Genomic repercussions of landscape modification on three lizard species

Abstract

Habitat destruction is the most pervasive threat to global biodiversity, leading to widespread population declines and range reductions. Land clearing can leave small, isolated populations persisting in remnant habitat, where demographic factors may erode genomic diversity and diminish adaptive potential. We compared the genomic structure, diversity, inbreeding and effective population sizes of fragmented populations on farms to nearby populations in large, continuous tracts of vegetation (national park) for three terrestrial lizard species in south-eastern Australia. Due to the small spatial scale of the study, observed levels of genomic differentiation among sampling locations were typically very low (FST < 0.1). The farm locality of one species, the painted dragon (Ctenophorus pictus), showed substantially more differentiation to national park localities (FST > 0.05) than the national park localities showed to one another (FST < 0.01), suggestive of genetic isolation due to the agricultural matrix. Genomic diversity and effective population sizes were lower in farm populations compared to national parks for two of the three species, the exception being shrubland morethia (Morethia obscura), where genomic diversity was similar across site types. Inbreeding coefficients were generally comparable between farm and national park populations. Our findings highlight the genetic consequences of land clearing including low population size, low genomic diversity and higher risk of inbreeding depression. Despite these challenges, habitat fragments can maintain high biodiversity value, which can be maximised by management initiatives such as translocations and establishing habitat corridors.

Full text

Vol.:(0123456789)
Conservation Genetics
https://doi.org/10.1007/s10592-025-01686-2
RESEARCH
Genomic repercussions oflandscape modification onthree lizard
species
DylanM.Westaway1,2· DaleG.Nimmo1,2· ChrisJ.Jolly1,3,4· DamianR.Michael1,2· DavidM.Watson1,2·
BrentonvonTakach5
Received: 1 December 2024 / Accepted: 28 February 2025
© The Author(s) 2025
Abstract
Habitat destruction is the most pervasive threat to global biodiversity, leading to widespread population declines and range
reductions. Land clearing can leave small, isolated populations persisting in remnant habitat, where demographic factors
may erode genomic diversity and diminish adaptive potential. We compared the genomic structure, diversity, inbreeding and
effective population sizes of fragmented populations on farms to nearby populations in large, continuous tracts of vegeta-
tion (national park) for three terrestrial lizard species in south-eastern Australia. Due to the small spatial scale of the study,
observed levels of genomic differentiation among sampling locations were typically very low (FST < 0.1). The farm locality
of one species, the painted dragon (Ctenophorus pictus), showed substantially more differentiation to national park localities
(FST > 0.05) than the national park localities showed to one another (FST < 0.01), suggestive of genetic isolation due to the
agricultural matrix. Genomic diversity and effective population sizes were lower in farm populations compared to national
parks for two of the three species, the exception being shrubland morethia (Morethia obscura), where genomic diversity was
similar across site types. Inbreeding coefficients were generally comparable between farm and national park populations.
Our findings highlight the genetic consequences of land clearing including low population size, low genomic diversity and
higher risk of inbreeding depression. Despite these challenges, habitat fragments can maintain high biodiversity value, which
can be maximised by management initiatives such as translocations and establishing habitat corridors.
Keywords Gene flow· Genetic diversity· Genomics· Habitat fragmentation· Landscape connectivity· Lizard
Introduction
With over half of all ice-free land on Earth converted for
agriculture (Hooke etal. 2013), habitat destruction is the
most widespread threat to global biodiversity (Bergstrom
etal. 2021; Jaureguiberry etal. 2022). Therefore, to succeed,
conservation must occur within agricultural landscapes.
Critical to informing conservation actions is understanding
the impacts of land modification on biodiversity (Fischer and
Lindenmayer 2007). Increasingly, genomic tools are being
used to provide such information, by identifying evolutionar-
ily significant units (Paplinska etal. 2011), assessing popula-
tion size (Frankham etal. 2014) and connectivity (Lettoof
etal. 2021), and revealing adaptive potential in the face of
environmental change (Hohenlohe etal. 2021).
In large, well-connected populations, gene flow freely
distributes alleles throughout the population, maintain-
ing genetic diversity, facilitating adaptation, and reducing
inbreeding depression and genetic drift (Frankham etal.
2002). Landscape modification, such as land clearing for
agriculture, can create isolated populations persisting in
remnant patches, with little or no gene flow throughout
the agricultural matrix (Driscoll etal. 2013). Fragmented
populations are more vulnerable to genetic risks such as
* Dylan M. Westaway
dwestaway93@gmail.com
1 Gulbali Institute forAgriculture, Water andEnvironment,
Charles Sturt University, Thurgoona, NSW2640, Australia
2 School ofAgricultural, Environmental andVeterinary
Sciences, Charles Sturt University, Thurgoona, NSW2640,
Australia
3 School ofNatural Sciences, Macquarie University,
MacquariePark, NSW2109, Australia
4 Research Institute fortheEnvironment andLivelihoods,
Charles Darwin University, Casuarina, NT0810, Australia
5 School ofMolecular andLife Sciences, Curtin University,
Perth, WA, Australia
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Conservation Genetics
inbreeding depression, loss of genetic diversity, and reduced
ability to adapt to environmental change (Frankham etal.
2017). These issues, combined with demographic and envi-
ronmental stochasticity, can lead to localised extinctions in
fragments (Fischer and Lindenmayer 2007) and—if a hos-
tile matrix precludes recolonisation—can trigger an ‘extinc-
tion rachet’ where each local extirpation brings the regional
or global population closer to extinction (Templeton etal.
1990).
Despite these issues, small patches can be of high con-
servation value (Wintle etal. 2019; Riva and Fahrig 2022).
Fragments can comprise the last vestiges of contracting
species’ ranges, and provide ‘insurance populations’, safe-
guarded against contagious disturbances like wildfire and
disease (Westaway etal. 2024b). Recognizing the high bio-
diversity value of small fragments raises the question of how
to best manage fragmented metapopulations (Lindenmayer
2019). Optimal management should strive to leverage the
benefits of fragments whilst addressing challenges faced by
populations in small fragments. However, management deci-
sions can be difficult when little is known about historical
connectivity of focal species in the area, population sizes in
contemporary habitat fragments, genomic diversity within
populations, or genomic structuring between populations.
Population genomics can inform all of these issues and pro-
vide critical information for designing conservation strate-
gies (von Takach etal. 2024). For example, low genomic
diversity and population size may warrant ‘genetic rescue’
via translocations (Whiteley etal. 2015). This involves intro-
ducing individuals from other locations to augment popula-
tion size and heterozygosity, increasing population viability
and adaptive potential (Wright etal. 2022). Such genetic
rescue attempts have been successful for a variety of taxa
including the European adder (Vipera berus; Madsen etal.
1999), mountain pygmy possum (Burramys parvus; Weeks
etal. 2017), Rocky Mountain bighorn sheep (Ovis canaden-
sis; Hogg etal. 2006) and Florida panther (Puma concolor
couguar; Hostetler etal. 2013).
We assessed genomic structure and diversity of three
terrestrial lizard species between fragmented populations
and populations in large, continuous tracts of vegetation
(national park). We used three lizard species (one agamid
and two scincids) for our study because such vertebrates:
(1) are likely to have dispersal impeded by land clearing,
and (2) may persist in small fragments due to their potential
for high population densities and small home ranges. We
expected land clearing and subsequent sheep grazing to ren-
der the agricultural matrix inhospitable to small lizards (due
to factors such as lack of vegetation structure, low resource
availability, and increased predation pressure), resulting in
reduced gene flow across the matrix. Thus, we predicted that
fragment populations would show some level of genomic
differentiation from adjacent national park populations for
all three species. We expected less differentiation for painted
dragon (Ctenophorus pictus) and robust ctenotus (Ctenotus
robustus) populations due to their larger body size, poten-
tially facilitating greater dispersal capacity (Jenkins etal.
2007), and tolerance of disturbed landscapes (Read 2002;
Shea 2010; Michael etal. 2011). These traits may enable
greater gene flow across the agricultural matrix compared to
shrubland morethia (Morethia obscura), which has a narrow
habitat breadth (Westaway etal. 2024b) and may be sensitive
to land modification (Driscoll 2004; Simpson etal. 2023).
We predicted lower genomic diversity, and higher levels
of inbreeding and relatedness in fragments compared to
national park populations for all three species, with greater
differences for the painted dragon due to substantially lower
population sizes compared to the two skink species (Westa-
way etal. 2024b) resulting in more rapid genetic drift.
Materials andmethods
Study area
Our study took place in the Little Desert National Park and
surrounding agricultural land in western Victoria, Australia
(Fig.1). The landscape is characterised by sandy soils and
contains a series of undulating dunes and swales, and expan-
sive plains hosting a variety of vegetation communities (see
S1.1 for further detail). Outside the national park, agricul-
ture (predominantly sheep grazing and cereal cropping) is
widespread. We targeted marginal farming land, abutting the
national park, where small fragments of vegetation were left
uncleared to provide shelter for livestock when land clearing
began in the study area in the 1960s. These fragments occur
on the same soils and support the same vegetation types as
the adjacent national park, providing an opportunity to test
landscape ecological questions relating to patch mosaics and
the habitat fragmentation matrix without the confounding
effects of different vegetation types and variation in resource
availability and productivity.
Study species
Shrubland morethia (M. obscura) is a small (snout-vent-
length up to 51mm), diurnal scincid lizard, which preys
upon small invertebrates. Despite its large distribution and
sometimes high local abundance, this species has a nar-
row habitat breadth (third lowest habitat breadth of the 17
detected reptile species in our study landscape; Westaway
etal. 2024b) and may be rare or absent from fragmented
landscapes (Driscoll 2004; Simpson etal. 2023). Robust
ctenotus (C. robustus) is a large (SVL up to 110mm), diur-
nal scincid lizard, which constructs shallow burrow systems,
and preys primarily upon arthropods, but occasionally eats
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Conservation Genetics
smaller skinks and native berries (Robertson and Coventry
2019). It is a habitat generalist with a widespread distri-
bution through eastern and northern Australia (Wilson and
Swan 2021). The painted dragon (C. pictus) is a small (SVL
up to 70mm), terrestrial agamid lizard that digs shallow
burrows for refuge (Westaway etal. 2024a). They are a pre-
dominantly ‘annual’ species, with around 90% of individuals
dying after their first year (Olsson etal. 2007), making popu-
lations vulnerable to localised extinctions should a breeding
season be unsuccessful.
Sample collection
Samples were collected from animals captured in a
network of pitfall traps established across 11 pairs of
fragments (n = 11), in the agricultural landscape, and
‘pseudo-fragments’ (n = 11), in the national park (Fig.1B;
Westaway etal. 2024b). Each fragment was paired with a
pseudo-fragment—an area of the same size as the frag-
ment, comprised of the same vegetation type, same fire
age class, but embedded within the continuous habitat of
the national park (MacNally and Bennett 1997; Johnstone
etal. 2014). All fragments were surrounded by cleared
grazing pasture, presumably inhospitable to our focal spe-
cies, and so were considered separate populations, whilst
pseudo-fragments in close proximity, connected by con-
tinuous vegetation, were grouped into the same population
(Fig.1C). Therefore, we only collected samples from two
of the 11 available fragments (Farm_W and Farm_C) as
preliminary work suggested that these two sites were most
likely to produce sufficient sample sizes of individuals for
Fig. 1 A National scale: study area location in Victoria, Australia.
B Regional scale: study fragments (solid triangles) and pseudo-frag-
ments (hollow triangles) across the Little Desert National Park and
surrounds. C Landscape scale: sampling site locations grouped by
colour into populations. Continuous vegetation of the national park
is shaded grey, while fragments of remnant native vegetation in the
agricultural landscape are hashed grey. Highways are represented by
black lines
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Conservation Genetics
population genomic analysis. Samples were collected from
all 11 pseudo-fragments (Fig.1C).
Tissue samples were taken by clipping 5mm off the end
of the tail of study lizards using sharp and sterilised scissors.
The removal of tail tips (< 10mm) has been shown to have
no effect on lizard behaviour (García-Muñoz etal. 2011).
Samples were stored in 90% ethanol and kept refrigerated
(~ 4°C) for up to 18months before being sent off for DNA
extraction and subsequent DNA sequencing.
DNA extraction, SNP genotyping andfiltering
DNA extraction for shrubland morethia samples was per-
formed by Museums Victoria staff. Extractions were con-
ducted using the consumables and standard protocol from
the Qiagen DNeasy Blood & Tissue Kit (Qiagen, Inc.). The
digestion protocol was adjusted to manage for the potential
interference by non-digestible scales. Specifically, samples
were left overnight to ensure thorough tissue digestion, and
afterwards, scales were removed from the solution. Extrac-
tions were eluted in Milli-Q purified water for quality testing
using a QIAxpert (Qiagen, Inc.). Where necessary, either to
increase DNA concentration or sample quality, Agencourt
AMPure XP beads (Beckman Coulter, Inc.) were used
according to a standard PCR purification protocol.
DNA extraction for the remaining two focal species, and
single-nucleotide polymorphism (SNP) genotyping for all
species, were performed by Diversity Arrays Technology
Pty. Ltd. (DArT Pty Ltd, Canberra, ACT, Australia). Diver-
sity Arrays Technology uses a combination of genome com-
plexity reduction methods and next generation sequencing
platforms (Sansaloni etal. 2011; Kilian etal. 2012). The
resulting datasets were read into R (Version 4.3.3; R Core
Team 2024) and exposed to the following filtering pipeline
in the dartRverse package (Gruber etal. 2018).
First, samples with > 35% missing data across SNPs were
filtered out to remove poorly genotyped individuals. Sam-
ples belonging to populations represented by < 3 individuals
were also filtered out, due to uncertainty associated with
such small sample sizes, followed by the removal of mono-
morphic loci. We then checked for sex-linked SNPs using
the dartR.sexlinked package (Robledo‐Ruiz etal. 2023) but
found none that were sex-linked. To remove potential bias
from sequencing or genotyping errors, we filtered out loci
according to read depth with a lower threshold of 10 and
an upper threshold of mean read depth plus two standard
deviations (Attard etal. 2024). We then examined repro-
ducibility between pairs of technical replicates, filtering out
loci with < 99% reproducibility. Missing data was managed
at both the locus (SNPs retained with per-locus genotype
call rate > 0.95) and individual (individuals retained with
genotype call rate > 0.9) level. For the painted dragon, we
lowered the per-locus call rate threshold to 0.90 to retain a
more robust number of SNPs, as the genotyping call rate
was substantially lower for this species (TableS3). Finally,
we excluded loci with a minor allele count of ≤ 3 to adjust
for potential sequencing errors, and filtered out secondary
SNPs, using the ‘best’ option where multiple SNPs occurred
on a single RAD locus.
To ensure our dataset was not biased by highly related
individuals, we investigated pairwise kinship between all
individuals of each species using the ‘beta.dosage’ func-
tion of the hierfstat package (Goudet 2005). For any pairs
of individuals with relatedness > 0.25, one individual was
removed from the dataset. Once filters had been applied,
we retained 21,527 SNPs from 89 shrubland morethia indi-
viduals (TableS1), 20,171 SNPs from 61 robust ctenotus
individuals (TableS2), and 7,650 SNPs from 45 painted
dragon individuals (TableS3).
Population genomic structure
Preliminary investigation of population genomic structure
included visualising structure via a principal coordinates
analysis (PCoA), calculation of genomic differentiation
among a priori populations, and investigation of the pat-
tern of isolation-by-distance. To construct the PCoA, we
calculated Euclidean genetic distance between all individu-
als via the ‘gl.dist.ind’ function of the dartR.base package
(Gruber etal. 2018), and the ‘cmdscale’ function of the
stats package (R Core Team 2024). To test the validity of
our a priori groupings of sampling sites into populations,
we first calculated pairwise FST values for all combinations
of sampling sites in the dartR.base package (Gruber etal.
2018) with 999 bootstraps used to generate confidence inter-
vals and p-values. Pairwise FST values were near zero for
all grouped sampling sites (TableS4), so we retained our a
priori population groupings. We then recalculated pairwise
FST values to estimate genomic differentiation between these
populations. To investigate patterns of isolation-by-distance,
we explored the correlation of pairwise FST values against
pairwise geographic distances using a Mantel test in the
package dartR.spatial (Mantel 1967; Gruber etal. 2018).
Genomic diversity
To investigate population genomic diversity, we estimated
observed heterozygosity (HO) and expected heterozygosity
(HE) using the dartR.base package (Gruber etal. 2018). An
inbreeding coefficient was calculated at the individual and
population level using the ‘snpgdsIndInbCoef’ function of
the SNPRelate package (Zheng etal. 2012). Multi-locus
heterozygosity (MLH) was calculated for each individual
using the inbreedR package (Stoffel etal. 2016). We calcu-
lated the number of private alleles present in each population
using the poppr package (Zheng etal. 2012) as an additional
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Conservation Genetics
measure of genomic distinctiveness. To standardise the
number of private alleles across populations with different
sample sizes, we bootstrapped private allele calculations
by resampling seven individuals per population 100 times
(without replacement), taking the mean and standard error
of all bootstraps.
Correlation betweengenomic diversity andbody condition
At the individual level, genomic metrics such as heterozy-
gosity and inbreeding can often be related to individual fit-
ness (Reed and Frankham 2003; Lettoof etal. 2021). Body
condition estimates using length and weight data have been
shown to accurately predict fat stores and organ mass (e.g.,
muscle tissue, liver) in reptiles (Madsen and Shine 2002;
Weatherhead etal. 2009), thus providing a useful proxy for
fitness. To evaluate individual fitness, body condition scores
were calculated according to the scaled mass index (SMI)
formula: SMI = Mi(L0 / Li) bSMA where Mi and Li are the
mass (grams) and length (snout-vent length) of individuals,
L0 is the arithmetic mean length of all sampled individuals,
and bSMA is the scaling exponent estimated by the standard-
ised major axis regression of mass on length of all sampled
individuals (Peig and Green 2009). We consider the SMI to
be an estimate of individual fitness with higher values cor-
responding to better condition. We calculated SMI, and per-
formed the following analyses, only for the painted dragon
as this was the only species for which we had body mass and
length data (recorded for another study).
We used two methods to explore the relationship between
painted dragon body condition and individual genomic met-
rics (MLH and inbreeding coefficient). First, we performed
Pearson’s correlation tests comparing MLH and SMI, and
then comparing individual inbreeding coefficient and SMI.
Second, we used generalised linear models, with a gaussian
error distribution, in the lme4 package (Bates etal. 2015)
with SMI as the response variable and site and one of MLH
or inbreeding coefficient as response variables. We carried
out these analyses on both the total dataset, and a subset of
the data which excluded two individuals (ID: 68,189 and
67,931) to test if the presence of outliers affected results.
Relatedness
To assess patterns of relatedness, we calculated pairwise
relatedness for all combinations of individuals for each spe-
cies using the ‘beta.dosage’ function of the hierfstat pack-
age (Goudet 2005). Here, we used the filtered SNP dataset
for each species but retained related individuals which were
omitted at the last stage of the filtering process for other
analyses (Tables S1, S2 and S3), and restricted compari-
sons to individual sampling sites, rather than our a priori
designated populations. This means that relatedness was
compared at equivalent spatial scales across the study area.
To explore the effect of land use on relatedness, we per-
formed linear mixed effect models with pairwise relatedness
values (within sampling sites) as the response variable, land
use (farm vs. national park) as a fixed effect and site as a
random effect. We added a constant and log-transformed
(log shift transformation) the response variable to improve
model fit based on residual plots. We used 999 bootstraps
to generate estimated means and confidence intervals for
sampling sites where n > 3.
Local effective population size
The effective population size (Ne) refers to the number of
breeding individuals in an idealised population that would
show the same amount of genetic drift or inbreeding as in
the study population (Luikart etal. 2010). A low Ne value
suggests a small, isolated population vulnerable to the
effects of genetic drift and inbreeding depression, while a
large Ne value suggests a large, genetically diverse and resil-
ient population (Frankham etal. 2014). We estimated Ne for
each population using the linkage disequilibrium method
within the ‘NeEstimator’ v2.1 software (Do etal. 2014) via
the ‘gl.LDNe’ function in the dartR.popgen package (Gruber
etal. 2018). As lack of independence becomes more severe
with increasing number of loci used and Ne estimate preci-
sion plateaus after a few thousand SNPs (Waples 2024), we
used a subset of 3,000 of the most informative SNPs for each
species (determined by the method = “pic” command). To
reduce bias in estimates, we screened out singleton alleles
(Waples 2024) and present Ne estimates and parametric con-
fidence intervals for each population.
Results
Population genomic structure
The three lizard species showed varying levels of genomic
structure across the sampled populations. While the top two
axes explained modest levels of variance (shrubland more-
thia: 3.2%; robust ctenotus: 5.4%; painted dragon: 9.4%), the
analysis highlighted differing patterns of genomic structure
among species (Fig.2). No clustering was evident for the
shrubland morethia (Fig.2A). A small amount of clustering
was evident for the robust ctenotus (Fig.2B) with Farm_C
and Farm_W populations showing some separation from
the national park population. For the painted dragon, two
distinct clusters were observed, with the farm fragment pop-
ulation separating from the two national park populations
(Fig.2C). Across all three species, there was some evidence
of farm populations clustering more loosely compared to
national park populations (Fig.2).
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Conservation Genetics
Pairwise FST values were generally low across all spe-
cies (Table1). All pairwise FST values were statistically
significant, except for some values involving the two most
poorly sampled populations (shrubland morethia popula-
tion NP W 2 and NP C 1; TableS6). Absolute values were
lowest for the shrubland morethia (FST = 0.001–0.003),
with very little genomic differentiation among all pop-
ulations. Pairwise FST values were larger for the robust
ctenotus (FST = 0.003–0.017) and broadly aligned with
geographic distance between populations. Painted dragons
showed the most differentiation (FST = 0.005–0.054). Nota-
bly, FST was substantially higher between the isolated farm
fragment and nearby national park site (FST = 0.051), com-
pared to between the two national park sites (FST = 0.005),
despite the former pair being geographically closer than
the latter (Fig.1). Mantel tests showed a significant
positive effect of isolation-by-distance for all three spe-
cies, although slopes of the effect varied across species
(TableS9, FiguresS1, S2 and S3).
Fig. 2 Multidimensional scaling (MDS) plot showing the genomic
structure of sampled populations in western Victoria, Australia for:
A shrubland morethia (Morethia obscura), B robust ctenotus (Cteno-
tus robustus), and C painted dragon (Ctenophotus pictus). Numbers
in parentheses after axis titles show the proportion of total variance
explained by each axis. Photos by Owen Lishmund
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Conservation Genetics
Genomic diversity
For the shrubland morethia and robust ctenotus, observed
and expected heterozygosity values were generally similar
across populations, although most populations showed mar-
ginally higher expected heterozygosity than observed het-
erozygosity (Table2). Values of expected heterozygosity
for populations of the painted dragon ranged from 0.168 to
0.189, and were substantially higher than values of observed
heterozygosity, which ranged from 0.123 to 0.125, suggest-
ing a general deficit of heterozygotes in these populations.
While, for all three species, almost all pairwise comparisons
of expected heterozygosity between sites were significantly
different, painted dragon heterozygosity showed the most
Table 1 Pairwise FST values for populations of three lizard species (shrubland morethia, robust ctenotus, and painted dragon) across isolated
farm fragments and national park in western Victoria, Australia
Shrubland morethia (Morethia obscura)
Farm W NP W 1 NP W 2 Farm C NP C 1 NP C 2
Farm W 0
NP W 1 0.001 0
NP W 2 < 0.001 < 0.001 0
Farm C 0.003 0.002 0.002 0
NP C 1 0.002 0.002 0.001 0.001 0
NP C 2 0.003 0.002 0.002 0.001 0.001 0
Robust ctenotus (Ctenotus robustus)
Farm W NP W 1 Farm C
Farm W 0
NP W 1 0.003 0
Farm C 0.017 0.016 0
Painted dragon (Ctenophorus pictus)
Farm W NP W 1 NP W 2
Farm W 0
NP W 1 0.051 0
NP W 2 0.054 0.005 0
Table 2 Genomic diversity
estimates for populations of
three lizard species (shrubland
morethia, robust ctenotus, and
painted dragon) across isolated
farm fragments and national
park sites in western Victoria,
Australia
Presented as mean values across all SNPs; n = number of samples, HO = observed heterozygosity,
HE = expected heterozygosity, standard error (SE) is reported in brackets with “ < ” representing SE < 0.01
Location n HOHEInbreeding coefficient Private alleles
Shrubland morethia (Morethia obscura)
Farm W 22 0.082 ( <) 0.086 ( <) 0.065 ( <) 1179.97 (2.86)
NP W 1 17 0.081 ( <) 0.085 ( <) 0.069 ( <) 1137.89 (2.51)
NP W 2 7 0.082 ( <) 0.081 ( <) 0.057 ( <) 1098.10 (2.12)
Farm C 19 0.082 ( <) 0.085 ( <) 0.065 ( <) 1161.88 (9.83)
NP C 1 8 0.081 ( <) 0.080 ( <) 0.056 ( <) 1062.99 (2.53)
NP C 2 16 0.082 ( <) 0.084 ( <) 0.060 ( <) 1093.39 (2.96)
Robust ctenotus (Ctenotus robustus)
Farm W 28 0.229 ( <) 0.234 ( <) 0.033 ( <) 1237.40 (5.16)
NP W 1 24 0.231 ( <) 0.235 ( <) 0.032 ( <) 1246.91 (6.47)
Farm C 9 0.229 ( <) 0.226 ( <) 0.061 ( <) 1422.71 (4.74)
Painted dragon (Ctenophorus pictus)
Farm W 14 0.123 ( <) 0.168 ( <) 0.324 (0.01) 588.30 (2.89)
NP W 1 20 0.124 ( <) 0.189 ( <) 0.374 (0.02) 825.01 (3.29)
NP W 2 11 0.125 ( <) 0.183 ( <) 0.377 (0.01) 835.44 (3.31)
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Conservation Genetics
variation (TableS12). The “Farm_W” population of the
painted dragon had a significantly lower expected heterozy-
gosity (HE = 0.168) compared to the two national park popu-
lations (HE = 0.183 & 0.189; TableS12). In accordance with
the ratio of observed to expected heterozygosity values for
the painted dragon, the inbreeding coefficients in all painted
dragon populations were much higher than in populations of
the other two species (Table2).
Genomic diversity was lower at farm populations com-
pared to national park populations for robust ctenotus and
painted dragon, but not for shrubland morethia (Table2).
Inbreeding coefficients were mostly similar between farm
and national park populations (Fig.3), although one farm
population (Farm_C) of the robust ctenotus showed sub-
stantially higher inbreeding, whilst the Farm_W population
of the painted dragon suggested less inbreeding (again in
agreement with the ratio of observed to expected heterozy-
gosity values for that population; Table2). Private alleles
were also notably higher for the robust ctenotus at Farm_C
and lower for the painted dragon at Farm_W (Table2). For
the shrubland morethia, private allele values were mostly
similar, but the highest values were recorded at the two farm
populations (Table2).
Correlation betweengenomic diversity andbody condition
Correlation tests showed no correlation between multi-
locus heterozygosity (MLH) and scaled mass index (SMI)
or between inbreeding coefficient and SMI (TableS10).
Similarly, modelling revealed no effect of MLH, inbreeding
coefficient, or site on SMI (TableS11).
Relatedness
Mean relatedness values were close to zero
(−0.0049–0.0178; TableS7, Fig.4) for shrubland morethia
and robust ctenotus, indicating that average relatedness of
individuals within individual sampling sites was similar to
mean relatedness of individuals across the entire landscape.
For painted dragons, site relatedness means were more var-
ied (−0.0282–0.1028; TableS7, Fig.4). Modelling revealed
a strong positive effect of land use (farm) on relatedness for
painted dragons (Coefficient ± 85% CI = 0.10 [0.08–0.12],
R2 = 0.31; TableS8). For shrubland morethia, an oppo-
site but smaller effect was detected (Coefficient ± 85%
CI = −0.01 [−0.02–−0.01], R2 = 0.02; TableS8) although
this model had low explanatory power (R2 = 0.024).
Land use had no effect on relatedness for robust ctenotus
(TableS8).
Local effective population size
For all three species, effective population size (Ne) was lower
at farm populations compared to national park populations
(except for shrubland morethia at NP_C_1 compared to
Farm_W; Table3). Shrubland morethia had the largest effec-
tive population sizes, with four out of six estimates exceed-
ing 500 (Table3). Estimates for the other two species were
relatively low (Ne < 400), particularly at farm populations
(Ne < 130; Table3).
Discussion
Land modification affects most landscapes on earth, isolating
small populations and inhibiting gene flow for some species.
Given the ubiquity of modified landscapes, biodiversity con-
servation must extend into such systems, identifying driv-
ers of diversity and designing incentive schemes to improve
on-ground management. Here, genomic approaches can
inform conservation decisions. We compared the genomic
structure, diversity, inbreeding, and effective population
Fig. 3 A Individual inbreeding coefficients, and B mean inbreed-
ing coefficients and 95% CIs for populations of shrubland morethia
(Morethia obscura), robust ctenotus (Ctenotus robustus), and painted
dragon (Ctenophorus pictus) sampled across western Victoria, Aus-
tralia. Samples from agricultural habitat fragments are shown with
black fill (A) or black diagonal bars (B) whilst samples from the
national park are depicted by open colours
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Conservation Genetics
sizes of populations of three Australian lizard species in
agricultural fragments to nearby populations in the Little
Desert NationalPark. While genomic differentiation among
sampling locations was typically very low, results for one
species, the painted dragon, suggested genetic isolation due
to the agricultural matrix. Genomic diversity and effective
population sizes were lower at farm populations compared
to national park populations for all three species (except for
genomic diversity of shrubland morethia). Inbreeding coef-
ficients were mostly similar between farm and national park
populations, although one farm population of robust ctenotus
showed notably higher inbreeding, whereas the farm popula-
tion of painted dragon showed lower inbreeding compared to
national park sites. These results highlight the genetic con-
sequences of land clearing such as low population size, low
genomic diversity and higher risk of inbreeding depression.
Fig. 4 Mean pairwise relatedness and 95% confidence intervals at farm fragment and national park sampling sites for: A shrubland morethia
(Morethia obscura), B robust ctenotus (Ctenotus robustus), and C painted dragon (Ctenophorus pictus) in western Victoria, Australia
Table 3 Local estimates of effective population size (Ne) with 95% parametric confidence intervals for three species of lizard in western Victo-
ria, Australia
Overall r2 provides a composite measure of linkage-disequilibrium across all pairs of loci. Sampling range refers to the maximum linear distance
between samples taken from the same population
Population Sampling
range (km)
Shrubland morethia (Morethia obscura) Robust ctenotus (Ctenotus robustus) Painted dragon (Ctenophorus pictus)
nNeOverall r2nNeOverall r2nNeOverall r2
Farm W 0.35 22 1415.2 (1072.4–2077.7) 0.053 28 129.9 (127.0–133.2) 0.042 14 30.9 (30.4–31.4) 0.106
NP W 1 2.60 17 4100.8 (1817.2–Inf) 0.071 24 387.5 (358.7–421.2) 0.048 20 163.3 (155.7–171.6) 0.067
NP W 2 0.55 7 1950.8 (592.6–Inf) 0.223 – – – 11 303.5 (252.4–380.2) 0.135
Farm C 0.31 19 209.6 (198.0–222.5) 0.064 9 32.7 (31.8–33.5) 0.159 – – –
NP C 1 0.24 8 263.2 (209.9–352.0) 0.187 – – – – – –
NP C 2 3.73 16 4665.8 (1831.4–Inf) 0.076 – – – – – –
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Conservation Genetics
Despite these challenges, habitat fragments can maintain
high biodiversity value, which can be maximised by man-
agement initiatives such as translocations and establishing
habitat corridors.
Population genomic structure
Genomic analysis revealed population structuring for the
painted dragon between the fragment and national park
populations, but not between the two national park popula-
tions, despite the former pair being geographically closer
than the latter pair. This indicates that the barrier created
by land clearing has a larger impact on population genomic
structure than geographic distance. Our results suggest that,
for painted dragons, natural gene flow between the farm
fragment and national park is very restricted, likely due to
the highly modified agricultural matrix inhibiting dispersal
(Driscoll etal. 2013). Similarly, Levy etal. (2010) observed
reduced gene flow in an agricultural landscape for a closely
related species, Ctenophorus ornatus. The monoculture of
grass, grazed low to the ground, in pasture paddocks greatly
contrasts vegetation structure in remnant habitat leading to
reduced likelihood of dispersal (Driscoll etal. 2013). Addi-
tionally, grazing paddocks offer few food or shelter resources
for small lizards (Driscoll 2004), and altered abiotic condi-
tions rendering small lizards more prone to desiccation (Tuff
etal. 2016). Thus, grazing paddocks likely restrict gene flow
for lizard meta-populations.
Compared to the painted dragon, we observed less
population structure among localities for the robust cten-
otus and shrubland morethia. Other studies have reported
low genomic differentiation between recently fragmented
(< 100years) populations, highlighting the difficulties of
distinguishing between historical connectivity and cur-
rent migration as a cause of observed genetic similarity
among populations (Sumner etal. 2004; Richmond etal.
2009; Virens etal. 2015). In some cases, gene flow may be
inhibited by the agricultural matrix, but other factors can
prevent the detection of isolation through population genom-
ics. Firstly, high local abundance can create ‘genetic inertia’
which may buffer the genetic effects (e.g., drift) of fragmen-
tation that are more readily detectable in small populations
(Richmond etal. 2009; Virens etal. 2015). This hypothesis
aligns with our results, as species abundance was negatively
correlated with observed genomic differentiation—the
shrubland morethia was the most abundant species with the
least genomic differentiation, and the painted dragon was the
least abundant species with the most genomic differentia-
tion (Westaway etal. 2024b). Secondly, species with shorter
generation times tend to exhibit more rapid genetic differ-
entiation (Sumner etal. 2004). The painted dragon is a pre-
dominantly ‘annual’ species with almost all animals reach-
ing sexual maturity in their first year (Olsson etal. 2007),
whilst robust ctenotus may not reach maturity until their
second year (Coventry 1996). No data exists on the gen-
eration time of shrubland morethia, but the closely related
Morethia boulengeri, reaches sexual maturity in either their
first or second year (Henle 1989). These demographic and
life-history factors, combined with the relatively recent land
clearing at study sites (50–60years since clearing occurred),
could contribute to a time-lag in the genetic response to
human disturbance that varies across species (Richmond
etal. 2009).
Alternatively, the population structure observed may
reflect true levels of gene flow across the agricultural matrix,
with varied levels of dispersal across the three focal species.
This explanation seems less likely as it would imply more
regular dispersal for shrubland morethia, which has a narrow
habitat breadth (Westaway etal. 2024b) and is often absent
in agricultural fragments (Driscoll 2004, Simpson etal.
2023, but see Westaway etal. 2024b), than robust ctenotus
or painted dragon, both of which occur in disturbed, structur-
ally simple habitats, as well as intact heath and mallee veg-
etation (Read 2002; Shea 2010; Michael etal. 2011). Thus,
robust ctenotus and painted dragon are arguably more likely
to disperse across the highly modified grazing pastures in
our study area than shrubland morethia, which does not align
with the population structuring reported here. Therefore, we
suggest that demographic and life-history factors, rather than
dispersal, may be responsible for the observed population
genomic structure patterns, which are incongruous with our
hypothesis of highest structuring in shrubland morethia.
Genomic diversity
Heterozygosity varied across the three species, with the
shrubland morethia and robust ctenotus approximating
Hardy–Weinberg equilibrium (i.e., HE ~ HO), whilst a het-
erozygote deficit was observed in all populations of the
painted dragon. This heterozygote deficit may be a result
of complex breeding behaviour of painted dragons, violat-
ing the Hardy–Weinberg Principle’s assumption of random
breeding (Graffelman etal. 2017). Male painted dragons
actively patrol and defend territories, with dominant males
largely monopolizing access to females within their home
range (Olsson etal. 2007). Additionally, polymorphism in
head and gular ‘bib’ colouration influences the outcome of
male–male contests, female mate choice, and sperm perfor-
mance (Healey etal. 2007; Olsson etal. 2009; McDiarmid
etal. 2017). Thus, for this species, strong sexual selection
may reduce genomic diversity if a subset of males with cer-
tain traits are more likely to successfully reproduce. Such
paternity skew has been observed in a wide range of spe-
cies, especially those with complex social structures such
as primates (Engelhardt etal. 2017) and social insects
(Loope etal. 2014). This can be problematic when trying
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Conservation Genetics
to maintain genetic diversity in small, isolated populations
and can undermine genetic rescue attempts (Manning etal.
2022).
Land clearing often results in reduced genomic diver-
sity, especially when fragments are small and disconnected
(Frankham etal. 2017), and has been reported for verte-
brate populations in habitat fragments (Levy etal. 2010;
Lino etal. 2019), as well as true islands (Frankham 1997;
von Takach etal. 2022; Crates etal. 2024). Our results show
some evidence of this pattern with lower genomic diversity
in farm fragment populations compared to national park
populations for the painted dragon and robust ctenotus. The
two farm populations with notably lower genomic diver-
sity (Farm_C for robust ctenotus and Farm_W for painted
dragon) contain the lowest effective population sizes, sug-
gesting that diversity will erode more quickly at those sites
(Frankham etal. 2014). Interestingly, these two populations
also differ in other metrics but in opposite directions, with
the former population showing the highest levels of inbreed-
ing and number of private alleles compared to other popula-
tions of this species, whilst the latter population shows the
lowest levels of inbreeding and number of private alleles
for that species. For the robust ctenotus, higher inbreeding
in the fragment follows theoretical expectations of small,
isolated populations. The higher number of private alleles in
this population may be explained by an isolation-by-distance
effect as this population was geographically distant from
the other two. For the painted dragon, lower diversity and
private alleles in the fragment may be a result of a bottleneck
event associated with land clearing (Hellborg etal. 2002).
Lower inbreeding in this population is more unexpected,
but may stem from an ‘Allee effect’ causing a breakdown
in sociality (Brashares etal. 2010). Here, the lower popula-
tion density in the fragment (Westaway etal. 2024b) may
reduce the ability of male painted dragons to monopolise
multiple females because fewer females occur within a given
male home range. Therefore, breeding becomes more ran-
dom which could reduce heterozygote deficit and inbreed-
ing in this population relative to larger populations existing
at higher densities. Future study using a greater number of
pairwise replicates would be useful to identify the key driv-
ers of these results.
Relatedness
Mean pairwise relatedness values were generally lower in
farm compared to national park populations for shrubland
morethia and robust ctenotus, potentially contributing to
farm populations showing more loosely clustered population
structure. Conversely, painted dragons in the fragment had
a mean pairwise relatedness estimate more than four times
higher than either national park site. This stark contrast may
be a result of land clearing preventing emigration from and
immigration to the isolated farm fragment. For dispersal-
limited species, such as these small lizards, even relatively
small sections of cleared agricultural land (< 500m) can
act as dispersal barriers, disrupting patterns of sex-biased
dispersal and altering the spatial organisation of related
individuals in remnant fragments (Sumner 2005; Hoehn
etal. 2007; Munguia-Vega etal. 2013). Increasing related-
ness in isolated populations may lead to further declines in
genomic diversity and increased risk of inbreeding depres-
sion (Frankham etal. 2017). Here, genetic rescue via trans-
locations may be necessary in the short-term (Weise etal.
2020), whilst revegetation of habitat corridors can increase
connectivity with neighbouring populations in the longer-
term (Michael etal. 2018).
Local effective population size (Ne)
Effective population size estimates were almost always
higher in national park populations compared to farm frag-
ments. In some cases, this may reflect the spatial extent of
sampling, which was larger for some national park popula-
tions than farm fragments. Nevertheless, low Ne in fragments
aligns with other work in fragmented populations (Sumner
etal. 2004; Lettoof etal. 2021; Senior etal. 2022). Sev-
eral characteristics of habitat fragments may lead to spe-
cies decline including: altered disturbance regimes reducing
habitat quality (Driscoll 2004), edge effects (Didham etal.
2007), and reduced dispersal interrupting meta-population
dynamics (Fischer and Lindenmayer 2007). Low effective
population sizes highlight the need for conservation inter-
ventions, and are especially useful for species with low
detectability, where traditional population metrics are dif-
ficult to ascertain (Stock etal. 2023; von Takach etal. 2023).
We identified two out of five farm fragment populations
under the Ne ≥ 100 threshold recommended to avoid inbreed-
ing depression over the next five generations, and a further
two populations under the Ne ≥ 1000 threshold recom-
mended to maintain evolutionary potential (Frankham etal.
2014). These populations show signs of genetic degrada-
tion including lower genomic diversity, increased inbreeding
and pairwise relatedness. Low population size also causes
increased likelihood of local extinction through stochastic
events (Frankham etal. 2017). Ongoing habitat destruction
across the region (Maron and Fitzsimons 2007) is likely to
further fragment populations, increasing the risk of such
extinction events, while simultaneously reducing overall
population size and population sizes in fragments. Translo-
cations to identified populations with low Ne is one way to
boost effective population size and restore genomic health
(Watson and Watson 2015). Maintaining large Ne has been
identified as a key conservation concern in the Convention
on Biological Diversity Kunming-Montreal Global Biodi-
versity Framework (CBD 2022), which aims to monitor the
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Conservation Genetics
proportion of populations with Ne < 500 across many species
and at a national scale.
Study limitations
The extent of sampling was relatively low overall, includ-
ing 12 populations across three species. In particular, robust
ctenotus was only sampled at two farm populations and one
national park population, and painted dragon sampled at one
farm population and two national park populations. This was
because these species were absent from other locations sam-
pled in the study area. As a result, comparison of species-
specific responses may be confounded by varied sampling.
Incomplete sampling and small samples sizes are common
limitations of genetic studies, particularly for cryptic spe-
cies like small lizards, due to the practical difficulties and
financial cost of collecting and sequencing genetic samples
across entire study areas (Driscoll and Hardy 2005; Senior
etal. 2022). Nevertheless, the limited sampling breadth in
this study should be considered when making inferences
from results presented here. Future studies should aim to
increase the number of habitat fragments, and paired areas
of continuous habitat, sampled to increase inferential power.
Management implications
Habitat fragments can provide important habitat for fauna
but may require management interventions to reduce genetic
risk. Low population size, low genomic diversity, and higher
risk of inbreeding depression were observed to varying
degrees in our study landscape but could be addressed by
management practices. In the short term, wildlife restoration
via regular translocations (sensu Watson and Watson 2015)
may be one means of reducing genetic risk. Translocating
individuals from surrounding areas acts as a surrogate for
natural dispersal, which is inhibited for many species in
modified landscapes (Doherty and Driscoll 2018; Hansen
etal. 2020), and can augment population size and genomic
diversity. Experimental translocations of lizards into farm
fragments in our study landscape had promising short-term
results with translocated individuals showing similar sur-
vival, body condition, and movement patterns compared to
resident animals (Westaway etal. 2024c), although ongoing
monitoring is required to assess longer-term objectives. We
recommend similar translocations be performed in an exper-
imental manner for a range of species across fragmented
urban and agricultural landscapes.
Over the longer term, re-vegetation of habitat corridors
has the potential to restore natural dispersal into fragmented
landscapes, potentially alleviating genomic issues of small,
isolated populations. Additionally, innovative agri-environ-
ment schemes such as directional sowing of crops between
habitat patches (Kay etal. 2016), and inclusion of palatable
woody shrubs and native grasses in grazing pastures (Col-
lard and Fisher 2010), could function to ‘soften’ the agri-
cultural matrix, further enhancing landscape connectivity.
Finally, protection of existing habitat fragments, including
linear strips and stepping stones that offer some connectivity,
is vital to retain whatever gene flow remains in fragmented
landscapes. While small fragments play a crucial role in sup-
porting fauna, proactive management strategies are essential
to mitigate genetic risks and ensure the long-term persis-
tence of populations in fragmented landscapes.
Supplementary Information The online version contains supplemen-
tary material available at https:// doi. org/ 10. 1007/ s10592- 025- 01686-2.
Acknowledgements We pay respects to the Wotjobaluk people, tradi-
tional custodians of the land on which this research was conducted. We
thank Bim Nash, Mick Shoulders and their families for allowing field-
work to be undertaken on private property. We thank Owen Lishmund,
Mirinda Thorpe, Iestyn Hosking and Amy Simpson for assistance with
fieldwork. We thank Claire McLean andJaclyn Harris fortheir guid-
ance and for extracting DNA from tissue samples for shrubland more-
thia.We are grateful for the support of the Gulbali Graduation Bridging
Scheme which made this work possible.
Author Contributions DMW: Conceptualization, data collection, for-
mal analysis, writing—original draft. DGN: Conceptualization, writ-
ing—review & editing, project administration. CJJ: Conceptualization,
writing—review & editing. DRM: Writing—review & editing. DMW:
Writing—review & editing. BvT: Conceptualization, formal analysis,
writing—review & editing.
Funding Open Access funding enabled and organized by CAUL and
its Member Institutions. Funding was provided by the Department of
Energy, Environment and Climate Action through a Victorian State
Government—Biodiversity Bushfire Response and Recovery Grant.
Data availability All data and scripts have been uploaded to the Dryad
Digital Repository (https:// doi. org/https:// doi. org/ 10. 5061/ dryad. s7h44
j1h4).
Declarations
Competing interests The authors have no relevant financial or non-
financial interests to disclose.
Open Access This article is licensed under a Creative Commons Attri-
bution 4.0 International License, which permits use, sharing, adapta-
tion, distribution and reproduction in any medium or format, as long
as you give appropriate credit to the original author(s) and the source,
provide a link to the Creative Commons licence, and indicate if changes
were made. The images or other third party material in this article are
included in the article’s Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in
the article’s Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will
need to obtain permission directly from the copyright holder. To view a
copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Conservation Genetics
References
Attard C, Sandoval-Castillo J, Lang A, Vernazzani B, TorreS L, Bald-
wiN R, JenneR K, Gill P, Burton C, Barceló A (2024) Global con-
servation genomics of blue whales calls into question subspecies
taxonomy and refines knowledge of population structure. Anim
Conserv 27:626–638. https:// doi. org/ 10. 1111/ acv. 12935
Bates D, Maechler M, Bolker B, Walker S (2015) Fitting linear mixed-
effects models using lme4. J Stat Softw 67:1–48
Bergstrom DM, WieneckE BC, Van Den Hoff J, Hughes L, Linden-
mayer DB, Ainsworth TD, Baker CM, BlanD L, Bowman DM,
Brooks ST (2021) Combating ecosystem collapse from the tropics
to the Antarctic. Glob Change Biol 27:1692–1703
Brashares JS, Werner JR, Sinclair A (2010) Social ‘meltdown’in the
demise of an island endemic: Allee effects and the Vancouver
Island marmot. J Anim Ecol 79:965–973
CBD (2022) Kunming-Montreal Global Biodiversity Framework. UN
Convention on Biological Diversity, Montreal, Canada
Collard SJ, Fisher AM (2010) Shrub-based plantings of woody
perennial vegetation in temperate Australian agricultural land-
scapes: What benefits for native biodiversity? Ecol Manag
Restor 11:31–35
Coventry A (1996) The herpetofauna of the Chinaman well area of
the big desert, Victoria. Proc R Soc Vic 108:107–120
Crates R, Von Takach B, Young CM, Stojanovic D, Neaves L,
Murphy L, Gautschi D, Hogg CJ, Heinsohn R, Bell P (2024)
Genomic insights into the critically endangered King Island
scrubtit. J Hered. https:// doi. org/ 10. 1093/ jhered/ esae0 29
Didham RK, Tylianakis JM, Gemmell NJ, RaND TA, Ewers RM
(2007) Interactive effects of habitat modification and spe-
cies invasion on native species decline. Trends Ecol Evol
22:489–496
Do C, WapleS RS, Peel D, Macbeth G, Tillett BJ, Ovenden JR (2014)
NeEstimator v2: re-implementation of software for the estima-
tion of contemporary effective population size (Ne) from genetic
data. Mol Ecol Resour 14:209–214
Doherty TS, Driscoll DA (2018) Coupling movement and landscape
ecology for animal conservation in production landscapes. Proc
R Soc B: Biol Sci 285:20172272
Driscoll DA (2004) Extinction and outbreaks accompany fragmenta-
tion of a reptile community. Ecol Appl 14:220–240
Driscoll DA, Hardy CM (2005) Dispersal and phylogeography of the
agamid lizard Amphibolurus nobbi in fragmented and continu-
ous habitat. Mol Ecol 14(6):1613–1629
Driscoll DA, Banks SC, Barton PS, Lindenmayer DB, Smith AL
(2013) Conceptual domain of the matrix in fragmented land-
scapes. Trends Ecol Evol 28:605–613
Engelhardt A, Muniz L, Perwitasari-Farajallah D, Widdig A (2017)
Highly polymorphic microsatellite markers for the assessment
of male reproductive skew and genetic variation in critically
endangered crested macaques (Macaca nigra). Int J Primatol
38:672–691
Fischer J, Lindenmayer DB (2007) Landscape modification and habi-
tat fragmentation: a synthesis. Glob Ecol Biogeogr 16:265–280
Frankham R (1997) Do island populations have less genetic variation
than mainland populations? Heredity 78:311–327
Frankham R, Ballou SEJD, Briscoe DA, Ballou JD (2002) Introduc-
tion to conservation genetics. Cambridge University Press
Frankham R, Bradshaw CJ, Brook BW (2014) Genetics in conser-
vation management: revised recommendations for the 50/500
rules, Red List criteria and population viability analyses. Biol
Cons 170:56–63
Frankham R, Ballou JD, Ralls K, Eldridge MDB, Dudash MR, Fen-
ster CB, Lacy RC, Sunnucks P (2017) Genetic management of
fragmented animal and plant populations. Oxford University
Press, Oxford
García-Muñoz E, Ceacero F, Pedrajas L, KaliontzopouloU A, Car-
retero MÁ (2011) Tail tip removal for tissue sampling has no
short-term effects on microhabitat selection by Podarcis boc-
agei, but induced autotomy does. Acta Herpetologica 6:223–227
Goudet J (2005) Hierfstat, a package for R to compute and test hier-
archical F-statistics. Mol Ecol Notes 5:184–186
Graffelman J, Jain D, Weir B (2017) A genome-wide study of Hardy-
Weinberg equilibrium with next generation sequence data. Hum
Genet 136:727–741
Gruber B, Unmack PJ, Berry OF, Georges A (2018) dartr: an r
package to facilitate analysis of SNP data generated from
reduced representation genome sequencing. Mol Ecol Resour
18:691–699
Hansen N, Driscoll D, Michael D, Lindenmayer D (2020) Movement
patterns of an arboreal gecko in fragmented agricultural land-
scapes reveal matrix avoidance. Anim Conserv 23:48–59
Healey M, Uller T, Olsson M (2007) Seeing red: morph-specific con-
test success and survival rates in a colour-polymorphic agamid
lizard. Anim Behav 74:337–341
Hellborg L, Walker CW, Rueness EK, Stacy JE, Kojola I, Valdmann
H, Vilà C, Zimmermann B, Jakobsen KS, Ellegren H (2002) Dif-
ferentiation and levels of genetic variation in northern European
lynx (Lynx lynx) populations revealed by microsatellites and mito-
chondrial DNA analysis. Conserv Genet 3:97–111
Henle K (1989) Population ecology and life history of the diurnal skink
Morethia boulengeri in arid Australia. Oecologia 78:521–532
Hoehn M, Sarre S, Henle K (2007) The tales of two geckos: does
dispersal prevent extinction in recently fragmented populations?
Mol Ecol 16:3299–3312
Hogg JT, Forbes SH, Steele BM, Luikart G (2006) Genetic rescue of
an insular population of large mammals. Proc R Soc B: Biol Sci
273:1491–1499
Hohenlohe PA, Funk WC, Rajora OP (2021) Population genomics for
wildlife conservation and management. Mol Ecol 30:62–82
Hooke RL, Martín Duque JF, Pedraza Gilsanz JD (2013) Land trans-
formation by humans: a review. Ene 12:43
Hostetler JA, Onorato DP, Jansen D, OlI MK (2013) A cat’s tale:
the impact of genetic restoration on Florida panther population
dynamics and persistence. J Anim Ecol 82:608–620
Jaureguiberry P, Titeux N, Wiemers M, Bowler DE, Coscieme L,
Golden AS, Guerra CA, Jacob U, Takahashi Y, Settele J (2022)
The direct drivers of recent global anthropogenic biodiversity loss.
Sci Adv 8:eabm9982
Jenkins DG, Brescacin CR, Duxbury CV, Elliott JA, Evans JA, Gra-
blow KR, Hillegass M, LyON BN, MetzgeR GA, Olandese ML
(2007) Does size matter for dispersal distance? Glob Ecol Bio-
geogr 16:415–425
Johnstone CP, Lill A, Reina RD (2014) Habitat loss, fragmentation and
degradation effects on small mammals: analysis with conditional
inference tree statistical modelling. Biol Cons 176:80–98
Kay GM, Driscoll DA, Lindenmayer DB, Pulsford SA, Mortelliti
A (2016) Pasture height and crop direction influence reptile
movement in an agricultural matrix. Agric Ecosyst Environ
235:164–171
Kilian, A., Wenzl, P., Huttner, E., CaRLING, J., Xia, L., Blois, H.,
CaiG, V., Heller-Uszynska, K., Jaccoud, D. & Hopper, C. 2012.
Diversity arrays technology: a generic genome profiling technol-
ogy on open platforms. Data production and analysis in popula-
tion genomics: Methods and protocols, 67–89.
Lettoof DC, Thomson VA, Cornelis J, Bateman PW, Aubret F, Gagnon
MM, Von Takach B (2021) Bioindicator snake shows genomic
signatures of natural and anthropogenic barriers to gene flow.
PLoS ONE 16:e0259124
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Conservation Genetics
Levy E, Kennington JW, Tomkins JL, Lebas NR (2010) Land clearing
reduces gene flow in the granite outcrop-dwelling lizard, Cteno-
phorus ornatus. Mol Ecol 19:4192–4203
Lindenmayer D (2019) Small patches make critical contributions to
biodiversity conservation. Proc Natl Acad Sci 116:717–719
Lino A, Fonseca C, Rojas D, Fischer E, Pereira MJR (2019) A meta-
analysis of the effects of habitat loss and fragmentation on genetic
diversity in mammals. Mamm Biol 94:69–76
Loope KJ, Chien C, Juhl M (2014) Colony size is linked to paternity
frequency and paternity skew in yellowjacket wasps and hornets.
BMC Evol Biol 14:1–12
Luikart G, Ryman N, Tallmon DA, Schwartz MK, Allendorf FW
(2010) Estimation of census and effective population sizes: the
increasing usefulness of DNA-based approaches. Conserv Genet
11:355–373
Macnally R, Bennett AF (1997) Species-specific predictions of the
impact of habitat fragmentation: local extinction of birds in the
box-ironbark forests of central Victoria, Australia. Biol Cons
82:147–155
Madsen T, Shine R (2002) Short and chubby or long and slim? Food
intake, growth and body condition in free-ranging pythons. Aus-
tral Ecol 27:672–680
Madsen T, Shine R, Olsson M, Wittzell H (1999) Restoration of an
inbred adder population. Nature 402:34–35
Manning TP, Austin JJ, Moseby KE, Jensen MA (2022) Skewed pater-
nity impacts genetic diversity in a small reintroduced population
of western quolls (Dasyurus geoffroii). Aust Mammal 45:199–209
Mantel N (1967) The detection of disease clustering and a generalized
regression approach. Can Res 27:209–220
Maron M, Fitzsimons JA (2007) Agricultural intensification and loss of
matrix habitat over 23 years in the West Wimmera, south-eastern
Australia. Biol Cons 135:587–593
Mcdiarmid C, Friesen CR, Ballen C, Olsson M (2017) Sexual col-
oration and sperm performance in the Australian painted dragon
lizard, Ctenophorus pictus. J Evol Biol 30:1303–1312
Michael DR, Cunningham RB, Lindenmayer DB (2011) Regrowth
and revegetation in temperate Australia presents a conservation
challenge for reptile fauna in agricultural landscapes. Biol Cons
144:407–415
Michael DR, Crane M, Florance D, Lindenmayer DB (2018) Reveg-
etation, restoration and reptiles in rural landscapes: Insights
from long-term monitoring programmes in the temperate euca-
lypt woodlands of south-eastern Australia. Ecol Manag Restor
19:32–38
Munguia-Vega A, Rodriguez-Estrella R, Shaw WW, Culver M (2013)
Localized extinction of an arboreal desert lizard caused by habitat
fragmentation. Biol Cons 157:11–20
Olsson M, Healey M, Wapstra E, Schwartz T, Lebas N, Uller T (2007)
Mating system variation and morph fluctuations in a polymorphic
lizard. Mol Ecol 16:5307–5315
Olsson M, Schwartz T, Uller T, Healey M (2009) Effects of sperm stor-
age and male colour on probability of paternity in a polychromatic
lizard. Anim Behav 77:419–424
Paplinska JZ, Taggart DA, Corrigan T, Eldridge MD, Austin JJ (2011)
Using DNA from museum specimens to preserve the integrity of
evolutionarily significant unit boundaries in threatened species.
Biol Cons 144:290–297
Peig J, Green AJ (2009) New perspectives for estimating body condi-
tion from mass/length data: the scaled mass index as an alternative
method. Oikos 118:1883–1891
R Core Team (2024) R: A language and environment for statistical
computing. R Foundation for Statistical Computing, Vienna, Aus-
tria. URL: https:// www.R- proje ct. org
Read JL (2002) Experimental trial of Australian arid zone reptiles as
early warning indicators of overgrazing by cattle. Austral Ecol
27:55–66
Reed DH, Frankham R (2003) Correlation between fitness and genetic
diversity. Conserv Biol 17:230–237
Richmond JQ, Reid DT, Ashton KG, Zamudio KR (2009) Delayed
genetic effects of habitat fragmentation on the ecologically spe-
cialized Florida sand skink (Plestiodon reynoldsi). Conserv Genet
10:1281–1297
Riva F, FahriG L (2022) The disproportionately high value of small
patches for biodiversity conservation. Conserv Lett 15:e12881
Robertson P, Coventry AJ (2019) Reptiles of Victoria: a guide to iden-
tification and ecology. CSIRO Publishing, Clayton
Robledo-Ruiz DA, Austin L, Amos JN, Castrejón-Figueroa J, Harley
DK, Magrath MJ, Sunnucks P, Pavlova A (2023) Easy-to-use R
functions to separate reduced-representation genomic datasets into
sex-linked and autosomal loci, and conduct sex assignment. Mol
Ecol Resour. https:// doi. org/ 10. 1111/ 1755- 0998. 13844
Sansaloni C, Petroli C, Jaccoud D, Carling J, Detering F, Grattapa-
glia D, Kilian A (2011) Diversity Arrays Technology (DArT)
and next-generation sequencing combined: genome-wide,
high throughput, highly informative genotyping for molecular
breeding of Eucalyptus. BMC Proc. https:// doi. org/ 10. 1186/
1753- 6561-5- S7- P54
Senior AF, Clemann N, Gardner MG, Harrisson KA, WhilE GM,
Chapple DG (2022) Genetic structure, diversity and distri-
bution of a threatened lizard affected by widespread habi-
tat fragmentation. Conserv Genet. https:// doi. org/ 10. 1007/
s10592- 021- 01408-4
Shea GM (2010) The suburban terrestrial reptile fauna of Sydney-
winners and losers. In: Lunney D, Hutchings P, Hochuli D (eds)
The natural history of Sydney. Royal Zoological Society of New
South Wales, Sydney, pp 154–197
Simpson R, Nimmo D, Wright L, Wassens S, Michael D (2023)
Decline in semi-arid reptile occurrence following habitat loss
and fragmentation. Wildlife Res. https:// doi. org/ 10. 1071/
WR230 34
Stock SE, Klop-Toker K, WallacE S, Kelly O, Callen A, Seeto R, Mah-
ony SV, Hayward MW, Mahony MJ (2023) Uncovering inbreed-
ing, small populations, and strong genetic isolation in an Austral-
ian threatened frog, Litoria littlejohni. Conserv Genet 24:575–588
Stoffel MA, Esser M, KaRDOS M, Humble E, Nichols H, David
P, Hoffman JI (2016) inbreedR: an R package for the analysis
of inbreeding based on genetic markers. Methods Ecol Evol
7:1331–1339
Sumner J (2005) Decreased relatedness between male prickly forest
skinks (Gnypetoscincus queenslandiae) in habitat fragments. Con-
serv Genet 6:333–340
Sumner J, Jessop T, Paetkau D, Moritz C (2004) Limited effect of
anthropogenic habitat fragmentation on molecular diversity in
a rain forest skink, Gnypetoscincus queenslandiae. Mol Ecol
13:259–269
Templeton AR, Shaw K, Routman E, Davis SK (1990) The genetic
consequences of habitat fragmentation. Ann Missouri Bot Gard.
https:// doi. org/ 10. 2307/ 23996 21
Tuff K, Tuff T, Davies K (2016) A framework for integrating thermal
biology into fragmentation research. Ecol Lett 19:361–374
Virens E, Krauss SL, Davis RA, Spencer PB (2015) Weak genetic
structuring suggests historically high genetic connectivity among
recently fragmented urban populations of the scincid lizard, Cten-
otus fallens. Aust J Zool 63:279–286
Von Takach B, Ranjard L, Burridge CP, Cameron SF, Cremona T,
Eldridge MD, Fisher DO, Frankenberg S, Hill BM, Hohnen R
(2022) Population genomics of a predatory mammal reveals pat-
terns of decline and impacts of exposure to toxic toads. Mol Ecol
31:5468–5486
Von Takach B, Sargent H, Penton CE, Rick K, Murphy BP, NeavE
G, DaVIES HF, Hill BM, BankS SC (2023) Population genom-
ics and conservation management of the threatened black-footed
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Conservation Genetics
tree-rat (Mesembriomys gouldii) in northern Australia. Heredity
130:278–288
Von Takach B, Cameron SF, Cremona T, Eldridge MD, Fisher DO,
Hohnen R, Jolly CJ, Kelly E, Phillips BL, Radford IJ (2024) Con-
servation prioritisation of genomic diversity to inform manage-
ment of a declining mammal species. Biol Cons 291:110467
Waples RS (2024) Practical application of the linkage disequilibrium
method for estimating contemporary effective population size: a
review. Mol Ecol Resour 24:e13879
Watson DM, Watson MJ (2015) Wildlife restoration: Mainstream-
ing translocations to keep common species common. Biol Cons
191:830–838
Weatherhead PJ, Knox JM, Harvey DS, Wynn D, ChiucchI J, GibbS
HL (2009) Diet of Sistrurus catenatus in Ontario and Ohio: effects
of body size and habitat. J Herpetol 43:693–697
Weeks AR, Heinze D, Perrin L, Stoklosa J, Hoffmann AA, Van Rooyen
A, Kelly T, Mansergh I (2017) Genetic rescue increases fitness
and aids rapid recovery of an endangered marsupial population.
Nat Commun 8:1071
Weise EM, Sard NM, Nehfer A, Scribner KT (2020) Adaptive genetic
management: assessing the benefits of translocations. Conserv
Genet 21:277–287
Westaway DM, Jolly CJ, Michael DR, Nimmo DG (2024a) Factors
affecting microhabitat use in two agamid species from south-
eastern Australia. Austral Ecol 49:e13567
Westaway DM, Jolly CJ, Watson DM, Jessop TS, MichaeL DR, Lin-
ley GD, Aristova A, Holmes B, Price JN, Ritchie EG, Nimmo
DG (2024b) Fragments maintain similar herpetofauna and small
mammal richness and diversity to continuous habitat, but com-
munity composition and traits differ. Landscape Ecol 39:1–21
Westaway DM, Jolly CJ, Watson DM, Watson MJ, Michael DR, Linley
GD, Holmes B, Ritchie EG, Buchan A, LoeffleR E, Nimmo DG
(2024c) Wildlife restoration in fragmented landscapes: trialling
wild-to-wild translocation with two common reptiles. Biol Cons
299:110780
Whiteley AR, Fitzpatrick SW, Funk WC, Tallmon DA (2015) Genetic
rescue to the rescue. Trends Ecol Evol 30:42–49
Wilson S, Swan G (2021) A complete guide to reptiles of Australia,
6th edn. Reed New Holland, Sydney, p 688
Wintle BA, Kujala H, Whitehead A, Cameron A, Veloz S, Kukkala A,
Moilanen A, Gordon A, Lentini PE, Cadenhead NC (2019) Global
synthesis of conservation studies reveals the importance of small
habitat patches for biodiversity. Proc Natl Acad Sci 116:909–914
Wright A, Anson J, Leo V, Wright B, Newsome T, Grueber C (2022)
Urban restoration of common species: population genetics of rein-
troduced native bush rats Rattus fuscipes in Sydney, Australia.
Anim Conserv 25:825–836
Zheng X, Levine D, Shen J, Gogarten SM, Laurie C, Weir BS (2012) A
high-performance computing toolset for relatedness and principal
component analysis of SNP data. Bioinformatics 28:3326–3328
Publisher's Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com