Content uploaded by Francesca Marini
Author content
All content in this area was uploaded by Francesca Marini
Content may be subject to copyright.
ORIGINAL PAPER
Coypu (Myocastor coypus) in a Mediterranean remnant
wetland: a pilot study of a yearly cycle with management
implications
Francesca Marini
•
Simone Ceccobelli
•
Corrado Battisti
Received: 26 February 2010 / Accepted: 20 December 2010
Ó Springer Science+Business Media B.V. 2011
Abstract Following an apparent increase of local
population density of coypu (Myocastor coypus)ina
Mediterranean remnant wetland, we developed a pilot
study aimed to evaluate a specific control program.
Inside the study area, we performed three transects
per month from August 2008 to July 2009, grouping
data in bimonthly periods. The water level in the
study area showed a maximum in December–Janu-
ary, significantly decreasing from late spring to
summer and significantly increasing from late sum-
mer to winter. Sampled individuals mainly occurred
in Phragmites reed beds and in rush beds (dominance
of Carex sp., Juncus sp. Bolboschoenus sp.). The
index of mean relative density of coypu individuals
ranged between 1.40 (February–March) and 5.72
(October–November) with an evident increase in late
summer–autumn. During this period, mean density of
runways was higher in reed beds than in rush beds,
with differences tending to significance. In summer,
the network of channels in reed beds, locally used for
fishery farm, may maintain a water level suitable for
the coypu. These results (preference for reed beds and
increase of coypu density in late summer–autumn)
should be considered when coypu populations are
under control program, at least in the Mediterranean
region where there is a scarcity of available data.
Keywords Myocastor coypus Reed beds
Rush beds Density Runways Control
Introduction
The coypu (Myocastor coypus Molina, 1782) is an
aquatic rodent native to South America which was
imported for fur farming into Europe, Asia, Africa
and North America (Carter and Leonard 2002;
Bertolino and Genovesi 2007). The rodent repeatedly
escaped from the farms and/or was released into the
wild, and several populations have become estab-
lished along river banks and in wetlands. In the areas
of introduction, the coypu is considered a pest
because of its negative impact on biological diversity,
ecological relationships, crop and irrigation systems
(Linscombe et al. 1981; Shaffer et al. 1992; Llewel-
lyn and Shaffer 1993; Kaplan et al. 1998; Carter et al.
1999; Cabral et al. 2004; Randall and Foote 2005).
For these reasons, coypu is on the list of the 100
World’s Worst Invasive Alien Species (Invasive
Species Specialist Group 2000; Bertolino 2009).
The presence of many invasive species in some cases
F. Marini (&) C. Battisti
Environmental Service (‘Protected Areas-Regional
Parks’), Via Tiburtina, 691, 00159 Rome, Province
of Rome, Italy
e-mail: f.marini@provincia.roma.it
S. Ceccobelli
Environmental Education Centre, Via del Martin
Pescatore, 66, 00124 Rome, Ostia, Italy
123
Wetlands Ecol Manage
DOI 10.1007/s11273-010-9208-9
implies a disruption in ecological relationships and
sometimes economic consequences, with establish-
ment of alien food chains (Amori and Battisti 2008).
Data on the seasonal density and dynamics of
coypu in the Mediterranean region are lacking,
despite the fact that this species was first introduced
to the region during the first half of the twentieth
Century and has significantly increased in both
numbers and the range during the last 20 years
(Reggiani et al. 1995; Cocchi and Riga 2001).
Following an apparent increase of the local
population density of coypu in a small remnant
wetland of Tyrrhenian Central Italy, the Local
Administration managing this area (Province of
Rome) developed a pilot study (sensu Sutherland
2004) on their population status focused on develop-
ing a control program in this area of conservation
concern. In this sense, data collected from this study
may be useful to develop management procedures in
specific habitats and times.
Therefore, the purpose of this study was to
estimate seasonal density of coypu around a yearly
cycle and use of two main habitat types (reed beds
and rush beds) during the period of increased
population density (late summer). In this last case,
we carried out an indirect approach, thus sampling
the density of runways (i.e. passages made by coypu
individuals in the vegetation alongside the banks of
the remnant wetland channels).
Methods
Study area
The study area is located inside the ‘Palude di Torre
Flavia’ Natural Monument (hereafter named, TFNM),
in Central Italy (41°58
0
N; 12°03
0
E), a protected
wetland on the Tyrrhenian coast (size-area: 40 ha).
This is also designated as a Special Protection,
according to the EU Directive 79/409 (Code
IT6030020). TFNM is the remnant of a larger wetland
that was drained and transformed in the second half of
the twentieth Century to an agricultural and urbanized
landscape. Linear density of channels is 800 m/10 ha.
TFNM shows seminatural patchiness with ponds and
channels, reedbeds (Phragmites australis), flooded
meadows, dune and backdune areas, patches of Carex
hirta, Juncus acutus and Cyperaceae.
TFNM is intensely managed for fish farming in a
network of channels (approximately 2,000 m-long,
see above) developing mainly in a Phragmites reed
bed core area. These channels have been artificially
built in the first half of twentieth Century for fish
farming activity. The water supply comes largely
from rainfall (mesomediterranean xeric region; Blasi
1994), while flow from surrounding areas is scarce.
The water level is variable in space and in time, with
an evident water stress induced by fishery farm
activity in late spring–late summer (Causarano and
Battisti 2009; Battisti et al. 2008).
Inside TFNM, a 10-ha-study area, a representative
sample of wetland, was selected. This area is charac-
terized by a network of channels (28 channel traits for
approximately 2,055 m/10 ha) with the presence of
water and covered by reed bed (Phragmites australis)
(core area) and rush bed (Juncus sp., Carex sp. and
Bolboschoenus maritimus) (buffer zone, surrounding
the reed bed core area). Therefore, this area could be
described as an ideal ‘coypu habitat complex’.
For data collection and analyses, we selected four
EU habitat types:
(i) Reed beds with dominance of Phragmites aus-
tralis (Ph; hereafter ‘reed-beds’) (with rare
occurrences of Calystegia sepium and Sambucus
nigra; approximately the 15.1% of TFNM);
(ii) Rush beds (Sc, hereafter ‘rush beds’), corre-
sponding to patches at Bolboschoenus maritimus,
Carex hirta, Juncus acutus and Cyperaceae
(Juncetalia maritimi 1410 EU Directive habitat
type; approximately the 31.4% of TFNM);
(iii) Mediterranean salt meadows (Sarcocornetea
fruticosi 1420 EU Directive habitat type;
approximately the 5% of TFNM);
(iv) Environment back dunes (embryonic shifting
dunes 2110 EU Directive habitat type and
annual vegetation of drift lines 1210 EU
Directive habitat type; approximately the 5%
of TFNM).
Data collection and analysis
Water level
From August 2008 to July 2009, we measured with a
metric pole (±1 cm) once every 10 days at only one
sampling point (hydrometric station of TFNM;
Wetlands Ecol Manage
123
Battisti 2006) the water level in the wetland channels
obtaining, a mean value of water level for each
bimonthly period (MWL; in cm). We performed three
measures per month (six per bimonthly period).
Channels were artificially built for fish farming in the
first half of the twentieth Century and show a similar
depth. Consequently, water level measured in this
sampling points could be considered representative of
water level in the network of fish farming channels of
the Torre Flavia wetland. We calculated also the
mean of water level in each channel (CWL) in
August by walking along the channels and measuring
the variable every 4 m. This last measure was
correlated to the number of runways (see below) in
summer period to indirectly evaluate the use of two
main wetland habitat types (reed beds and rush beds)
during the period of increase population density.
Relative density of detected individuals
We used direct counts of coypus along a line transect
(see Sutherland 2004) to estimate relative abundance.
Inside the study area, we selected a representative line
transect (416.3 m-length; in Phragmites reed bed:
258.58 m, 62.12%, and in Bolboschoenus rush bed:
157.71 m, 37.88%). We replicated the sessions on the
transect performing three visits/month (approximately
one each 10 days) from August 2008 to July 2009.
The individuals were watched inside a main-belt
50 m-wide along single side of the transect (sampled
area: approximately: 4.16 ha), through a 10 9 50
binocular from one observer (FM), travelling along
1 h before sunset. Sunset was widely considered the
better daily period for sampling the coypu density
(Gosling 1979, S. Bertolino, personal communication).
Data were elaborated obtaining a simple index of
relative density (calculated as number of detected
individuals/100 m). We grouped data in bimonthly
periods obtaining an averaged index of relative density
of coypu individuals.
Density of runways
To calculate the numbers of active coypu runways
(Prigioni et al. 2003 for Mediterranean area) made in
the vegetation alongside the banks of the remnant
wetland channels, we ran through 26 channels (i.e. 52
banks; total channel length: 2006.5 m), checking the
number of runways alongside on both channel banks
and obtaining a measure of density of runways,
calculated as number of runways/100 m of the
wetland channel banks. We have considered active
the runways that showed tracks or signs related to
coypu individuals: in particular, faeces and altered
vegetation (as Phragmites or Juncus stems). We
assumed that locally coypu is the only species that
may track the runways because, in the TFNM, there
are no other large mammals (e.g. dogs, foxes).
To compare the mean density of active runways
(i.e. number of runways/number of channel banks) in
different EU habitat types along the channels banks,
we checked the prevailing habitat type along each
bank in a mean point with a buffer of 15 m outside
(total length of banks: 4012.9 m). We obtained a
representative number of data only for two EU
habitat type: Ph and Sc, although, in Mediterranean
salt meadows and environment back dunes, some
records occurred. As pointed by Prigioni et al. (2003),
we assumed that the runway density was directly
correlated with coypu density.
Statistical analysis
Variables were tested for normality and homosce-
dasticity prior to applying any test. If data distribution
was not normal, then we applied non-parametric
statistical tests. For instance, water level was not
normal, and hence it was always analysed using non-
parametric tests. To compare either the median values
of water level or the coypu relative density, we
performed a Kruskal–Wallis ANOVA along a yearly
cycle, and U-Mann–Whitney test between consecu-
tive bimonthly periods. To correlate the mean water
levels (either MWL or CWL) to both mean relative
density and density of runways, we performed a non-
parametric Spearman rank correlation test (2 tail). We
set alpha level to 0.05, using the SPSS 13.0 software
for Windows (SPSS 2003).
Results
Water level
The MWL in TFNM ranged between 75.50 cm
(±5.28) and 111.83 cm (±5.98) during the period
from August–September to December–January. Dif-
ferences were significant around a yearly cycle
Wetlands Ecol Manage
123
(v
2
= 24.331, P \ 0.001, Kruskal–Wallis test). The
water levels in the channels were at a maximum in
December–January, and steadily decreased during
period from the February–March through April–May
(U = 3.5, z =-2.326, P = 0.020), and significantly
increased between August–September through Octo-
ber–November (U = 3, z =-2.196, P = 0.028),
and October–November through December–January
(U = 3, z =-2.191, P = 0.028; Mann–Whitney
U test) (Fig. 1).
Relative density of detected individuals
Around the yearly cycle, the averaged index of
relative density ranged between 1.40 (±0.77) ind./
100 m (February–March) and 5.72 (±3.22) (Octo-
ber–November). No yearly difference in density
could be detected (Kruskal–Wallis v
2
= 10.137,
df = 5, P = 0.071) (Fig. 2).
No correlation between MWL and mean relative
density could be detected (r
s
= 0.257, P = 0.623;
Spearman rank correlation test, 2 tail).
Density of runways
We sampled 997 runways along 52 channel banks
belonging to four habitat types. Considering only the
more represented habitat types (Ph and Sc, n = 47
banks), the mean density of runways was higher in
Phragmites reed beds (29.53 ± 13.69 runways) when
compared to Bolboschoenus rush beds, but the
differences were not statistically significant (22.86 ±
10.62; Mann–Whitney U test; P = 0.052).
CWL of each channel did not result in significant
correlation with the density of runways in each
channels (r
s
= 0.397, P = 0.103, Spearman rank
correlation test; 2 tail, n = 18).
Discussion
The TFNM remnant wetland represents an area of
conflict where many invasive species occur together
with many species of conservation concern (Battisti
et al. 2008; Amori and Battisti 2008; Battisti et al.
2008).
In this study area, coypu local population showed a
change in relative density index around a yearly cycle
(increase in late summer; decrease in winter) but the
differences were not statistically significant. Coypu
populations have been known to seasonally fluctuate in
many parts of their range in the Northern hemisphere
(Doncaster and Micol 1989) and in the Mediterranean
area; the few available research has highlighted that an
increase of density occurs from spring to autumn
consequent upon immigration from surrounding areas
(e.g. Velatta and Ragni 1991; Reggiani et al. 1995;
Bertolino et al. 2005). In winter, cold climate controls
the density of coypus by increasing reproductive
failure, abortion, mortality and decreasing adult sur-
vival (Doncaster and Micol 1989, 1990).
We collected data on a small subset of a
larger regional coypu population. Consequently, our
0
20
40
60
80
100
120
140
Aug-Sep
Oct-Nov
Dec-Jan
Feb-Mar
Apr-May
Jun-Jul
water level (cm)
Fig. 1 Mean water level (MWL) in channels of the Torre
Flavia Natural Monument (Central Italy) around a yearly cycle
(August 2008–July 2009)
0
1
2
3
4
5
6
7
8
9
10
Aug-Sep
Oct-No v
Dec-Jan
Feb-Mar
Apr-May
Jun-Jul
mean density
Fig. 2 Relative density index of detected individuals (mean
values, ±SD) calculated as number of individuals detected/
100 m
Wetlands Ecol Manage
123
observations may be affected by daily or seasonal
dynamics occurring at a larger scale (in the order of
1–10 km; Kim 1980; Linscombe et al. 1981). It is
likely that coypus move between the TFNM and other
wetlands that are 2–3 km away utilizing a network of
channels in the reclaimed land as corridors. However,
further research would be needed to substantiate this
hypothesis.
In this pilot study, we have shown a higher runway
density in reed beds in August. Assuming that the
density of runways was directly correlated with the
coypu density and activity (Prigioni et al. 2003), we
may hypothesize that coypu in TFNM occurs more
frequently in reed beds during the summer. Reed beds
could be used by coypu as cover, nesting and feeding
sites (personal observations). Moreover, for this
species, reed beds may represent a suitable habitat
in summer. In TFNM, rush beds surrounding the reed
beds tend to dry out in late spring, and water is only
present in fishery farm channels crossing the reed bed
core area (Causarano and Battisti 2009). The excess
in coypu use of reed bed habitat in summer may
imply a disturbance on this habitat type and on its
components (e.g. plants, birds). Indeed, as pointed by
several authors (e.g. Gilbert et al. 2003), the disrup-
tion of reed beds may affect local populations of a set
of specialized area-sensitive, sedentary and/or
migrant bird species of conservation concern. Among
these species, we may cite some Rallidae, Botaurus
stellaris, Ixobrychus minutus, Acrocephalus scirpac-
eus and Acrocephalus arundinaceus, which are
strictly related to reed beds (Benassi et al. 2009).
Nevertheless, differences of density in rush beds
are not significant, and further research is required on
this topic that should be focused on a larger area and
on a wider time range at spatial scale suitable to the
local coypu meta-population.
Management implications
The local Administrators managing these remnant
wetlands should keep in mind that the occurrence of
coypu in reed beds during summer may disrupt the
structure of this habitat type and their suitability for
reed bed related species. The occurrence of coypu in
rush beds should be considered as a priority, because
this habitat type corresponding to Juncetalia maritimi
is of conservation concern sensu Habitat Directive.
To reduce damage to the components of these
remnant wetlands (e.g. reed beds, rush beds) impor-
tant to many specialized bird species and plants a
coypu control program will be needed. This program
should include a monitoring of the population density
and distribution patterns of coypu at multiple scales
(e.g. local and landscape scale; Bertolino et al. 2005;
Curtet 2008). This pilot study has highlighted that, in
Mediterranean remnant wetland (i) coypu increase
their density, during a yearly cycle, in late summer
and autumn; and (ii) in reed beds, a higher frequent-
ation (indirect data by runway density) may occur in
summer with implication of habitat and bird conser-
vation. Therefore, in our study area, a control
program may hypothesize that the captures may be
done in late summer when numbers and water level
are low and coypu were concentrated in reed-beds.
The above data could be useful also to develop
further research on (i) coypu ecology and distribu-
tion/abundance pattern, and (ii) their impact on the
native plant communities and water-related bird
species aimed to define a coypu control in TFNM
that would permit to reduce the damage in remnant
wetlands.
Acknowledgments This study has been performed within the
activities of the Environmental Service—Province of Rome,
that manages ‘Torre Flavia’ Natural Monument. The authors
would like to thank S. Bertolino (DIVAPRA Torino),
P. Genovesi (ISPRA) and A. Monaco (Regional Park
Agency, Latium) for the useful suggestions during the initial
phase of this study; and Luca Luiselli for the careful reading of
English language.
References
Amori G, Battisti C (2008) An invaded wet ecosystem in
central Italy: an arrangement and evidence for an alien
food chain. Rend Fis Acc Lincei 19:161–171
Battisti C (a cura di) (2006) Biodiversita
`
, gestione, conserv-
azione di un’area umida del litorale tirrenico: la Palude di
Torre Flavia. Provincia di Roma, Gangemi editore, Rome
Battisti C, Luiselli L, Pantano D, Teofili C (2008) On threats
analysis approach applied to a Mediterranean remnant
wetland: is the assessment of human-induced threats
related into different level of expertise of respondents?
Biodiv Conserv 16:1529–1542
Benassi G, Battisti C, Luiselli L, Boitani L (2009) Area-sen-
sitivity of three reed bed bird species breeding in Medi-
terranean marshland fragments. Wetl Ecol Manag
17:555–564
Bertolino S (2009) Species account of the 100 of the most
invasive alien species in Europe: Myocastor coypus
Wetlands Ecol Manage
123
(Molina), coypu, nutria (Myocastoridae, Mammalia). In:
DAISIE handbook of alien species in Europe. Invading
nature-Springer series in invasion ecology, vol 3.
Springer, Dordrecht, The Netherlands
Bertolino S, Genovesi P (2007) Semiaquatic mammals intro-
duced into Italy: Case studies in biological invasion. In:
Gherardi F (ed) Biological invaders in inland water.
Profiles, distribution and threats. Springer, Netherlands,
pp 175–191
Bertolino S, Perrone A, Gola L (2005) Effectiveness of coypu
control in small Italian wetland areas. Wildl Soc Bull
33:714–720
Blasi C (1994) Fitoclimatologia del Lazio. Carta del Fitoclima
del Lazio, Univ. La Sapienza, Regione Lazio, Roma
Cabral JA, Miero CL, Marques JC (2004) Environmental and
biological factor influence the relationship between a
predator fish, Gambusia holbrooki, and its main prey in
rice fields of the Lower Mondego River Valley (Portugal).
Hydrobiologia 382:41–51
Carter J, Leonard BP (2002) A review of the literature on the
worldwide distribution, spread of, and efforts to eradicate
the coypu (Myocastor coypus). Wildl Soc Bull 30:162–175
Carter J, Foote AL, Johnson-Randall LA (1999) Modelling the
effects of nutria (Myocastor coypus) on wetland loss.
Wetlands 19:209–219
Causarano F, Battisti C (2009) Effect of seasonal water level
decrease on a sensitive bird assemblage in a Mediterra-
nean wetland. Rend Fis Acc Lincei 20:211–218
Cocchi R, Riga F (2001) Linee guida per il controllo della
Nutria (Myocastor coypus). Quad. Cons. Natura. 5. Min,
Ambiente, Ist. Naz. Fauna Selvatica
Curtet L, Benmergui M, Broyer J (2008) Le dispositif exclos/
te
´
moin, un outil pour e
´
valuer l’efficacite
`
de la regulation
du ragondin. Faune Sauvage 280:16–23
Doncaster CP, Micol T (1989) Annual cycle of a coypu
(Myocastor coypus) population: male and female strate-
gies. J Zool (London) 217:227–240
Doncaster CP, Micol T (1990) Response of coypus to cata-
strophic events of cold and flooding. Holoartic Ecol
(Copenhagen) 13:98–104
Gilbert G, Tyler G, Smith KW (2003) Nestling diet and fish
preference of Bitterns Botaurus stellaris in Britain. Ardea
91:35–44
Invasive Species Specialist Group (2000) 100 of the wolrd’s
worst invasive alien species: a selection from the global
invasive species database—special lift-out in aliens,
Auckland, New Zealand, IUCN/SSC (ISSG) 12:11. IUCN
SSC Invasive Species Specialist Group, IUCN, Gland
Switzerland/Cambridge, UK
Kaplan D, Oron T, Gutman M (1998) Development of mac-
rophytic vegetation in the Agmon Wetland of Israel by
spontaneous colonization and reintroduction. Wetlands
Ecol Manage 6:143–150
Kim P (1980) The coypu (Myocastor coypus) in the Nether-
lands: reproduction, home range and manner of seeking
food. Lutra 23:55–64
Linscombe G, Kinledr N, Wright U (1981). Nutria population
density and vegetative changes in brackish marsh in
coastal Louisiana. In: Chapman JA, Pursley D (eds) Pro-
ceeding of the Worldwide Furbearer Conference 1:
129-141
Llewellyn DW, Shaffer GP (1993) Marsh restoration in the
presence of intense herbivory. The role of Justicia
lanceolata (Chapm.) small. Wetlands 13:176–184
Prigioni C, Balestrieri A, Remonti L (2003) Efficacia degli
interventi di controllo della Nutria (Myocastor coypus)in
Lombardia. Regione Lombardia. Direzione Generale
Agricoltura. Struttura Sviluppo e tutela del territorio ru-
rale e montano. Dip. di Biologia Animale, Universita
`
di,
Pavia
Randall LAJ, Foote AL (2005) Effects of managed impound-
ments and herbivory on wetland plant production and
stand structure. Wetlands 25:38–50
Reggiani G, Boitani L, De Stefano R (1995) Population
dynamics and regulation in the coypu Myocastor coypus
in central Italy. Ecography 18:138–146
Shaffer GP, Sasser CE, Gosselink JG, Rejmanek M (1992)
Vegetation dynamics in the emerging Atchafalaya Delta,
Lousiana, USA. J Ecol 80:677–687
SPSS Inc (2003) SPSS for Windows—Release 13.0 (1 Sep
2004), Leadtools (c), Lead Technologies Inc
Sutherland WJ (2004) Ecological census techniques. Blackwell
Science, Massachussets
Velatta F, Ragni B (1991) The coypu population in Lake
Trasimeno. Population parameters and numerical control.
In: Spagnesi M, Toso T (eds) II Congresso Nazionale
dei Biologi della Selvaggina, Abstract, Suppl Ric Biol
Selvaggina 19: 311–326
Wetlands Ecol Manage
123
