Estimating population parameters for the Critically Endangered Bermuda skink using robust design capture-mark-recapture modelling
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Estimating population parameters for the Critically
Endangered Bermuda skink using robust design
capture–mark–recapture modelling
HELENA TURNER, RICHARD A. GRIFFITHS
M A R K E . O U T E R B R I D G E and G E R A R D O G A R C I A
Abstract Reliably estimating population parameters for survival (Krebs, ; Besbeas et al., ), which are critical
highly secretive or rare animals is challenging. We report determinants of population viability (White & Burnham,
on the status of the two largest remaining populations of ). Conventionally, marking is used to uniquely identify
the Critically Endangered Bermuda skink Plestiodon lon- individuals in successive samples. For lizards, photographic
girostris, using a robust design capture–mark–recapture identification using natural markings (Sacchi et al., ),
analysis. Skinks were tagged with passive integrated trans- tagging with passive integrated transponders (Germano &
ponders on two islands and captured on sampling occa- Williams, ) and individual recognition from DNA sam-
sions per year over years. The models provided precise pling (Moore et al., ) have increased the utility and
estimates of abundance, capture and survival probabilities application of the capture–mark–recapture approach.
and temporary emigration. We estimated skink abundance The Bermuda skink Plestiodon longirostris is the coun-
to be ± SE . on Southampton Island and ± SE . try’s only extant endemic reptile (Bacon et al., ). The
on Castle Island. The populations do not appear to be stable species primarily feeds on terrestrial and leaf litter arthro-
and fluctuated at both sites over the -year period. Although pods such as ants and woodlice (Wingate, ) but also
the populations on these two islands appear viable, the on a variety of other food sources such as prickly pear cactus
Bermuda skink faces population fluctuations and remains Opuntia sp. and fruits (Wingate, ; Davenport et al., ,
threatened by increasing anthropogenic activities, invasive ; Edgar et al., ). Unlike many other reptiles, the
species and habitat loss. We recommend these two popula- skinks are scavengers with a keen sense of smell that attracts
tions for continued monitoring and conservation efforts. them to carrion (Garber, ; Davenport et al., ), and
will make use of seasonally abundant food sources such as
Keywords Bermuda skink, capture–mark–recapture ana-
broken eggs, dead chicks and dropped fish from nesting col-
lysis, conservation, Plestiodon longirostris, population moni-
onies of the native white-tailed tropicbird Phaethon lepturus
toring, reptile, robust design, survival
catesbyi and the endemic Bermuda petrel Pterodroma cahow
Supplementary material for this article is available at (Garber, ; Edgar et al., ).
https://doi.org/./S The skinks were once common throughout the islands of
Bermuda but populations have undergone significant de-
clines since the s when they were reportedly rarely
seen on the mainland because of increased anthropogenic
Introduction disturbance, habitat loss and degradation, and the intro-
duction of invasive flora and fauna (Davenport et al., ;
M any threatened species are cryptic, elusive and chal-
lenging to survey. The development of efficient
survey methods to detect them is vitally important for
Glasspool & Outerbridge, ). The species is now re-
stricted to the edges of rocky coastal habitat and is cate-
conservation management. A wide variety of methods gorized as Critically Endangered on the IUCN Red List
have been used in monitoring studies, including capture– (Conyers & Wingate, ), with , , individuals
mark–recapture techniques to estimate abundance and thought to be left on Bermuda (Edgar et al., ; Turner,
). Recent surveys have confirmed skinks at only
sites within nature reserves and offshore islands (Turner,
HELENA TURNER (Corresponding author, orcid.org/0000-0002-6067-888X) ), and the continued threats are likely to have a
and RICHARD A. GRIFFITHS ( orcid.org/0000-0002-5533-1013) Durrell
Institute of Conservation and Ecology, School of Anthropology and
major impact on the remaining fragmented and isolated
Conservation, University of Kent, Canterbury, UK populations.
E-mail hturnerjsy@hotmail.com We used a robust design model to monitor trends in
MARK E. OUTERBRIDGE Government of Bermuda, Department of Environment abundance, survival, capture probabilities and temporary
and Natural Resources, Flatt’s Village, Bermuda
emigration of the two largest skink populations on
GERARDO GARCIA ( orcid.org/0000-0002-3007-8998) Chester Zoo, Cedar Bermuda. The robust design model of Pollock () is an
House, Chester, UK
extension of the Cormack–Jolly–Seber model and has
Received July . Revision requested September .
Accepted December . become increasingly popular as it combines the advantages
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use,
Downloadeddistribution,
from https://www.cambridge.org/core. IP address:
and reproduction in any medium, provided 46.4.80.155, on 22isDec
the original work 2020cited.
properly at 04:16:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605318001485
Oryx, Page 1 of 8 © 2019 Fauna & Flora International doi:10.1017/S00306053180014852 H. Turner et al.
of the live-recapture model and the closed-capture models. (M. Outerbridge, , pers. comm.) and because reaching
The model and its assumptions are described in detail by the islands in adverse weather was challenging. On both is-
Cormack (), Otis et al. (), Seber (, ) and lands we collected data over a -day period per month during
Kendall et al. (). The current management plan for May–July of , and , resulting in a total of sam-
the Bermuda skink calls for monitoring every – years pling occasions per island and year. Each sampling occasion
(Edgar et al., ) but the last capture–mark–recapture sur- consisted of five trap checks per day, with a total trapping
veys were undertaken in , on Southampton Island, and time of hours per year per island (Supplementary Table ).
the current status of the population was unknown. More fre- We weighed all captured individuals and measured their
quent estimates of population parameters will be necessary snout–vent length, tagged unmarked animals and recorded
to inform conservation management and future monitoring tags of recaptured ones. All individuals in good health
programmes. (excluding gravid females, and juveniles with a snout–vent
length , mm or body mass , . g) were tagged with
passive integrated transponders ( × . mm, mg, model
Study area IDB FDX-B; Trovan, Douglas, Isle of Man) operating at a
frequency of . kHz. Tags were inserted subcutaneously
This study was undertaken at two sites in Castle Harbour,
in either the left or right side of the body with a syringe
Bermuda: the . ha Southampton Island and . ha
implanter and .-inch -gauge sterile disposable hypo-
Castle Island (Fig. ). We chose these two sites because
dermic needle. Prior to tagging, the injection site was
they are thought to harbour the largest known Bermuda
wiped with antiseptic. Immediately after injection, a drop
skink populations, are both within protected nature reserves
of cyanoacrylate glue was applied over the injection site
and are considered the most suitable areas for targeted con-
to prevent tag loss and aid wound healing (Germano &
servation efforts. The islands are separated by a m wide
Williams, ; Gibbons & Andrews, ). Once implanted,
water channel and their abiotic factors are similar. Castle
the tags were checked using a tag reader that revealed the
Island is closer to the main island ( m) and faces more
individual’s unique identification code. Even though tagging
threats from invasive species and anthropogenic activities,
may induce temporary stress (Langkilde & Shine, ), if
which have been the main reasons for the skink’s decline
inserted properly the tags do not otherwise affect the animals
at this location. We surveyed additional sites across Bermuda
(Gibbons & Andrews, ; Connette & Semlitsch, ).
but sample sizes were too small (, individuals) to provide
Once processed, we immediately released individuals at the
reliable estimates of abundance.
capture site.
Methods
Statistical analyses
Skink capture and marking We compiled capture histories as time series of binary va-
lues, with captures coded as and non-captures as , using
We adapted methods from Davenport et al. () and used
a standard x-matrix format (Otis et al., ; Nichols, ).
a consistent survey protocol across the two locations. We
Rows represented capture histories of each captured individ-
used glass jars (c. × mm, volume . l) as traps,
ual and columns represented capture occasions.
set up in × m grid networks (skink home ranges are
We based the capture–mark–recapture modelling on a ro-
c. m; Davenport et al., ). Traps were mounted at a
bust design model (Nichols, ), which assumes population
–° angle, with rocks or vegetation placed at the opening
closure between secondary sampling occasions or days (i.e. no
so skinks could gain access. We used flagging tape to label
births, deaths, emigration or immigration during each -day
each trap, recorded locations with a handheld GPS and
sampling period), but assumes a population open to demo-
used dried palmetto Sabal bermudana fronds and small
graphic changes between primary sampling occasions or
towels to shade the traps. The pitfall traps were baited daily
years (over years, –; Fig. ). The resulting encounter
with ml of canned sardines, which were placed in a small
history consisted of capture occasions with unequal time
sealed tea strainer to prevent consumption, which could af-
spacing; i.e. the days were not always consecutive, nor was
fect recapture rates. We smeared ml of cod liver oil around
sampling undertaken on the same dates each year, because of
the rim of each trap to prevent skinks escaping and to deter
weather conditions and logistical challenges.
ants (Davenport et al., ). Traps were checked hourly dur-
ing .–. and were then closed by removing the bait
and turning the jar upside down to prevent accidental cap- Model selection
tures. We did not carry out surveys at temperatures , °C
or when heavy rainfall or winds (. km/h) were forecast To monitor population trends between the two skink popu-
because the lizards are less active under such conditions lations, we constructed the robust design model to estimate
Oryx, Page 2 of 8 © 2019 Fauna & Flora International doi:10.1017/S0030605318001485
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 22 Dec 2020 at 04:16:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605318001485Bermuda skink population modelling 3
FIG. 1 (a) Bermuda with the
locations of Castle and
Southampton Islands within
Castle Harbour, and locations
of traps for the Bermuda skink
Plestiodon longirostris on
(b) Castle Island and
(c) Southampton Island.
the probability of survival w, probability of capture p, heterogeneity effects Mth (the mean probabilities of cap-
temporary emigration γ and population size N, using the ture for each occasion) and the null model with no time
packages marked, descr and Rcapture in R .. (Rivest & or heterogeneity effects M (the capture probability at
Baillargeon, ; R Core Team, ). We derived estimates any capture occasion; Rivest & Baillargeon, ). The
of the demographic parameters using maximum likelihood Mth and Mh models were additionally fitted with four
estimates of the loglinear parameters with the R function heterogeneity estimators: Chao (Chao, ), Poisson
glm and calculated standard errors by linearization. and Gamma (Rivest & Baillargeon, ) and Darroch
Ten loglinear models were used to account for the time (Darroch et al., ). In addition, we tested temporary
(t) and heterogeneity (h) effects of capture probabilities emigration (between and ; and between
on the two islands. These included the model with hetero- and ) for each island, because the probability of cap-
geneity effects Mh (the mean probability of capture), the turing an individual may vary between capture occasions.
model with time effects Mt (the capture probabilities for The models were compared based on their Akaike infor-
each capture occasion), the model with both time and mation criterion (AIC) values.
Oryx, Page 3 of 8 © 2019 Fauna & Flora International doi:10.1017/S0030605318001485
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 22 Dec 2020 at 04:16:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S00306053180014854 H. Turner et al.
FIG. 2 Robust design example, with
three primary trapping sessions each
consisting of five secondary occasions.
Results and temporary emigration was highest on
Southampton Island but was not apparent on Castle
Over years (–) we captured a total of skinks Island. Between and temporary emigration was
at the study sites. We did not tag individuals caught on higher on Southampton but also occurred on Castle
Castle Island ( were too small and four escaped) and Island (Table ).
individuals from Southampton Island ( were too small,
four gravid and escaped). Overall, individuals were
tagged and we used only these in further analyses. More Discussion
than half of the marked skinks were recaptured over the
years: .% ( of ) on Southampton Island and Population trends over time
.% ( of ) on Castle Island.
Applying a robust design model, the Mth model with This study provided precise estimates of abundance, capture
Chao’s estimator for each period had the lowest AIC and and survival probabilities of a Critically Endangered lizard
hence the best fit to the data for both islands (Table ). at two study sites. Our results demonstrate that passive
According to the model, the estimated abundance was integrated transponder tags provide a reliable method for
N = ± SE . on Southampton Island and N = ± SE long-term marking of lizards, and that a robust design
. on Castle Island. The robust design model provided model is effective for monitoring skink population trends.
precise abundance estimates, which were calculated for Previous population estimates of the Bermuda skink
each location over years (Fig. a). Estimates appeared to were calculated using the simple Lincoln–Petersen method,
fluctuate between years on both islands. When comparing which may be subject to bias (Seber, ). Using a tempo-
mean abundance between Southampton Island (mean = rary marking method (acrylic paint spots) can lead to some
± SE ) and Castle Island (mean = . ± SE .) esti- recaptures going undetected and the population size being
mates followed a normal distribution (P , .). overestimated as a result. In addition, previous studies did
Estimates of annual capture probability derived using the not account for juveniles (because of the trapping method
model Mth were slightly higher on Southampton Island used), and some were undertaken during the breeding sea-
(mean P = . ± SE .) compared to Castle Island (mean son (May–June), with brooding females assumed to be un-
P = . ± SE .). Capture probabilities were lowest in available for capture. Davenport et al. () estimated the
(Southampton Island: mean P = . ± SE .; skink population on Southampton Island to be in ,
Castle Island: mean P = . ± SE .) and highest in and Glasspool & Outerbridge () derived an estimate of
(Southampton Island: mean P = . ± SE .; Castle in . On Castle Island, Hammond () estimated
island: mean P = . ± SE .). The estimates were very individuals in . Population estimates were also cal-
precise (small standard errors) as a result of the relatively culated in at four other sites on Bermuda; individuals
high capture probabilities (Fig. b). on Palm Island in Sandys Parish, on Inner Pear Island and
Annual survival of skinks was higher on Southampton on Charles Island (both in St. George’s Parish), and
Island (w = . ± SE .) than on Castle Island (w = . ± at Spittal Pond, a mainland nature reserve in Smith’s Parish
SE .). Survival was lower at both sites from to (Raine, ; Wingate, ). Although these populations
but increased by .% on Southampton Island and .% showed viable proportions of juveniles and adults with
on Castle Island from to . breeding potential at the time, there are now concerns for
When comparing models, the best fit models for both their long-term survival.
islands included temporary emigration between years, Estimating survival is important in demographic studies.
indicating that a small number of individuals were not For example, if skink recruitment is low, then persistence,
available for capture within the sampling areas. Between repeated breeding and longevity may be key factors for
Oryx, Page 4 of 8 © 2019 Fauna & Flora International doi:10.1017/S0030605318001485
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 22 Dec 2020 at 04:16:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605318001485Bermuda skink population modelling 5
TABLE 1 Model selection criteria for Bermuda skink Plestiodon
longirostris capture–mark–recapture data from Southampton Island
and Castle Island, Bermuda. Table shows skink abundance
estimate ± SE, Akaike information criterion (AIC), difference of
AIC to best-performing model (ΔAIC), degrees of freedom (df)
and deviance for each model.
Model Estimate ± SE AIC ΔAIC df Deviance
Southampton Island
Mth Chao 547.2 ± 63.5 571.53 0.00 32,740 332.39
Mth Poisson 595.5 ± 80.4 574.31 2.78 32,743 336.17
Mth Gamma 724.1 ± 91.6 574.44 2.91 32,745 340.30
Mth Darroch 719.0 ± 86.3 576.44 4.91 32,745 337.30
Mt 482.2 ± 43.5 589.20 17.67 32,746 357.06
Mh Darroch 718.7 ± 86.2 607.07 35.54 32,757 396.93
Mh Poisson 601.1 ± 81.7 609.85 38.32 32,755 395.72
Mh Gamma 722.5 ± 91.3 609.86 38.33 32,757 399.72
Mh Chao 555.2 ± 65.0 612.34 40.81 32,752 392.20
M0 490.9 ± 44.7 622.94 51.41 32,758 414.80
Castle Island
Mth Chao 294.8 ± 31.5 416.95 0.00 32,745 249.36
Mt 274.7 ± 27.5 418.64 1.69 32,747 251.66
Mth Gamma 196.6 ± 41.8 420.45 3.50 32,743 247.36
Mth Darroch 183.6 ± 24.1 421.06 4.11 32,743 247.79
Mth Poisson 175.2 ± 30.7 421.55 4.60 32,743 248.27
M0 292.7 ± 30.6 521.41 104.46 32,759 380.12
Mh Gamma 193.8 ± 37.1 523.69 106.74 32,755 374.41
Mh Chao 294.8 ± 31.5 523.73 106.78 32,757 378.45
Mh Darroch 183.5 ± 24.1 523.99 107.04 32,755 374.71
Mt Poisson 176.9 ± 32.5 524.27 107.32 32,755 374.98
recovery. Although previous surveys state the Bermuda
skink may live for up to years (Davenport et al., ),
the survival estimates from our study suggest shorter life
spans, similar to the – years observed in other closely
related Plestiodon species in the wild (Clark et al., ).
At both sites survival was low between and ,
which resulted in a population decrease, followed by high
survival between and , which led to an increase
in recruitment of the larger size classes. This was most likely FIG. 3 (a) Comparison of Mth Chao robust design model of
the result of El Niño (a cyclic shift in atmospheric patterns; Bermuda skink abundance estimates with standard error
Cai et al., ) between and . Although few major (vertical bars) on Southampton Island and Castle Island,
impacts such as tropical storms and hurricanes affected Bermuda, during –. (b) Comparison of Mth Chao
Bermuda during that time, precipitation was increased robust design model of Bermuda skink capture probability
with standard error (vertical bars) on Southampton Island
and winds were stronger compared to (Bermuda
and Castle Island, Bermuda, during –.
Weather Service, ). These conditions would have been
unfavourable for skinks, with a reduction in seasonal food
abundance and fewer opportunities for foraging, basking,
breeding and incubating eggs. In turn, these factors may plants (asparagus fern Asparagus densiflorus and casuarina
have driven the population fluctuations on the two islands. trees Casuarina equisetifolia) and anthropogenic distur-
Southampton Island continues to harbour the highest bance are largely absent. However, these threats are present
density of skinks on Bermuda, mostly because landing is on Castle Island and may explain why populations there
prohibited and threats such as introduced predators and take longer to recover from decreases (Fig. a). Studying
competitors (rats Rattus sp., kiskadee flycatchers Pitangus population dynamics over a longer period may reveal
sulphuratus, cats Felis catus, anolis lizards Anolis sp. and whether these estimates are cyclic or stable and may
yellow-crowned night herons Nyctanassa violacea), invasive uncover factors limiting population growth.
Oryx, Page 5 of 8 © 2019 Fauna & Flora International doi:10.1017/S0030605318001485
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 22 Dec 2020 at 04:16:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S00306053180014856 H. Turner et al.
TABLE 2 Estimated P. longirostris survival probability and tempo- Overall, the different models in this study produced a
rary emigration, ± SE, on Southampton Island and Castle Island, broad range of abundance estimates, .–. individuals
Bermuda, during –, using an Mth Chao robust design on Castle Island and .–. individuals on Southampton
model.
Island (both excluding juveniles). However, the abundance
Sampling Survival probability Temporary emigration estimates based on Mth are reliable and using Chao’s estima-
period w ± SE γ ± SE tor provided lower bound conservative estimates that pro-
Southampton Island duced better fits than other estimators such as Darroch,
2015–2016 0.372 ± 0.08 39.1 ± 32.2 Poisson and Gamma, which can be highly variable (Rivest
2016–2017 0.794 ± 0.19 36.7 ± 49.8 & Daigle, ), especially in small scale capture–mark–re-
Castle Island capture studies (Chao, ). To estimate abundance with lit-
2015–2016 0.309 ± 0.06 0.0 ± 0.0 tle bias, capture probability must be relatively high (Otis et al.,
2016–2017 0.490 ± 0.19 26.0 ± 11.9
; Burnham & Overton, ). Skinks had a moderate
probability of capture on both Southampton and Castle
Island (.–. and .–., respectively), and therefore
the trapping method is adequate to describe the dynamics
Meeting model assumptions of these populations. In general, the Mth estimator works
well if most individuals are captured many times and when
Lizards may shift territories frequently during the breeding the population size is estimated to be . individuals
season (Ruby, ), or in response to fluctuations in food (Otis et al., ).
availability (Hews, ). Consequently, animals are more In , population sizes were low at both sites. However,
likely to be captured at some locations and times than at this time capture probability was highest. Although a con-
others. This violates the standard assumptions of basic siderable number of skinks were caught during this time
capture–mark–recapture models (Seber, ; Hammond, ( individuals in total), % were recaptured individuals
). However, the interval between the primary capture (compared to % in and % in ), which explains
trapping sessions (c. year) was sufficiently long to ensure the lower abundance estimates.
that gains (births and immigration) and losses (deaths and
emigration) would occur and the sample size collected was
large enough to detect heterogeneity and for the Mth model Conclusion
to fit well.
Temporary emigration may occur when a proportion The Bermuda skink receives the highest legislative pro-
of the population remains unavailable for capture. For ex- tection under the Protected Species Act (), and the
ample, skinks often spend a considerable time in rock cre- Government of Bermuda’s Department of Environment
vices where they are difficult to capture. During the breeding and Natural Resources is committed to undertaking
season many females may be sedentary while guarding their conservation activities that ensure the continued survival
nests (Glasspool & Outerbridge, ) and are therefore un- of this unique species.
available for capture. If temporary emigration is not tested Although we identified two relatively large populations,
for it can lead to negatively biased population estimates, and the fluctuations we observed suggest they remain vulnerable
model assumptions may be violated (Hammond, , and may not be viable in the long term without management
). (e.g. creation of artificial burrows for both seabirds and
For the management of small, threatened populations, the skinks, predator control and habitat restoration), especially
potential ramifications of underestimating (negative bias) or during periods of low abundance or when population
overestimating (positive bias) abundance are clearly impor- growth is slow. We advise the continuation of passive inte-
tant. As the robust design uses two levels of sampling it allows grated transponder tagging as a long-term marking method
for more parameters to be estimated and for finer control and that capture–mark–recapture surveys are undertaken
over the relative precision of each parameter (Kendall & annually, alternating between sites to monitor population
Pollock, ; Kendall et al., ). The only major problem trends across Bermuda. Additionally, we recommend using
associated with the robust design is the large trapping effort robust design models for evaluating population parameters
required (Pollock, ); a minimum of days at each site is when samples are taken over multiple days and years, as
recommended (Otis et al., ; Nichols, ) and intensive these can provide timely insights into population trends
sampling can be costly. We therefore recommend this design and the mechanisms driving them, with important implica-
for future capture–mark–recapture studies aimed at estimat- tions for future conservation and research efforts.
ing reptile demographic parameters, particularly for species Acknowledgements We thank the Government of Bermuda
for which population declines need to be detected before they Department of Environment and Natural Resources for providing
become critical. the necessary permissions and for their continued support;
Oryx, Page 6 of 8 © 2019 Fauna & Flora International doi:10.1017/S0030605318001485
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 22 Dec 2020 at 04:16:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605318001485Bermuda skink population modelling 7
J. Maderios, A. Copeland, M. Meijas, N. Wellman, J. Labisko, D AV E N P O R T , J., H I L L S , J., G L A S S P O O L , A. & W A R D , J. () A Study of
S. Clayton-Green, D. Muldoon, N. Wright, J. Carney, D. D’Afflitto, Populations of the Bermudian Rock Lizard (Skink), Eumeces
M. Alonso, S. Massey, M. Shailer, P. Drew, R. Frith, K. Trott, longirostris, Cope (), on the Islands of Nonsuch and
R. Marirea, J.P. Rouja, P. Rouja and L. Thorne for assistance with Southampton, Bermuda. Government of Bermuda, Department of
fieldwork; R. McCrea and E. Matechou for assistance with statistical Agriculture and Fisheries, Flatts, Bermuda.
analyses; and the reviewers for their critiques. This research was funded D AV E N P O R T , J., H I L L S , J., G L A S S P O O L , A. & W A R D , J. () Threats
by J. Summers Shaw, Chester Zoo, the States of Jersey, the British to the Critically Endangered endemic Bermudian skink Eumeces
Herpetological Society and the Bermuda Zoological Society (Eric longirostris. Oryx, , –.
Clee Fund). This is Contribution #278, Bermuda Biodiversity Project E D G A R , P., K I T S O N , L., G L A S S P O O L , A. & S A R K I S , S. () Recovery
(BBP), Bermuda Aquarium, Museum and Zoo, Department of Plan for the Bermuda Skink Eumeces longirostris. Government of
Environment & Natural Resources. Bermuda, Department of Conservation Services, Flatts, Bermuda.
G A R B E R , S. () Behaviour and ecology of the Endangered endemic
Author contributions Analysis and writing: HT; experimental and Bermuda rock lizard (Eumeces longirostris). Rutgers University,
statistical design: HT, RG; field work: HT, RG, MO, GG; revisions: RG, New Jersey, USA.
GG, MO. G E R M A N O , D. & W I L L I A M S , D. () Field evaluation of using passive
integrated transponder (PIT) tags to permanently mark lizards.
Conflicts of interest None. Herpetological Review, , –.
G I B B O N S , J.W. & A N D R E W S , K.M. () PIT tagging: simple
Ethical standards This research abided by the Oryx guidelines on technology at its best. BioScience, , –.
ethical standards, was approved by Chester Zoo and the University G L A S S P O O L , A. & O U T E R B R I D G E , M. () A Population Re-Survey of
of Kent Research and Ethics Committee and was conducted under the Bermuda Skink, Eumeces longirostris Cope (), on
permits issued by the Government of Bermuda’s Department of Southampton Island, Castle Harbour. Bermuda Zoological Society,
Environment and Natural Resources (DENR 15-06-12-27). Handling Flatts, Bermuda.
and tagging of skinks were undertaken in accordance with the H A M M O N D , P.S. () Estimating the size of naturally marked whale
conditions of the licence. populations using capture–recapture techniques. Reports of the
International Whaling Commission, , –.
H A M M O N D , P.S. () Heterogeneity in the Gulf of Maine?
References Estimating humpback whale population size when capture
probabilities are not equal. Reports of the International Whaling
B AC O N , J.P., G R AY , J.A. & K I T S O N , L. () Status and conservation Commission, , –.
of the reptiles and amphibians of the Bermuda islands. Applied H A M M O N D , M.P. () An Analysis of Bermuda Skink (Eucemes
Herpetology, , –. longirostris) Populations on Castle Island, Bermuda. Bermuda
B E R M U D A W E A T H E R S E R V I C E () Climate Data. Http://www. Aquarium, Museum and Zoo, Flatts, Bermuda.
weather.bm/climate.asp [accessed May ]. H E W S , D.K. () Food resources affect female distribution and male
B E S B E A S , P., F R E E M A N , S.N., M O R G A N , B.J.T. & C AT C H P O L E , E.A. mating opportunities in the iguanian lizard Uta palmeri. Animal
() Integrating mark–recapture recovery and census data to Behaviour, , –.
estimate animal abundance and demographic parameters. K E N D A L L , W.L. & P O L LO C K , K.H. () The robust design in
Biometrics, , –. capture–recapture studies: a review and evaluation by Monte Carlo
B U R N H A M , K.P. & O V E R T O N , W.S. () Robust estimation of simulation. In Wildlife : Populations (eds D.R. McCullough &
population size when capture probabilities vary among animals. R.H. Barrett), pp. –. Elsevier Science Publishers, Barking, UK.
Ecology, , –. K E N D A L L , W.L., P O L LO C K , K.H. & B R OW N I E , C. () A
C A I , W., W A N G , G., S A N T O S O , A., L I N , X. & W U , L. () Definition likelihood-based approach to capture–recapture estimation of
of extreme El Niño and its impact on projected increase in demographic parameters under the robust design. Biometrics, ,
extreme El Niño frequency. Geophysical Research Letters, , –.
–. K R E B S , C.J. () Ecological Methodology. nd edition.
C H AO , A. () Estimating the population size for capture–recapture Addison-Wesley Educational Publishers, Inc., Menlo Park, USA.
data with unequal catchability. Biometrics, , –. L A N G K I L D E , T. & S H I N E , R. () How much stress do researchers
C L A R K , K., H E N D R I C K S , A. & B U R G E , D. () Molecular inflict on their study animals? A case study using a scincid lizard,
identification and analysis of Borrelia burgdorferi sensu lato in Eulamprus heatwolei. Journal of Experimental Biology, ,
lizards in the Southeastern United States. Applied Environmental –.
Microbiology, , –. M O O R E , J., C H A R L E S , A., D A U G H E R T Y , H. & N E L S O N , N.J. ()
C O N N E T T E , G.M. & S E M L I T S C H , R.D. () Successful use of a passive Large male advantage: phenotypic and genetic correlates of
integrated transponder (PIT) system for below-ground detection of territoriality in tuatara. Journal of Herpetology, , –.
plethodontid salamanders. Wildlife Research, , –. N I C H O L S , J.D. () Capture–recapture models. Bioscience, , –.
C O N Y E R S , J. & W I N G AT E , D. () Plestiodon longirostris. In The O T I S , D.L., B U R N H A M , K.P., W H I T E , G.C. & A N D E R S O N , D.R. ()
IUCN Red List of Threatened Species . Http://dx.doi.org/. Statistical inference from capture data on closed animal populations.
/IUCN.UK..RLTS.TA.en [accessed May Wildlife Monograph, , –.
]. P O L LO C K , K.H. () A capture–recapture design robust to unequal
C O R M AC K , R.M. () The statistics of capture–recapture methods. probability of capture. Journal of Wildlife Management, , –.
Oceanography and Marine Biology Annual Review, , –. R C O R E T E A M () R: a Language and Environment for Statistical
D A R R O C H , S.E., F I E N B E R G , G., G LO N E K , B. & J U N K E R , B. () A Computing. R Foundation for Statistical Computing, Vienna,
three sample multiple capture–recapture approach to the census Austria.
population estimation with heterogeneous catchability. Journal of RAINE, A. () A study of the morphological differentiation,
the American Statistical Association, , –. fluctuating asymmetry and the threats facing isolated
Oryx, Page 7 of 8 © 2019 Fauna & Flora International doi:10.1017/S0030605318001485
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 22 Dec 2020 at 04:16:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S00306053180014858 H. Turner et al.
populations of the critically endangered Bermuda Rock S E B E R , G.A.F. () The Estimation of Animal Abundance and Related
Lizard (Eumeces longirostris). MSc thesis, University College, Parameters. nd edition. Macmillan, New York, USA.
London, UK. S E B E R , G.A.F. () A review of estimating animal abundance.
R I V E S T , L.P. & B A I L L A R G E O N , S. () Applications and extensions Biometrics, , –.
of Chao’s moment estimator for the size of a closed population. T U R N E R , H. () Population status and conservation of the Critically
Biometrics, , –. Endangered Bermuda skink. PhD thesis, University of Kent,
R I V E S T , L.P. & B A I L L A R G E O N , S. () R Capture: Loglinear models Canterbury, UK.
for capture–recapture experiments. R package version .-. Http:// W H I T E , G.C. & B U R N H A M , K.P. () Program MARK: survival
CRAN.R-project-org/package=Rcapture [accessed March ]. estimation from populations of marked animals. Bird Study, ,
R I V E S T , L.P. & D A I G L E , G. () Loglinear models for the robust –.
design in mark–recapture experiments. Biometrics, , –. W I N G AT E , D.B. () Terrestrial herpetofauna of Bermuda.
R U B Y , D.E. () Seasonal changes in the territorial behavior of the Herpetologica, , –.
iguanid lizard Sceloporus jarrovi. Copeia, , –. WINGATE, R.S. () A comparison of demography and morphological
S AC C H I , R., S C A L I , S., P E L L I T T E R I -R O S A , D., P U P I N , F., G E N T I L L I , A., variation in two insular populations of the Bermuda rock lizard
T E T T A M A N T I , S. et al. () Photographic identification in reptiles: (Eumeces longirostris). MSc thesis, University of Wales, Swansea,
a matter of scales. Amphibia-Reptilia, , –. UK.
Oryx, Page 8 of 8 © 2019 Fauna & Flora International doi:10.1017/S0030605318001485
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 22 Dec 2020 at 04:16:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605318001485You can also read
