Content uploaded by Marc Cheylan
Author content
All content in this area was uploaded by Marc Cheylan on Feb 17, 2023
Content may be subject to copyright.
23
ecologia mediterranea – Vol. 48 (2) – 2022
Estimating the demographics of an ocellated
lizard (Timon lepidus Daudin, 1802)
population through photo identification
capture-recapture
Michel BELAUD1, Albert BERTOLERO2, Marc CHEYLAN3, *
1. 8 quartier Saint Pancrace, 06830 Gilee, France
2. Associació Ornitològica Picampall de les Terres de l’Ebre, La Galera 53, 43870 Amposta, Spain
3. École pratique des hautes études, Université Paris Sciences Leres, 4-14 rue Ferrus, Paris, France
* Corresponding author: marc.cheylan@gmail.com
Received: 28 Oct., 2022; First decision: 12 Nov., 2022; Revised: 18 Nov., 2022;
Second decision: 7 Dec., 2022; Revised: 9 Dec., 2022; Accepted: 20 Dec., 2022
the high “recapture” (via photo identification)
rates (between 0.465 ± 0.053 and 0.986 ± 0.014 in
spring, lower in autumn) and the isolation of the
study site.
Résumé
L’étude utilise la photo-identification pour suivre
une population de lézards ocellés (Timon lepidus
Daudin, 1802) dans le sud-est de la France. Les
données de capture-marquage-recapture (CMR)
issues de la photo-identification fournissent les
premières données sur la démographie de cee
espèce dans la nature. Le suivi a été réalisé sur
une période de six ans dans la vallée du fleuve
Var (Alpes-Maritimes), avec un rythme moyen
de 25visites par an. Mille huit cent soixante-dix
photographies exploitables ont permis d’iden-
tifier 138individus différents le long d’un parcours
linéaire de 1600 mètres. Les données d’histoire
de vie ont été analysées pour 114 lézards âgés de
plus d’un an, avec le logiciel MARK en utilisant
le modèle Cormack-Jolly-Seber (CJS). L’analyse
permet d’estimer la survie annuelle et saison-
nière (de mi-mai à fin septembre versus octobre
à mi-mai) chez les deux sexes. La survie annuelle
Abstract
This study used photo identification “capture-
recapture” to monitor a population of ocellated
lizards (Timon lepidus Daudin, 1802), providing
the first data on the demography of this species
in the wild. The monitoring was conducted over
a period of six years in a valley of the Var River in
southeast France, with an average rate of 25 visits
per year. This resulted in 1870 exploitable photo-
graphs that allowed the identification of 138 dif-
ferent individuals along a 1,600m linear path. For
the 114 lizards that were over the age of one, the
life history data was analysed with MARK soware
using the Cormack-Jolly-Seber (CJS) model. This
analysis enabled survival estimates –both annual
and seasonal (mid-May to end of September
versus October to mid-May)– for both sexes.
Annual survival appears to be higher in males than
in females (0.65 versus 0.48), with a life expectancy
of 2.3 years for adult/subadult males and 1.4 years
for adult/subadult females. During the winter
period, monthly survival is close to 1 for both sexes.
It is lower during the summer period: 0.92 in males
and 0.87 in females. These survival and longevity
rates are surprisingly low for a species of this size.
However, they can be considered reliable given
Mots-clés: Timon Lepidus, démographie,
CMR, sud de la France, identification photo,
monitoring; lézard ocellé.
Keywords: Timon lepidus, demography, CMR,
south of France, photo identification, monitoring,
ocellated lizard.
Michel Belaud, Albert Bertolero, Marc Cheylan
24 ecologia mediterranea – Vol. 48 (2) – 2022
capture-mark-recapture surveys. As a result,
no reliable data is available on the demog-
raphy of this threatened species (Grillet et al.
2006; Cheylan et al. 2011) (near threatened
species on the IUCN species list; Pleguezuelos
et al. 2009).
To improve demographic knowledge of this
species, this study used a technique that has
not yet been widely used on reptiles: photo
identification (Jackson et al. 2006; Bengsen
et al. 2011; Bolger et al. 2012; Cruickshank
& Schmidt 2017). It was possible to use it in
this case because of the distinctive patterns of
this species, but also because of the singular
configuration of the study site (the site is iso-
lated; see site description below). The findings
allow the first estimates of survival rates and
longevity in the wild for this species.
Materials and methods
The study site
The monitoring was carried out at a site located
in the valley of the Estéron River, which flows
into the Var River in the south of France, about
16km north of Nice (43.821°N/7.183°E). The
study area covered a length of 1,600 m and an
average width of about 50m along the western
bank of the Estéron River, stretching south
from the Nucera bridge (Figure 1). It was
embanked between 1920 and 1970 to protect
the valley from recurrent severe flooding of
the Var River, although the exact date of these
works on this part of the river is unknown.
The embankment consists of a concrete foun-
dation covering the side of the embankment at
an approximate 45° angle, with a 70cm wide
horizontal lip at the top (Figure 2). Concrete
blocks are stacked on the slope. Since the end
of 2014, the top of the embankment has been
made into a cycle path with a railing on the
river side to make walking and cycling safer.
Due to its configuration, the study area is vir-
tually isolated: to the east by the river bed,
to the west by a road with heavy traffic and
then a water body, all in a highly urbanised
context. Exchanges with the surroundings
are therefore very limited and, although the
species is present at the edge of the water body
on the other side of the road, no individuals
have ever been observed alive or dead on the
road despite regular visits to the site. The ocel-
lated lizards at the site shelter in the maze of
concrete blocks lining the bank and venture
semble être plus élevée chez les mâles que chez les
femelles (0,65 contre 0,48), avec une espérance de
vie de 2,3 ans pour les mâles adultes et subadultes
(> 1 an) et de 1,4 an pour les femelles adultes et
subadultes. Pendant la période hivernale, la survie
mensuelle est proche de 1 pour les deux sexes.
Elle est plus faible pendant la période estivale :
0,92 chez les mâles et 0,87 chez les femelles. Les
taux de survie mis en évidence sont étonnamment
bas pour une espèce de cee taille. Ils peuvent
cependant être considérés comme fiables étant
donné l’isolement du site d’étude et les forts taux
de « recapture » obtenus via la photo d’identifi-
cation (entre 0,465 ± 0,053 et 0,986 ± 0,014 au prin-
temps, plus faibles en automne).
Introduction
Knowledge of the demographic parameters of
a species – age at maturity, adult survival, mean
generation time – is fundamental to under-
standing its population dynamics, especially
when modelling the viability of a threatened
population (Akçakaya & Sjögren-Gulve
2000; Brook et al. 2000; Beissinger 2002).
Unfortunately, acquiring this information is
laborious, especially for discreet and uncap-
turable species: the case of many reptiles.
Estimating the survival rate or life expectancy
of a species in the wild requires multi-year
monitoring involving recognition of indi-
viduals – generally by individual marking –
which is difficult to put into practice for
most lizard species. As a result, demographic
studies on lizards mainly concern abundant and
easily captured species such as, for example,
in Europe, Zootoca vivipara, Lacerta agilis or
Podarcis sp. (Barbault & Mou 1988; Strijbosch
& Creemers 1988; Galán 1999; Berglind 2000;
Le Galliard et al. 2010; Rotger et al. 2020;
Diego-Rasilla et al. 2021).
The ocellated lizard (Timon lepidus Daudin,
1802) is a shy species that is difficult to capture
except in very specific situations: rabbit
burrows allowing the use of traps, artificial or
natural shelters facilitating the manual capture
of lizards (Grillet et al. 2010; Cheylan et al.
2011; Doré et al. 2015). The demographic data
available on this species is therefore almost
non-existent, apart from information derived
from skeletochronology almost four decades
ago (Castilla & Castanet 1986), an invasive
technique that can only be performed on dead
animals or on an amputated phalange (Comas
et al. 2016). Furthermore, this technique does
not provide information on survival rates,
which require multi-annual monitoring by
Estimating the demographics of an ocellated lizard (Timon lepidus Daudin, 1802) population
through photo identification capture-recapture
25
ecologia mediterranea – Vol. 48 (2) – 2022
Lizards cross the cycling path to assess grassy
areas between the cycling path and the road
where they feed (one lizard was found dead
on the cycle path, probably due to a collision
with a bike over the study period). Excluding
the area of the cycle track, which is not used
by lizards for feeding, the site exactly covers
81,400m².
little outside this refuge. This zone has the
advantage of being relatively inaccessible
to humans due to the incline of the bank and
a safety barrier separating the embankment
from the cycle path. The concrete blocks
provide valuable hiding places for the lizards,
especially the gap that has developed over
time between the blocks and the concrete base.
Figure 1 – View of the study site. The route taken to photograph the lizards is indicated in red
(photo M. Belaud).
Figure 2 – View of the study site – Details of the dike with the blocs and cycle path (photo M. Belaud).
Michel Belaud, Albert Bertolero, Marc Cheylan
26 ecologia mediterranea – Vol. 48 (2) – 2022
from a distance and of detecting distinguishing
marks between individuals (Steinicke et al.
2000; Perera & Perez-Mellado 2004; Sacchi
et al. 2016). From this point of view, the ocel-
lated lizard has the identification advantage
that individuals have varied lateral patterns.
Moreover, the size of this lizard allows an
image resolution that is sufficiently satis-
factory to examine the colouration details
(see photo examples in Figure 3). During
the survey visits, whenever possible lizards
were photographed from both sides, so that
the right and left side patterns were available.
Individuals photographed only on one side
were excluded from the results to allow for
unambiguous subsequent identification.
For the identification of individuals, we did
not use image recognition software (Matthé et
al. 2017), given the relatively small number of
individuals (138 different individuals over the
six years of the study). The identification was
carried out by directly comparing the photos
onscreen. This allowed us to use several cri-
teria to identify individuals: size, sex, side pat-
terns (blue markings, flank reticulation), head
scales and wounds. Flank patterns are a good
means of recognition (as shown in Figure 3),
as well as both the number and arrangement
of lateral blue markings and some dorsal
patterns. Head scales, particularly the labial
scales, can also be used as an identification
criterion given their variability. They also have
the advantage of being very similar on both
sides of the head. Wounds, which are quite
common in males due to fighting, are also an
aid to identification, as is the state of regener-
ation of the tail if it has been severed (14% of
individuals in our study). The main difficulty
lies in the identification of juveniles under
one year of age, for which profound marking/
colouration changes are observed. Without
obtaining frequent close-up photographs over
Data collection
The cycle path provides an ideal vantage point
to view and photograph the lizards below.
Accustomed to frequent human traffic, they
are much more confident here than elsewhere.
Over a period of six years (2015-2020), from
March to November, an observer (MB) walked
the cycle path from the Nucera bridge and
then back (a distance of 3,200m), generally
in the morning (between 9:00 and 13:00),
equipped with 10×40 binoculars and a 300mm
lens mounted on a digital SLR camera (Canon
7D mark II – 20.2 megapixels). The observer
walked slowly along the railing, stopping
every 10m. The immediate vicinity along the
track was examined first, as that would offer
the closest view of any lizards hidden in the
grass. Attention was then focused on the con-
crete blocks to the left and right and then on
the lower part of the embankment.
During the six-year study, the number of
annual monitoring visits varied between a
minimum of 4 (2015) and a maximum of
38 visits (2019) (Table 1), representing a total
of 315 hours of prospecting over a cumulative
distance of 451km. During the 150 visits of
the site, 996 ocellated lizards were observed
and 4026 photographs were taken of indi-
viduals. The precise time of the photograph
was also noted so that it could be referenced
to the previously calibrated time-stamped
photos, which were then labelled with the
specific code for the lizard.
Identification of individuals
Photo identification is a commonly used
monitoring technique today, especially for
mammals (Kelly 2001; Nipko et al. 2020).
It is less frequently used for reptiles, due to
the difficulty of photographing these animals
Table 1 – Sampling effort during the multi-year survey: number of annual visits; distance travelled during
visit; time spent on site (hours: minutes); number of photographs taken and used; number of
ocellated lizard individuals seen; number of different individuals observed.
Year # of visits km Time spent # of photographs # of lizards # of individuals
taken used seen different total
2015 4 1 3:10 66 14 4 4 4
2016 10 16 12:08 267 68 28 19 19
2017 36 115 98:12 1225 431 286 53 66
2018 34 109 78:39 1034 407 283 64 100
2019 38 122 71:20 925 543 252 58 125
2020 28 88 52:15 509 407 143 36 138
150 451 315:04 4,026 1,870 996 138
Estimating the demographics of an ocellated lizard (Timon lepidus Daudin, 1802) population
through photo identification capture-recapture
27
ecologia mediterranea – Vol. 48 (2) – 2022
The recapture probability for autumn 2016
was fixed at p = 0 (no sampling was carried
out during this period; Table 2). In order to
distinguish between summer and winter sur-
vival, we selected observations made in spring
(April to June) and autumn (September to
October) and eliminated observations from
July and August (Table 2). The time interval
between spring and autumn of each year was
set at 4.5 months to estimate summer sur-
vival (mid-May to end of September), and the
interval between autumn and the following
spring was set at 7.5 months to estimate winter
survival (October to mid-May). In this way, the
monthly survival estimates (
φ
) were directly
comparable between summer (su) and winter
(wi). In the survival modelling, we tested dif-
ferent biological hypotheses: variation due
time, for example from week to week, their
identification is almost impossible, as shown
in the photographs (Figure 4). For this reason,
juveniles under the age of one were excluded
from the demographic analyses.
Demographic analyses
Only capture histories of adult and subadult
lizards (age > 1 year) observed during the
period 2016 to 2020 were used for the analyses.
Each “recapture” corresponded to an obser-
vation validated by at least one photograph. To
estimate apparent survival (
φ
) and recapture
rates (p), we used the Cormack-Jolly-Seber
model (CJS: Lebreton et al. 1992) implemented
in program MARK (White & Burnham 1999).
Figure 3 – Example of paerns allowing the identification of the same adult individual over three years
(photo M. Belaud).
Figure 4 – Change in the paern of a juvenile photographed on four occasions over less than a year
(photo M. Belaud).
Michel Belaud, Albert Bertolero, Marc Cheylan
28 ecologia mediterranea – Vol. 48 (2) – 2022
model with the smallest value (ΔAICc) and
the Akaike weight (ωi) were used to select the
best model.
The graphs were made with R software (R
Core Team 2018) and the package tidyverse
(Wickham et al. 2019).
Results
Recapture rate
Recapture rates varied according to the obser-
vation effort made in each season (Figure 5;
Table 1). In all cases, they were higher in
spring than in autumn (between 0.465 ± 0.053
and 0.986 ± 0.014 in spring and between
0.219 ± 0.052 and 0.439 ± 0.052 in autumn;
first model in Table 4), reflecting both the
intensity of lizard activity and the frequency
of site visits.
Population size and density
Autumn sampling did not allow for reliable
population estimates (Table 2). The popu-
lation estimates are therefore based on spring
2017 to 2020, using closed population models.
The population size varied between a maximum
de 54.0 ± 4.9 (± SE) lizards in spring 2018
and a minimum of 34.2 ± 7.8 lizards in spring
2020 (Table 3), with no significant difference
to (1) sexual differences (s), (2) differences
between seasons and years (t), and (3) differ-
ences between summer and winter (season),
but without year differences for each season
(i.e. no difference between years for a given
season). We also assessed the existence of a
linear relationship between observation effort
(e: number of hours spent searching for lizards
each season) and recapture probability. Factors
and covariates were tested with interactions
(noted by *) or as additive effects (noted by +).
The average life expectancy for lizards older
than one year was calculated with the formula
LE = -1/ ln(φ), where φ was the annual survival
probability of each sex. Annual survival was
calculated from seasonal survival φ
annual
=
φ
su4.5
* φ
wi7.5
and the variation in φ
annual
was
estimated by the Delta method (Seber 1982).
Because the site is considered as isolated,
population size (spring data only) was cal-
culated with closed population models (Otis
et al. 1978), using the Huggins design in the
program MARK (White & Burnham 1999) (to
make sure, open population models were also
tested (Jolly-Seber model, POPAN program in
MARK) but did not provide good estimates).
We evaluated models that were constant (M0),
time dependent (Mt), dependent on behaviour
on first capture (Mb), with individual hetero-
geneity between capture probability (M
h
), and
a combination of these models (Mth, Mtb and
Mtbh; Otis et al. 1978; Williams et al. 2001).
Prior to the survival analysis, we verified the
goodness-of-fit of the overall model (φ
s*t
,
ps*t
) and the CJS model with U-Care software
(Choquet et al. 2009). The selection of sur-
vival and population size models was carried
out with the Akaike information criterion
corrected for small samples (AICc; Burnham
& Anderson 2002). The difference between
the AICc of each model and the AICc of the
Table 2 – Number of ocellated lizards identified in each season
and the seasonal observation effort.
Year Season Observation effort (hours) Lizards identified
2016 Spring 918
Autumn 0 0
2017 Spring 61 40
Autumn 7 8
2018 Spring 38 45
Autumn 17 24
2019 Spring 32 38
Autumn 12 17
2020 Spring 18 24
Autumn 12 8
Figure 5 – Resighting probability in spring (S) and
autumn (A) over 2017-2020 as estimated
from the model φs+season
, pe (resighting
probabilities can only be estimated from
2017 onward). The bars show the 95%
confidence intervals.
Estimating the demographics of an ocellated lizard (Timon lepidus Daudin, 1802) population
through photo identification capture-recapture
29
ecologia mediterranea – Vol. 48 (2) – 2022
Discussion
The unique conditions of this study – good
visibility of the animals, their habituation
to human presence, and the isolation of the
population – allowed a multi-year survey
via photo identification of this shy species,
which has been little accessible to population
dynamics studies. The study also allowed the
reliability of photo identification to be tested
as a long-term population monitoring method.
We found that the technique was effective for
the identification of subadults and adults older
than one year, but was problematic in juve-
niles, due to rapid changes in their colouration
pattern. In contrast to capture, which is gen-
erally traumatic for individuals and difficult to
carry out on this species, photo identification
offers the advantage of not disturbing the
between the four analysed years of monitoring
(2017-2020). Based on the most reliable popu-
lation estimate (54 individuals in 2018), the
density of individuals older than age one is
estimated to be in the range of a minimum of
5.98 ind/ha and a maximum of 8.59 ind/ha.
Sex ratio
In spring 2018, 25 adult females and 21 adult
males were identified, resulting in a balanced
sex ratio (1 female to 0.84 males; χ2 = 0.044,
df = 1, p = 0.835). In the same season, the
estimated population size was 34.2 ± 7.4
females (± SE; model Mh) and 30.4 ± 11.2
males (model Mb), i.e. also a balanced sex
ratio (1 female to 0.89 males; χ
2
= 0.031,
df = 1, p = 0.860).
Survival rate and life expectancy
Survival rates were calculated from the life
histories of 114 lizards older than one year
(66 females and 48 males). The data fitted
the CJS model satisfactorily (global GOF
test χ
2
= 29.514, df = 30, p = 0.491). The
retained survival model was supported with
an Aikake weight of 89.1% (Table 4). It shows
that the survival rates were higher for males
than females and, for both sexes, lower in
summer than winter (Figure 6). The annual
survival was 0.480 ± 0.062 (± SE) for females
and 0.650 ± 0.064 for males. From the second
year, average life expectancy was 1.4 years for
females and 2.3 years for males.
Table 3 – Estimated ocellated lizard population size (n) in spring 2017-2020.
Models: Mh = capture probability varies with individual; Mbh = capture
probability varies with individual and behavioural response to
capture; and Mtbh = capture probability varies with behavioural
response to capture, time and individual (Otis et al. 1978).SE = 95%
confidence interval.
Spring Model nSE Inf95 Sup95
P2017 Mh 53.08 10.88 44.80 99.43
P2018 Mh 54.04 4.88 48.68 70.12
P2019 Mtbh 43.37 6.54 38.83 72.79
P2020 Mbh 38.84 43.63 24.78 307.70
Table 4 – Models of apparent survival (φ) and recapture probability (p) for the
ocellated lizard population in the study site from 2016 to 2020. Models
were evaluated by comparing the ΔAICc, ωi is the Akaike weight
and k is the number of estimable parameters. Subscript notations
in the models: t = time; s = sex; season = summer and winter;
e= observation effort for recapture; and constant = no variation.
Model AICc ΔAICc ωikDeviance
φs+season , pe407.993 00.891 5166.846
φs+season , pt413.166 5.173 0.067 10 161.224
φseason , pt415.156 7.163 0.025 9165.415
φs , pt417.975 9.982 0.006 9 168.235
φs+t , pt418.953 10.959 0.004 15 155.670
φs , ps+t 420.168 12.175 0.002 10 168.226
φt , pt420.468 12.475 0.002 13 161.790
φs+t , ps+t 420.748 12.755 0.002 16 155.128
φconstant , pt421.210 13.217 0.001 8173.146
φt , ps+t 422.546 14.552 0.001 14 161.577
φconstant , ps+t 422.886 14.893 0.001 9173.146
φs+t , ps*t 428.170 20.176 0 20 152.959
φs*t , ps*t 436.455 28.462 024 151.248
Figure 6 – Monthly survival probability in summer
(mid-May to September) and winter
(October to mid-May) estimated from
the model φs+season
, pe. The bars show 95%
confidence intervals.
Michel Belaud, Albert Bertolero, Marc Cheylan
30 ecologia mediterranea – Vol. 48 (2) – 2022
to most small lacertid species, which have
a body mass 40 times lower than the ocel-
lated lizard. However, we found that one male
reached the age of 7, two reached the age of 6,
and two reached the age of 5.
Two studies using the skeletochronology
method allow comparisons with our findings.
In the south of France, the examination of
16 individuals resulted in an average age of
4.6 years (taking into account animals older
than one year) and a maximum age of 10
(Cheylan 1984). In central Spain, an average
age of 4.9 years was found in a sample of 76 indi-
viduals over the age of one, with a maximum
age of at least 11 (Castilla & Castanet 1986).
In captivity, the longest recorded life spans are
14 and 17 years (Decaux 1897; Flower 1925).
Skeletochronology studies therefore indicate
much higher life expectancy and longevity
than the survival rates we obtained, which
could be due to either (1) strong predation
pressure at the site or (2) an underestimation
of the estimated survival rate. Although it is
impossible to conclusively determine which
of these two hypotheses is at play in our case,
it should be noted that apparent survival rates
estimated by capture-mark-recapture are
always intrinsically lower than the reality,
since they do not distinguish between an indi-
vidual that has died and one that has left the
study site. Despite the fact that our study site
is highly isolated, it can be assumed that a few
individuals leave the site, especially subadult
individuals during the dispersal phase.
The size of the subadult and adult population,
estimated at between 48 and 70 individuals,
shows that this is a small population, although
we know that this estimate is low given that
some insufficiently photographed individuals
were excluded from the analysis. Due to the
small size and the isolation of this popu-
lation, it can be considered vulnerable in the
medium term. The risk of its disappearance is
exacerbated by the fact that the possibility of
exchange with the closest populations is near
non-existent. Previous studies on the ocel-
lated lizard in Liguria in Italy and the French
region of Alpes-Maritimes show the extent
to which these populations are threatened
– several have already become extinct (Deso
et al. 2015). As major redevelopment work on
the Var River is currently being considered, it
is to be feared that this will hasten the disap-
pearance of the ocellated lizard in this eastern
part of its distribution.
animals. However, the quality of the photos
can lead to significantly less information
compared to physical capture, and only pho-
tographing both sides of an individual’s body
allows for unambiguous subsequent identifi-
cation. The use of this method with other pop-
ulations of ocellated lizards or other species
of lizards may be problematic, except in very
specific situations allowing good visibility of
the animals.
Given the sampling effort (an average of
25 visits per year, with a maximum of 38 visits
in 2019), recapture rates per season were high,
ranging from a minimum of 29% in autumn
2017 to a maximum of 98% in spring 2017.
As would be expected, the findings show that
the “recapture” probability of an animal is
closely linked to the “capture” effort as well
as to the activity of the animals, which is more
intense in spring (April, May and June) than
in autumn (September and October). These
high “recapture” rates provide satisfactory
survival estimates, allowing an analysis of
whether these are differentiated by sex and
by season. Seasonally, there was a strong
contrast between apparent winter survival
(October to mid-May) and apparent summer
survival (mid-May to late September). Both
sexes taken together, survival was 99% in the
winter and 89% in the summer, showing that
the mortality risk, mainly due to predation,
is almost nil in the lizard’s winter phase and
higher during its active period. During the
active period (in this case, mid-May to end
September), females have a survival rate of
around 86%, while that of males is 92%, a dif-
ference of 6%. This difference in survival can
probably be explained by the risks involved in
reproduction: e.g. high food and thermoreg-
ulation requirements during the gestation
phase and less agility during pregnancy. In
our study site, the main predator could be
the Montpellier snake (Malpolon monspes-
sulanus), as it was frequently observed on
site and is a great consumer of lizards, in
particular ocellated lizards (Pleguezuelos
2021). The other potential diurnal predators
(corvids, gulls and a pair of short-toed snake
eagles nesting nearby) do not hunt on this site
because the cycling path, which is highly fre-
quented, is a disturbance.
Based on the estimated survival rates, the
average life expectancy of subadult and adult
ocellated lizards in this population is only
1.4 years for females and 2.3 years for males,
which are very low values for a lizard of this
size. Indeed, these are similar survival rates
Estimating the demographics of an ocellated lizard (Timon lepidus Daudin, 1802) population
through photo identification capture-recapture
31
ecologia mediterranea – Vol. 48 (2) – 2022
skeletochronology. Animal Biodiversity and
Conservation 39, 237-240.
Cruickshank, S.S. & Schmidt, B.R. (2017). Error
rates and variation between observers are reduced
with the use of photographic matching software
for capture-recapture studies. Amphi. Reptil. 38,
315-325.
Decaux C. (1897). Un lézard ocellé conservé en cap-
tivité depuis 14 ans. La Nature 49, 43-44.
Deso, G., Cevasco, J-M., Salvidio, S., Ottonella, D.,
Oneto, F. & Cheylan, M. (2015). Statut des popu-
lations franco-italiennes de Lézard ocellé Timon
lepidus (Daudin, 1801). Bull. Soc. Herp. France
156, 45-53.
Diego-Rasilla, F. J., & Pérez-Mellado, V. (2004). Density
dependence of survival in the Iberian wall lizard
(Podarcis hispanica). In: The Biology of Lacertid
lizards. Evolutionary and Ecological Perspectives
(eds Pérez-Mellado V., Riera N., Perera A.). Institut
Menorquí d’Estudis. Recerca 8, 177-200.
Doré, F., Cheylan, M. & Grillet, P. (2015). Le lézard
ocellé. Un géant sur le continent européen. Biotope,
Mèze.
Flower, S.S. (1925). Contributions to our Knowledge
of the Duration of Life in Vertebrate Animals. III.
Reptiles. Proceedings of the Zoological Society of
London 95(3), 911-981.
Galán, P. (1999). Demography and population dynamics
of the lacertid lizard Podarcis bocagei in north-west
Spain. Journal of Zoology 249, 203-218.
Grillet, P., Cheylan, M. & Dusoulier, F. (2006). Évolution
des habitats et changements climatique : quelles
conséquences pour les populations de lézard ocellé,
Lacerta lepida (Saurien, Lacertidés) en limite nord
de distribution? Ecologia Mediterranea 32, 63-72.
Grillet, P., Cheylan, M., Thirion, J.M., Doré, F., Bonnet,
X., Dauge, C., Chollet, S. & Marchand, M.A.
(2010). Rabbit burrows or artificial refuges are a
critical habitat component for the threatened lizard
Timon lepidus (Sauria, Lacertidae). Biodiversity and
Conservation 19, 2039-2051.
Jackson, R.M., Roe, J.D., Wangchuk, R. & Hunter, D.O.
(2006). Estimating snow leopard population abun-
dance using photography and capture-recapture
techniques. Wildl. Soc. Bull. 34, 772-781.
Kelly, M. J. (2001). Computer-aided photograph mat-
ching in studies using individual identification:
an example from Serengeti cheetahs. Journal of
Mammalogy 82, 440-449.
Lebreton, J.-D., Burnham, K.P., Clobert, J. & Anderson,
D.R. (1992). Modeling survival and testing biolo-
gical hypothesis using marked animals: a unified
approach with case studies. Ecological Monographs
62, 67-118.
Le Galliard, J.F., Marquis, O. & Massot, M. (2010).
Cohort variation, climate effects and population
dynamics in a short-lived lizard. Journal of Animal
Ecology 79, 1296-1307.
Matthé, M., Sannolo, M., Winiarski, K., Spitzen-van der
Sluijs, A., Goedbloed, D., Steinfartz, S. & Stachow,
U. (2017). Comparison of photo-matching algo-
rithms commonly used for photographic capture-
recapture studies. Ecol. Evol. 7, 5861-5872.
Nipko, R.B., Holcombe, B.E. & Kelly, M.J. (2020).
Identifying Individual Jaguars and Ocelots
via Pattern-Recognition Software: Comparing
HotSpotter and Wild-ID. Wildl. Soc. Bull. 44,
424-433.
Acknowledgements
We would like to thank our naturalist col-
leagues Patrick Kern, Gérard Joannes,
Monique and Guy Poncet for the photos of
geo-localised individuals they kindly sent us
and Elise Bradbury for the translation of this
article into English.
References
Akçakaya, H.R. & Sjögren-Gulve, P. (2000). Population
viability analyses in conservation planning: an
overview. Ecological Bulletins 48, 9-21.
Barbault, R. & Mou, Y. P. (1988). Population dynamics
of the common wall lizard, Podarcis muralis, in sou
-
thwestern France. Herpetologica 44, 38-47.
Beissinger, S. R. (2002). Population viability analysis:
past, present, future. In: Population viability ana-
lysis (ed Beissinger S.R.). University of Chicago
Press, Chicago, 5-17.
Bengsen, A., Butler, J. & Masters, P. (2011). Estimating
and indexing feral cat population abundances using
camera traps. Wildl. Res. 38, 732-739.
Berglind, S.Å. (2000). Demography and management
of relict sand lizard Lacerta agilis populations on
the edge of extinction. Ecological Bulletins 48,
123-142.
Bolger, D.T., Morrison, T.A., Vance, B., Lee, D. & Farid,
H. (2012). A computer-assisted system for photo-
graphic mark-recapture analysis. Methods Ecol.
Evol. 3, 813-822.
Brook, B.W., O’Grady, J.J., Chapman, A.P., Burgman,
M.A., Akcakaya, H.R. & Frankham, R. (2000).
Predictive accuracy of population viability ana-
lysis in conservation biology. Nature, 404 (6776),
385-387.
Burnham, K.P. & Anderson, D.R. (2002): Model
Selection and Inference, second edition. A Practical
Information-Theoretic Approach. Springer, New
York.
Castilla, M.A. & Castanet, J. (1986). Growth, age and
longevity of Lacerta lepida assessed by skeletochro-
nology. In: Studies in Herpetology (ed Rocek Z.).
Charles University, Prague, 331-336.
Cheylan, M. (1984). Croissance et détermination de
l’âge chez le lézard ocellé (groupe Lacerta lepida,
Sauria, Lacertidae) de France et du Maroc à partir
de la squelettochronologie. Bull. Mus. Hist. Nat.
Marseille 44, 29-37.
Cheylan, M., Grillet, P. & Doré, F. (2011). Dynamique
populationnelle du lézard ocellé. In: Thompson J.
& Gauthier P. (eds), Activités humaines et dyna-
mique de la biodiversité en région méditerranéenne,
CEFE, Montpellier, 25-26.
Choquet, R., Lebreton, J.-D., Reboulet, A.-M., Pradel,
R. & Gimenez, O. (2009). U-CARE: Utilities for
performing goodness-of-fit tests and manipu-
lating CApture-REcapture data. Ecography 32,
1071-1074.
Comas, M., Reguera, S., Zamora-Camacho, F.J.,
Salvadó i Cabré, H. & Moreno-Rueda, G. (2016).
Comparison of the effectiveness of phalanges vs.
humeri and femurs to estimate lizard age with
Michel Belaud, Albert Bertolero, Marc Cheylan
32 ecologia mediterranea – Vol. 48 (2) – 2022
Sacchi, R., Scali, S., Mangiacotti, M., Sannolo, M.
& Zuffi, M.A. (2016). Digital identification and
analysis. In: Reptile Ecology and Conservation. A
Handbook of Techniques (ed Dodd C.K.). Oxford
University Press, New York, 59-72.
Seber, G.A.F. (1982). The Estimation of Animal
Abundance and Related Parameters, 2nd ed.
Charles Griffin & Company Ltd, London.
Steinicke, H., Ulbrich, K., Henle, K. & Grosse,
W.R. (2000). Eine neue Methode zur fotogra-
fischen Individualidentifikation mittelerupäischer
Halsbandeidechsen (Lacertidae). Salamandra 36,
81-88.
Strijbosch, H. & Creemers, R.C.M. (1988). Comparative
demography of sympatric populations of Lacerta
vivipara and Lacerta agilis. Oecologia 76, 20-26.
White, G. C., & Burnham, K. P. (1999). Program
MARK: survival estimation from populations of
marked animals. Bird study 46, 120-139.
Wickham, H. et al. (2019) Welcome to the tidyverse.
Journal of Open Source Software 4, 1686. https://
doi.org/10.21105/joss.01686
Williams, B.K., Nichols, J.D. & Conroy, M.J. (2001).
Analysis and management of animal populations,
Academic Press. San Diego.
Otis, D.L., Burnham, K.P., White, G.C. & Anderson,
D.R. (1978). Statistical inference from capture data
on closed animal populations. Wildlife Monogr. 62,
1-135.
Perera, A. & Perez-Mellado, V. (2004). Photographic
identification as a non-invasive marking technique
for Lacertid lizard. Herpetol. Rev. 35, 349-350.
Pleguezuelos, R. M, Sá-Sousa, P., Pérez-Mellado, V.,
Marquez, R., Cheylan, M., Corti, C. & Martínez-
Solano, I. (2009). Timon lepidus. The IUCN Red List
of Threatened Species 2009: e.T61583A12498949.
https://dx.doi.org/10.2305/IUCN.UK.2009.RLTS.
T61583A12498949.en Accessed on 04 January
2023.
Pleguezuelos, J.M. (2021). Culebra bastarda - Malpolon
monspessulanus. In: Enciclopedia Virtual de los
Vertebrados Españoles (eds López P., Martín J.,
Martínez-Freiría F.). Museo Nacional de Ciencias
Naturales, Madrid. http://www.vertebradosibericos.
org/
R Development Core Team. (2018). R: A Language
and Environment for Statistical Computing. R
Foundation for Statistical Computing, Vienna,
Austria.
Rotger, A., Igual, J. M., & Tavecchia, G. (2020).
Contrasting size-dependent life history strategies
of an insular lizard. Current Zoology 66, 625-633.
