DNA samples were deposited in the collection of Laboratorio de
Ecología Molecular, CREG, UNLP and are preserved in eppendorf test tubes at -20°C for
later genetic analyses.
The Invasive Beaver Castor Canadensis in the Tierra Del Fuego …
9
Mitochondrial DNA Amplification and Sequencing
A D-loop fragment (500 bp) was amplified using the universal primers: Thr-L15926 (5´
CAATTCCCCGGTCTTGTAAACC-3´), located in proximity to the proline tRNA gene, and
DL-H16340 (5´CCTGAAGTAGGAACCAGATG-3´), after Vilà et al. (1999). PCR was
performed following the protocol in Fasanella et al. (2010). The amplification was done on a
total volume of 25 µl with the following components: 25-100 ng DNA, 1x Buffer Taq
Polymerase, 1.5 mM Cl2Mg, 200 µM of each dNTP´s, 25 mM of primer and 1.25 U Taq
Polymerase (PB-LTM, INVITROGENTM). PCR was performed on an automated cycler
THERMO HYBAID MBS o.2S and consisted of denaturation at 94°C for 30 sec, annealing at
55-57°C for 30 sec and extension at 72-74°C for 1-5 min, the cycle was repeated 40 times.
For each individual, two PCR of 25 µl each were performed. Double-stranded PCR products
were purified by precipitation and directly sequenced in both directions using the same
primers for amplifications. Sequencing was conducted at MACROGEN Inc. (Korea).
Sequences for both strands were determined and chromatograms were edited and aligned
using the software program BIOEDIT 7.0 (Hall 1999).
Data Analysis
Genetic diversity of populations was studied estimating parameters of haplotype diversity
(h) and nucleotide diversity (π) with the program DNAsp 4.10 (Rozas et al. 2003). Haplotype
diversity is the probability for two DNAmt sequences randomly taken from a population to be
different; while nucleotide diversity is the number of differences in nucleotides between two
pairs of sequences taken randomly. An haplotype network was done following the “median-
joining” method using the program NETWORK 4.6.0.0. (Fluxus Technology Ltd. 1999-
2011). Haplotype networks (defined as plots connected by circles) are more appropriate to
show relationships within a species.
Hierarchical analyses of molecular variance (AMOVA) were performed with the program
GenAIEx 6.0 (Peakall and Smouse 2006) in order to study different levels of population
structure. Such analysis considers genetic distances between haplotypes and their frequencies;
groups of populations have to be defined a priori in order to evaluate if the genetic structure
proposed by the operator is significant. AMOVA decomposes variance in: a) differences in
the composition of haplotypes between individuals from different populations (i.e. variance
within a population); b) differences in the composition of haplotypes of individuals from
different populations (i.e. variance between populations); and c) differences in the
composition of haplotypes between groups of populations (i.e. variance between regions). For
this analysis, the population of C. canadensis (Argentina-Chile) was divided into 5
subpopulations defined in Figure 2.
In order to determine if there is any relationship between genetic distance and geographic
distance, a Spatial Autocorrelation Analysis was performed for those samples from Parque
Nacional Tierra del Fuego (PNTDF). PNTDF is located within the distribution range of
subpopulation B, and is the only zone which has an abundant number of geographically-close
samples (n = 72).
Mariana Fasanella and Marta Susana Lizarralde
10
Figure 2. Location of the 5 subpopulation (A-E) in the Archipelago of Tierra del Fuego.
Correlation diagrams show the correlation index (r) which provide as proxy for genetic
similarity between pairs of individuals whose spatial separation falls within a distance range.
This analysis was performed with the program GenAIEx 6.0 (Peakall and Smouse 2006).
Since there is no way of inferring genetic structure a priori, 9 distance classes were
considered so that each of them would have a similar number of associated samples (i.e. 0-
800, 801-1300, 1301-2000, 2001-3000, 3001-4000, 4001-5000, 5001-6000, 6001-7000, and
7001-10500 m). Significance tests for the “r” statistic were calculated for each distance class,
defining the upper and lower limits for the confidence interval; this statistical procedure is
described in Peakall and Smouse (2006). Positive values of “r” (r > 0) indicate a positive
autocorrelation, meaning that those individuals genetically closer are also closer
geographically. On the other hand, if r < 0, then the spatial autocorrelation is negative and
therefore genetically similar individuals are further appart geographically.
In order to evaluate the relationship between haplotypes of C. canadensis from
populations introduced to TDFA (Argentina-Chile) and northern Hemisphere populations, an
haplotype network was done using Northern Hemisphere samples and D-loop haplotypes
deposited in GenBank (Ducroz et al. 2005), with the software NETWORK 4.6.0.0. (Fluxus
Technology Ltd., 1999-2011). For northern Hemisphere, 15 samples from Alaska and
Minnesotta (USA) (see Acknowledgements section) and three haplotypes of C. canadensis
deposited in GenBank (Ca1, Ca2 and Ca3; Ducroz et al. 2005) were used.
RESULTS
A ~500 bp D-loop fragment was amplified and sequenced successfully in the 255 beaver
samples considered. Haplotype (h) and nucleotide (π) diversity was 0.684 and 0.00444
respectively, indicating a high haplotype diversity and relatively moderate nucleotide
diversity. The higher diversity haplotype (h) at the subpopulation level was found in
subpopulation B (h = 0.765), the only subpopulation which presents the 7 haplotypes. On the
other hand, values for nucleotide diversity oscillated between 0.00391 and 0.00503, which are
very similar values. This index is related with the number of nucleotide differences between
haplotypes (1.87 in average). For this analysis, the Archipelago was divided into five
subpopulations according to the degree of advancement of beaver invasion (Anderson et al.
The Invasive Beaver Castor Canadensis in the Tierra Del Fuego …
11
2009), the type of environment (steppe, ecotone or forest), the density of colonies km-1 and
the type of hydrographic basin (Lizarralde 1993) (Figure 2). The following units or
subpopulations were determined accordingly:
1. Subpop A (n = 67) representative of the “buffer” area surrounding the original
introduction site (Río Claro), this is the founding population;
2. Subpop B (n = 87) representative of the forest and wetlands area with higher
individual density (4.72 - 5.85 colonies km-1; Lizarralde 1993);
3. Subpop C (n = 35) representative of an ecotone with intermediate densities (~ 2
colonies km-1; Lizarralde 1993);
4. Subpop D (n = 45) representative of the steppe with wide rivers of low slope, where
density is lower than in the other areas (0.2 colonies km-1; Lizarralde 1993); and finally
5. Subpop E (n = 21) from Isla Dawson (Chile).
Seven DNAmt haplotypes (Haplotype A – F and Haplotype H, deposited in GenBank)
were identified; they differed in six polymorphic sites in positions 39, 114, 128, 201, 212 and
236 (all transitions, Table 1). Haplotypes B (~ 40%), D (~ 20%) and H (~ 34%) were the most
abundant ones, the rest of the haplotypes representing less than 2.5% (Table 1).
All subpopulations shared haplotypes B, D and H; two subpopulations shared four
haplotypes (B, D, F and H), while only Subpop B (in Parque Nacional Tierra del Fuego)
presented all the haplotypes detected in this study.
Table 1. Polymorphic sites at the control region of mitochondrial DNA for Castor
canadensis. GenBank accession numbers are shown
Nucleotide Position
39
114
128
201
212
236
GenBank
Accession
Number
HAPLOTYPE
A
A
C
C
G
T
G
AY787822
HAPLOTYPE
B
A
AY787823
HAPLOTYPE
C
G
AY787824
HAPLOTYPE
D
T
A
A
AY787835
HAPLOTYPE
E
T
A
AY787826
HAPLOTYPE
F
T
A
C
AY787827
HAPLOTYPE
H
T
A
C
A
EU476079
Mariana Fasanella and Marta Susana Lizarralde
12
The haplotype network (Figure 3) shows that three haplotypes were exclusive to Subpop
B, while the rest of the haplotypes were shared by five subpopulations, except for haplotype F
which was shared by Subpop A and B. Figure 3 also evidences that two hypothetic
haplotypes would be the ancestral haplotypes, since they connect the rest of the haplotypes.
AMOVA evidenced a 6% difference in population structure between subpopulations,
while within each subpopulation diversity was 94% (Figure 4). The difference between the
five subpopulations was ϕst = 0.058 (p = 0.002) indicating that the population is lightly
structured. Pairwise comparisons indicate a smaller gene flow between subpopulations B and
C (ϕst = 0.100; p = 0.001), followed by B and E (ϕst = 0.091; p = 0.007), which are the
subpopulations located further apart and separated by the Strait of Magellan, which might
represent a barrier to dispersal. Summing up, Subpop B presents larger genetic differences
with respect to the rest of subpopulations (Table 2).
Figure 3. Haplotype network of Castor canadensis. Circle diameters are proportional to the number of
individual sequences per haplotypes, colors indicates subpopulations (violet (subpopA), orange
(subpopB), brown (subpopC), green (subpopD) and red (dubpopE)) and numbers represent nucleotide
differences. Small red circles represent hypothetical haplotypes (mv1, mv2).
The Invasive Beaver Castor Canadensis in the Tierra Del Fuego …
13
Figure 4. Results of Analysis of Molecular Variance. Percentages of variation among and within
subpopulations.
Table 2. Pairwise population differentiation values between five subpopulations of
Castor canadensis. p values are shown above diagonal, ϕst values are
shown below diagonal
SubPop
A
SubPop
B
SubPop
C
SubPop
D
SubPop
E
SubPop
A
0.069
0.358
0.291
0.323
SubPop
B
0.032
0.001
0.001
0.007
SubPop
C
0.000
0.100
0.019
0.079
SubPop
D
0.003
0.082
0.053
0.354
SubPop
E
0.002
0.091
0.050
0.000
Spatial Autocorrelation
The Spatial Autocorrelation Analysis between individuals from PNTDF showed a slight
but significant positive relationship between genetic and geographic distances (r = 0.094; p =
0.002; Figure 5). This pattern was also shown (albeit not significantly) for the first two
distance classes (up to 1.3 km), where the correlation coefficient was positive, but non-
significant, indicating that beavers 5 km apart from each other are more genetically related
than those that are located are smaller and greater distances. On the other hand, beavers that
are between 2 and 5 km and > 6 km apart, presented a negative, non-significant correlation
coefficient.
Relationship between Northern Hemisphere and TDFA Haplotypes
Samples coming from Alaska (USA) presented a single haplotype (Haplotype G), while
of a total of five samples from Minnesota (USA), four presented haplotype MN2 and the
other one, haplotype MN3. All haplotypes were deposited in GenBank (Table 3).
The haplotype network (Figure 6) evidenced that the haplotypes found in TDFA are
closely related, with an average nucleotide number between haplotypes of 1.873; while those
haplotypes from the northern Hemisphere present a higher number of changes between
haplotypes (k = 6.733).
Mariana Fasanella and Marta Susana Lizarralde
14
Figure 5. Spatial Autocorrelation Analysis. Spatial genetic structure autocorrelograms for the National
Park. The 95% confidence intervals for the autocorrelation coefficients (r) are also shown (broken line).
Even if none of the haplotypes detected in TDFA were found in the northern Hemisphere
haplotypes, it can be mentioned that haplotypes D, H and B (all haplotypes from TDFA) are
closer to two haplotypes found in the northern Hemisphere (i.e. Haplotypes G and MN2)
(Figure 6). The separation between them is three polymorphic sites (e.g. Hap H vs MN2; Hap
B vs Hap G). Given the short time since the introduction of North American beavers to
TDFA, it is unlikely that new haplotypes have arisen due to mutations, and therefore the
haplotypes found in the Archipelago must be present in the northern Hemisphere due to their
origin.
DISCUSSION
Population Structure
The colonization of new areas generally produces a marked founder effect with genetic
drift, which reduced the genetic variability within the populations and increases the genetic
differentiation between populations (Wang et al. 2008). However, on the beaver population in
TDF, a high intrapopulation variability (94%) was detected together with a low genetic
differentiation (6%) among subpopulations. The same pattern was found by Refoufi and
Esnault (2006).
According to Lizarralde et al. (2008), a maximum of twenty-five hypothetical
mitochondrial lineages could be the founders of the invasive population in TDFA; of those,
seven lineages were identified in this study.
Chances are that the twenty-five hypothetical lineages (1) might not have been present in
the founding population; or (2) if present in the founders, might have disappeared or became
extinct through selection during the processes of colonization, expansion and invasion.
The haplotype network showed that three haplotypes (B, D and H) were the most
abundant ones, and were detected in five subpopulations, while the rest were registered in
lower proportions in one or two subpopulations. Two hypothetical haplotypes were also
detected, which might be considered “ancestral” given that they connect all the haplotypes
and that: a) they might really exist in the population but were not included in the sampling; b)
they might have been originally introduced in TDF but later became extinct; or c) they might
only be present in populations from the northern Hemisphere. The haplotype network was
built including northern Hemisphere haplotypes, and one of the hypothetic haplotypes (mv2)
The Invasive Beaver Castor Canadensis in the Tierra Del Fuego …
15
was directly connected to these, which might indicate their ancestral nature. On the other
hand, the three most abundant haplotypes (B, D and H) which are directly connected to the
ancestral haplotype (mv2) only differ from this one in a base pair. The latter might indicate
that haplotypes closer to those from the northern Hemisphere are the most common in TDFA.
Lastly, the hypothetical haplotype (mv2) would be the most ancestral one, given that it
originates six haplotypes. This hypothetical haplotype might probably be found in Alberta,
Canada where the founding stock came from.
Table 3. Polymorphic sites at the control region of mitochondrial DNA for Castor canadensis from South America (Hap A–F and Hap H)
and North America (CA1, CA2 and CA3 (Ducroz et al. 2005), HAPLOTYPE G, MN2, MN3). GenBank accession numbers are shown
Nucleotide Position
39
114
128
144
201
205
212
223
233
236
268
276
300
353
354
401
416
420
421
422
Acceso
GenBank
HAPLOTYPEA
A
C
C
T
G
T
T
C
G
G
C
G
A
T
C
C
A
C
T
C
AY787822
HAPLOTYPE
B
A
AY787823
HAPLOTYPE
C
G
AY787824
HAPLOTYPE
D
T
A
A
AY787825
HAPLOTYPE
E
T
A
AY787826
HAPLOTYPE
F
T
A
C
AY787827
HAPLOTYPE
G
A
T
A
AY968083
HAPLOTYPE
H
T
A
C
A
EU476079
MN2
A
A
A
G
JN655158
MN3
T
C
A
T
A
A
T
C
T
T
JN655159
CA1
A
C
A
C
T
C
T
AY623644
CA2
A
C
A
A
C
T
C
T
AY623645
CA 3
A
C
A
A
C
T
C
T
AY623646
Mariana Fasanella and Marta Susana Lizarralde
17
Figure 6. Haplotype network of Castor canadensis. In black: Tierra del Fuego haplotypes. In yellow:
North Hemisphere haplotypes. The numbers represent the 20 nucleotide differences.
Assuming an island model, Wright (1978) proposed that values of Fst > 0.25 would
evidence a great differentiation among populations; values between 0.15 and 0.25 would
represent a moderate differentiation, while there would be no differentiation with values of
Fst < 0.05. In practice, values of Fst rarely exceed 0.5; on the contrary, they are much less.
The complex hydrographic network in TDFA would provide means for beavers to move
freely across the region, behaving as a large, panmictic population, which is backed by the
results presented here. A small but significant divergence between TDFA subpopulations was
registered (ϕst = 0.058; p = 0.002).
The ways in which animals make use of the landscape are determined by their habitat
requirements and their social structure; therefore, populations can exhibit a spatial genetic
structure even in the absence of physical barriers (Latch et al. 2008). However, in this work,
those subpopulations that exhibited a smaller gene flow were Subpop B and C (ϕst = 0.100; p
= 0.001) and Subpop B and E (ϕst = 0.091; p = 0.007), which are located further apart
geographically. Subpopulation E is separated from B by the Strait of Magellan, which might
pose a barrier to dispersal (Hoffmann 1985), although the limited gene flow might also be due
to the great geographic distance (> 230 km) between both subpopulations. Subpopulations B
and E are very different in their landscape use and density patterns. Subpopulation B inhabits
a forested zone and presents a higher abundance, while subpopulation E is located on the
steppe and presents the lowest density.
Finally, the Spatial Autocorrelation Analysis provided a finer-scale resolution for the
population genetic structure. According to these results, beavers 5 km apart (in linear
dimensions) are more genetically similar among themselves than those that are separated by
smaller and larger distances. This might indicate that juveniles migrating from a source
colony to start their own colonies, might do so within a 5 km radius.
Mariana Fasanella and Marta Susana Lizarralde
18
Management Strategies for the North American Beaver in TDF
As ecosystem engineers and keystone species, beavers have great impacts on the
ecosystems and biodiversity of TDFA, and on the local economies. The North American
beaver had a surprising population expansion shortly after its introduction, which brought
large-scale environmental changes. Deciding upon its eradication, a control strategy or even if
the species should be tolerated, is no doubt a complex choice.
Firstly, the North American beaver has long become a successfully-established invasive
species in TDFA, partly due to the absence of predators and competitors, favorable
environmental conditions, the inaccessibility of many remote areas where it thrives, and most
importantly the complex hydrographic network that facilitates the invasion. In turn, all these
factors considered together call for any management action to be decided upon multiple
environmental and social aspects. For instance, it is largely unknown what potential effects
the eradication of the North American beaver would have over the structure and functioning
of current TDFA ecosystems. Some topics to be considered firsthand include the effects of
the breakdown of dams and flooding of large forest areas, the redistribution of the sediments
trapped in dams and their downstream transport to the ocean, the ecological restoration of
impacted areas and their recovery times in terms of biogeochemical cycles. Worldwide, only
a few mammal eradication programs have been successful, and almost exclusively they took
place in islands or small populations (Towns and Broome 2003; Clout and Rusell 2006;
Martins et al. 2006; Rodríguez et al. 2006). Therefore, just considering an eradication
strategy for the North American beaver in Isla Grande (with an area of ~ 48,000 km2), might
represent an effort two-orders of magnitude above any successful rodent eradication program
implemented to date (e.g. 110 km2; Towns and Broome 2003), and probably the least
convenient alternative from a logistics point of view. Technical difficulties (e.g. environment,
climate, inaccessibility to invaded sites, a complex hydrographic network) would prevent the
extraction of 100% of individuals. Moreover, financial constraints related to the very high
costs of eradication (estimated in US $30 million; Parkes et al. 2008) and its estimated long
period of implementation (over 20 years; Parkes et al. 2008), would induce managers to
decide against it.
Social aspects should also be considered together with the above-mentioned problems. As
the invasion progressed, beavers paradoxically become “emblem” animals for the tourism
industry and for the inhabitants of TDFA (Parkes et al. 2008), and so the community’s
opinion on management matters should be taken into account. On the other hand, for an
eradication program to be successful, the detection of (natural or artificial) barriers to
dispersion is paramount, so that each discrete population (< 110 km2) could be considered an
Eradication Unit. However, results from this study indicate that it is not possible to clearly
identify EU in Isla Grande, given that the Strait of Magellan would be the only geographic
barrier that would prevent gene flow in the population. The scenario then becomes complex,
and makes it difficult to decide whether eradicating, controlling or even tolerating the species
is the more convenient strategy, and calls for a combination of all alternative measures into a
single strategy.
It has been shown that the TDFA population does not present a marked spatial structure,
while the five subpopulations identified at the molecular and demographic levels, have
smaller size and therefore would be easier to manage. Therefore, using diagnostic tools from
molecular biology techniques, five management units (MU) can be defined, which are
The Invasive Beaver Castor Canadensis in the Tierra Del Fuego …
19
representative of the “typical” environments invaded by beavers. Two types of activities are
suggested:
1. Eradication in basins or pilot water streams selected for each MU, and
2. Control of the expansion to other occupied areas.
The objective would then be to eradicate individuals in these MU, in accordance to the
technical and economic resources available, and to generate the missing scientific information
that would enable an objective analysis of the eradication in TDFA. For each MU, sustained
capture efforts would be put into practice, and they would be considered “Eradication Units”.
In that sense, the “best practices” would be used based on previous studies (Lizarralde and
Escobar 2000) sustaining superior capture efforts than those reported for the management
plan Lizarralde and Escobar (1999). Control strategies would emphasize and stimulate
extraction by poachers and independent hunters, together with local inhabitants especially in
forested areas, promoting a renewed interest in sound poaching practices, with basis in
environmental conservation. These measures would tend to lower the pressure exerted by the
species and its re-invasion of MU. The ultimate control goal would be to effectively poach all
beavers from all colonies. Accordingly, the “best practices”, which indicate poaching efforts
of 30,000 to 38,000 traps per night (Lizarralde and Escobar 1999; Lizarralde 2008), would be
carried out. The extraction could be further promoted with a renewed interest in the
exploitation of beaver fur and meat (Garriz and Lizarralde 2007), therefore generating an
increased participation of public and private entities in the control of the exotic species, and
indirectly contributing to the design of economically-viable management plans. In this way,
an economic income could be instated for a region that lacks any major economic
developments of its natural resources, and that would generate funds that could be destined
for the conservation of the regional biodiversity.
The recent invasion of a continental section of Chile (Brunswick Peninsula) forced
management agencies from both countries to sign a bi-national agreement in 2008 for the
restoration of those ecosystems affected by beavers. Beaver invasion is no doubt a complex
topic, and a molecular biology approach to it would allow to amalgamate different strategies
in order to ensure a successful eradication and control program that works in the short- and
mid-terms, in which the scientific community, private interests and non-government
organizations would participate.
A sustainable management unit that includes the criteria outlined in this chapter, is likely
to alleviate economic losses due to reduced wood production and flooding of foraging
grounds. Moreover it would contribute to the restoration of Nothofagus forests, helping
restore specific diversity in the habitat currently occupied by the North American beaver.
Lastly, an effective control plan would reduce dispersal of numerous exotic plant species that
are associated to the habitats modified by beavers.
ACKNOWLEDGMENTS
We thanks CONICET for financial support (Grants PIP 0123/2008 and iBOL project) and
also Universidad Nacional de La Plata for providing services. Special thanks to Secretaría de
Mariana Fasanella and Marta Susana Lizarralde
20
Ambiente y Recursos Naturales de TDF (Argentina), Parque Nacional TDF (Argentina),
Servicio Agrícola Ganadero (Mr. Nicolás Soto, SAG, Chile), Mr. Derek Corcorán
(Universidad Católica de Chile), and Mr. Claudio Moraga (Wildlife Conservation Society,
WCS, Parque Karukinka). Samples from Northern hemisphere populations were made
available by Dr. Thomas Hanley (US Forest Service, Juneau, Alaska, USA), and Dr. Karla
Pelz-Serrano (School of Natural Resources, University of Arizona, USA).
REFERENCES
Administración de Parques Nacionales (APN). (2006). Programa de Control de “Castor
Americano” Castor canadensis. 20 pp.
Aljanabi, S. M.and Martínez, E. I. (1997). Universal and rapid salt-extraction of high quality
genomic DNA for PCR- based techniques. Nucleic Acids Research. 25:4692-4693.
Anderson, C. B., Griffith, C. R., Rosemond, A. D., Rozzi, R., and Dollenz, O. (2006). The
effects of invasive North American beavers on riparian plant communities in Cape Horn,
Chile - Do exotic beavers engineer differently in sub-Antarctic ecosystems? Biological
Conservation 128(4): 467-474.
Anderson, C. B., Martínez Pastur, G., Lencinas, M. V., Wallem, P. K., Moorman, M. C., and
Rosemond, A. D. (2009). Do introduced North American beavers engineer differently in
southern South America? An overview with implications for restoration. Mammal Review.
39:33–52.
Anderson, C. H., Soto, N., Cabello, J. L., Martínez Pastur, G., Lencinas, M. V., Wallem, P.,
Antúnez, D., and Davis, E. (2011). Building effective alliances between research and
management to mitigate the impacts of an invasive ecosystem engineer: Lessons from the
pioneering example of the study and control of North American beaver in the Fuegian
archipelago. In: R. Francis (Ed.). A handbook of global freshwater invasive species.
Earthscan Press, London, UK. Section 5, Chapter 29, pp 347-359 (ISBN 978-1-849-71228-
6).
Andrade, B. S. (2005). Gestión para la caza y aprovechamiento de fauna dañina en XII
Región. Informe preparado para el Servicio Agrícola y Ganadero de Magallanes, 107 pp.
Barnes, D. M. and Mallik, A. U. (1997). Habitat factors influencing beaver dam establishment
in a northern Ontario watershed. Journal of Wildlife Management. 61:1371-1377.
Briones, M., Schlatter, R., Wolodarsky, A. Andvenegas, C. (2001). Clasificación ambiental
para hábitat de Castor canadensis (Kuhl 1820, Rodentia), de acuerdo a características de
cuencas en un sector de Tierra del Fuego. Anales del Instituto de la Patagonia, 29, 75–93.
Busher, P. E. (1996). Food caching behavior of beavers (Castor canadensis): selection and use
of woody species. American Midland Naturalist. 135: 343-348.
Castillo, N. (2006). Determinación de la composición botánica de la dieta de Castor (Castor
canadensis) en la isla Tierra del Fuego. Tesis título de Ingeniero en Ejecución
Agropecuario. Universidad de Magallanes, Punta Arenas.
Clout, M. N. and Rusell, J. C. (2006). The eradication of mammals from New Zealand
islands. In: Koike, F., Clout, M. N., Kawamichi, M., De Poorter, M., and Iwatsuki, K.
(Eds.). Assesment and Control of Biological Invasion Risks. (Pp. 127-141). Shoukadoh
Book Sellers, Kyoto, Japan and UICN, Gland, Switzerland.
The Invasive Beaver Castor Canadensis in the Tierra Del Fuego …
21
Collen, P. and Gibson, R. J. (2001). The general ecology of beavers (Castor spp.) as related to
their influence on stream ecosystems and riparian habitats, and the subsequent effects on
fish – a review. Review in Fish Biology and Fisheries 10: 439-461.
Curtis, P. D. and Jensen, P. G. (2004). Habitat factors affecting beaver occupancy along
roadsides in New York State. Journal of Wildlife Management. 68(2):278-287.
Ducroz, J. F., Stubbe, M., Saveljev, A. P., Heidecke, D., Samjaa, R., Ulevicius, A., Stubbe,
A., and Durka, W. (2005). Genetic variation and population structure or the Eurasian
Beaver Castor fiber in Eastern Europe and Asia. Journal of Mammalogy. 86(6): 1059-1067.
Durka, W., Babik, W., Ducroz, J. F., Heidecke, D., Rossel, F., Samjaa, R., Saveljev, A. P.,
Stubbe, A., Ulevicius, A., and Stubbe, M. (2005). Mitochondrial phylogeography of the
Eurasian beaver Castor fiber L. Molecular Ecology. 14: 3843-3856.
Elton, C. S. (1958). The Ecology of Invasions by Animals and Plants. Methuen and Co.,
London, UK.
Fasanella, M., Poljak, S. and Lizarralde, M. (2010). Invasive North American Beaver (Castor
canadensis): the distribution of mitochondrial variation across the Archipelago of Tierra del
Fuego. Mastozoología Neotropical. 17(1):43-52.
Fluxus Technology LTD. (1999-2011). NETWORK ver 4.6.0.0. Available from
http://www.fluxus-engineering.com.
Garriz, C. and Lizarralde, M. (2007). Subproductos Cárnicos de Fauna Silvestre: Posibilidad
de aprovechamiento de carne de castor. Publicación Científico Técnica Recursos Naturales
(Formato Digital). SPD Provincia de Tierra del Fuego. 70 pp.
Hall, T. A. (1999). BioEdit: a user-friendly biological sequence alignment editor and analysis
program for Windows 95/98/NT. Nucleic Acids Symposium Series. 41:95-98.
Hartman, G. (1995). Patterns of spread of a reintroduced beaver Castor fiber population in
Sweden. Wildlife Biology. 1:97-103.
Hoffmann, M. (1985). Zur Ansiedlung der Bisamratte,Ondatra zibethica (L.) In: Südamerika
(On the settlement of the muskrat, Ondatra zibethica (L.) in South-America). Journal Pest
of Science. 58: 93-97.
Iriarte, A. (2000). Impacto de la fauna silvestre sobre la flora nativa de Chile. In: Baldini, A.
and Pancel, L. (Eds). Agentes de daño en el bosque nativo. Santiago, Chile. Editorial
Universitaria. 319-350.
Jaksic, F. M., Yánez, J. L. and Rau, J. R. (1983). Trophic relations of the southernmost
population of Dusicyon in Chile. Journal of Mammalogy 64: 693-697.
Kolar, C. S. and Lodge, D. M. (2001). Progress in invasion biology: predicting invaders.
Trends in Ecology and Evolution. 16(4): 199-204.
Latch, E. K., Scognamillo, D. G., Fike, J. A., Chamberlain ,M. B., and Rhodes, O. E. (2008).
Deciphering ecological barriers to North American River Otter (Lontra canadensis) gene
flow in the Louisiana landscape. Journal of Heredity 99(3): 265-274.
Lever, C. (1994). Naturalized Animals: the ecology of successfully introduced species. T. and
A. D. Poyser, London. 354 pp.
Lizarralde, M. (1989). El Castor (Castor canadensis) en Tierra del Fuego. Praxis 1: 34-35.
Lizarralde, M. (1993). Current status of the introduced beaver (Castor canadensis) population
in Tierra del Fuego, Argentina. Ambio 22(6): 351-358.
Lizarralde, M. S., Deferrari, G. A., Alvarez, S. E., and Escobar, J. M. (1996). Effects of
beaver (Castor canadensis) on the nutrient dynamics of the Southern Beech forest of Tierra
del Fuego (Argentina). Ecología Austral. 6: 101-105.
Mariana Fasanella and Marta Susana Lizarralde
22
Lizarralde, M. and Escobar, J. (1999). Plan de Manejo de la especie Castor canadensis. Sub.
De Rec. Naturales, Prov. TDF.
Lizarralde, M. and Escobar, J. M. (2000). Mamíferos exóticos en la Tierra del Fuego. Ciencia
Hoy 10(56): 52-63.
Lizarralde, M., Escobar, J. and Deferrari, G. (2004). Invader species in Argentina: A review
about the beaver (Castor canadensis) population situation on Tierra del Fuego ecosystem.
Interciencia 29(7): 352-356.
Lizarralde, M. S., Bailliet, G., Poljak, S., Fasanella, M., and Giulivi, C. (2008). Assessing
genetic variation and population structure of invasive North American beaver (Castor
canadensis Kuhl, 1820) in Tierra del Fuego (Argentina). Biological Invasions 10: 673-683.
Martínez Pastur, G., Lencinas, M. V., Escobar, J., Quiroga, P., Malmierca, L., and Lizarralde,
M.. (2006). Understory succession in Nothofagus forests affected by Castor canadensis in
Tierra del Fuego (Argentina). Applied Vegetation Science 9: 143-154.
Martins, T. L. F., Brooke, M. L., Hilton, G. M., Farnsworth, S., Gould, J., and Pain, D .J.
(2006). Costing eradications of alien mammals from islands. Animal Conservation. 9:439-
444.
Massoia, E. and Chebez, J. C. (1993). Mamíferos silvestres del archipiélago fueguino.
Editorial L.O.L.A., Buenos Aires.
Moorman, M. C., Anderson, C. B., Gutiérrez, A. G., Charlin, R., and Rozzi, R. (2006).
Watershed conservation and aquatic benthic macroinvertebrate diversity in the Alberto
D’Agostini National Park, Tierra del Fuego, Chile. Anales del Instituto de la Patagonia. 34,
41–58.
Moritz, C. (1994). Defining “Evolutionary significant units” for conservation. Trends in
Ecology and Evolution. 9:373-375.
Müller-Schwarze, D. and Sun, L. (2003). The Beaver. Natural History of a Wetlands
Engineer. Cornell University Press, Ithaca, New York.
Naiman, R. J., Johnston, C. A. and Kelley, C. A. (1988). Alteration of North American
streams by beaver: The structure and dynamics of streams are changing as beaver recolonize
their historic habitat. BioScience 38:753-762.
Palsbøll, P. J., Berubé, M. and Allendorf, F. W. (2006). Identification of management units
using population genetic data. Trends in Ecology and Evolution 13:146-158.
Parkes, J. P., Paulson, J., Donlan, C. J., and Campbell, K. (2008). Control of North American
Beavers in Tierra del Fuego: Feasibility of Eradication and Alternativa Management
Options. Landcare Research Contract Report LC0708.
Peakall, R. and Smouse, P. E. (2006). GENAIEX 6: genetic analysis in Excel. Population
genetic software for teaching and research. Molecular Ecology Notes. 6: 288-295.
Pimentel, D., Lach, L., Zúñiga, R., and Morrison, D. (2000). Environmental and economic
costs of nonindigenous species in the United States. BioScience. 50, 53–65.
Ramírez Silva, M. D. (2006). Cuantificación de la biomasa leñosa removida por Castor
canadensis (Kuhl 1820, Rodentia) en bosques nativos de Tierra del Fuego (XII región de
Magallanes, Chile). Tesis de doctorado, Departamento de Ciencias Forestales, Pontificia
Universidad Católica de Chile, 152 pp.
Refoufi, A. and Esnault, M. A. (2006). Genetic diversity and population structure of Elytrigia
pycnantha (Godr.) (Triticeae) in Mont Saint-Michel bay using microsatellite markers. Plant
Biology. 8:234-242.
Formatted: Spanish (Argentina)
The Invasive Beaver Castor Canadensis in the Tierra Del Fuego …
23
Robertson, B. C. and Gemmell, N. J. (2004). Defining eradication units to control invasive
pests. Journal of Applied Ecology. 41:1042-1048.
Rodríguez, C., Torres, R. and Drummond, H. (2006). Eradicating introduced mammals from a
forested tropical island. Biological Conservation. 130:98-105.
Rozas, J., Sánchez-Del Barrio, J. C., Messeguer, X., and Rozas, R. (2003). DNAsp. DNA
polymorphism analyses by the coalescent and other methods. Bioinformatics. 19:2496-
2497.
Sala, O. E., Chapin, F. S., Armesto, J. J., Berlow, E., Bloomfield, J., Dirzo, R., Huber-
Sannwald, E., Huenneke, L. F., Jackson, R. B., Kinzig, A., Leemans, R., Lodge, D. M.,
Mooney, H. A., Oesterheld, M., Leroy, Poff, N., Sykes, M. T., Walker, B. H., Walker, M.,
and Wall, D. H. (2000). Global biodiversity scenarios for the year 2100. Science. 287: 1770-
1774.
Sambroock, J., Fritsch, E. F. and Maniatis, T. (1989). Molecular cloning, a Laboratory
Manual. 2nd Edition. Cold Spring Harbor Laboratory Press.
Silva, C. A. and Saavedra, B. (2008). Actas del Taller Internacional para el Control de
Castores en la Patagonia. Edición digital. Wildlife Conservation Society - Chile.
Skewes, O., González, F., Rubilar, L., Quezada, M., Olave, R., Vargas, V., and Ávila, A.
(1999). Investigación, aprovechamiento y control de castor (Castor canadensis) en las islas
Tierra del Fuego y Navarino. Informe Final. Servicio Agrícola y Ganadero (SAG) XII
Región, Magallanes y Antártica chilena. 200 pp.
Skewes, O., Gonzalez, F., Olave, R., Ávila, A., Vargas, V., Paulsen, P., König, H. E. (2006).
Abundance and distribution of American beaver, Castor canadensis (Kuhl 1820), in Tierra
del Fuego and Navarino islands, Chile. European Journal of Wildlife Research. 52: 292–
296.
Suzuki, N. and McComb, W. (1998). Habitat classification models for Beaver (Castor
canadensis) in the streams of the central Oregon coast range. Northwest Science. 72:102-
110.
Taylor, R. H., G. W. Kaiser and M. C. Drever. (2000). Eradication of Norway rats for
recovery of seabird habitat on Langara Island, Britsh Columbia. Restoration Ecology.
8:151-160.
Towns, D. R. and Broome, K.,G. (2003). From small Maria to massive Campbell: forty years
of rat eradications from New Zealand islands. New Zealand Journal of Zoology. 30: 377-
398.
Vázquez, D. P. (2002). Multiple effects of introduced mammalian herbivores in a temperate
forest. Biological Invasions. 4: 175-191.
Vilà, C., Amorim, I. R., Leonard, J. A., Posadas, D., Castroviejo, J., Petrucci-Fonseca, F.,
Crandall, K. A., Ellegren, H., and Wayne, R. K. (1999). Mitochondrial DNA
phylogeography and population history of the grey wolf Canis lupus. Molecular Ecology. 8:
2089-2103.
Wallem, P. K., Jones, C. G., Marquet, P. A., and Jaksic, F. M. (2007). Identificación de los
mecanismos subyacentes a la invasión de Castor canadensis (Kuhl 1820, Rodentia) en el
archipiélago de Tierra del Fuego, Chile. Revista Chilena de Historia Natural. 80: 309-325.
Wang, T., Su, Y. and Chen, G. (2008). Population genetic variation and structure of the
invasive weed Mikania micrantha in Southern China: consequences of rapid range
expansion. Journal of Heredity. 99(1):22-33.
Mariana Fasanella and Marta Susana Lizarralde
24
Wright, J. P., Jones, C. G. and Flecker, A. S. (2002). An ecosystem engineer, the beaver,
increases species richness at the landscape scale. Oecologia. 132(1):96-101.
Wright, S. (1978). Evolution and the Genetics of Populations. In: Variability Within and
Among Natural Populations. Vol. 4. University of Chicago Press, Chicago.
