Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
PRIMARY RESEARCH ARTICLE
Human land use promotes the abundance and diversity of
exotic species on Caribbean islands
Wendy A. M. Jesse
1
|
Jocelyn E. Behm
1,2
|
Matthew R. Helmus
2
|
Jacintha Ellers
1
1
Department of Ecological Science –Animal
Ecology, Vrije Universiteit Amsterdam,
Amsterdam, the Netherlands
2
Integrative Ecology Lab, Center for
Biodiversity, Department of Biology,
Temple University, Philadelphia,
Pennsylvania
Correspondence
Wendy A. M. Jesse, Department of
Ecological Science –Animal Ecology, Vrije
Universiteit Amsterdam, 1081 HV
Amsterdam, the Netherlands.
Email: w.a.m.jesse@vu.nl
Funding information
Koninklijke Nederlandse Akademie van
Wetenschappen, Grant/Award Number:
UPS/375/Eco/J1515; Nederlandse
Organisatie voor Wetenschappelijk
Onderzoek, Grant/Award Number:
858.14.041
Abstract
Human land use causes major changes in species abundance and composition, yet
native and exotic species can exhibit different responses to land use change. Native
populations generally decline in human-impacted habitats while exotic species often
benefit. In this study, we assessed the effects of human land use on exotic and
native reptile diversity, including functional diversity, which relates to the range of
habitat use strategies in biotic communities. We surveyed 114 reptile communities
from localities that varied in habitat structure and human impact level on two Carib-
bean islands, and calculated species richness, overall abundance, and evenness for
every plot. Functional diversity indices were calculated using published trait data,
which enabled us to detect signs of trait filtering associated with impacted habitats.
Our results show that environmental variation among sampling plots was explained
by two Principal Component Analysis (PCA) ordination axes related to habitat struc-
ture (i.e., forest or nonforest) and human impact level (i.e., addition of man-made
constructions such as roads and buildings). Several diversity indices were signifi-
cantly correlated with the two PCA axes, but exotic and native species showed
opposing responses. Native species reached the highest abundance in forests, while
exotic species were absent in this habitat. Human impact was associated with an
increase in exotic abundance and species richness, while native species showed no
significant associations. Functional diversity was highest in nonforested environ-
ments on both islands, and further increased on St. Martin with the establishment
of functionally unique exotic species in nonforested habitat. Habitat structure, rather
than human impact, proved to be an important agent for environmental filtering of
traits, causing divergent functional trait values across forested and nonforested envi-
ronments. Our results illustrate the importance of considering various elements of
land use when studying its impact on species diversity and the establishment and
spread of exotic species.
KEYWORDS
environmental filtering, functional trait diversity, habitat structure, Lesser Antilles, native and
exotic species, reptiles, squamates, tropics, urbanization
----------------------------------------------------------------------------------------------------------------------------------------------------------------------
This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any
medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
© 2018 The Authors. Global Change Biology published by John Wiley & Sons Ltd.
Received: 30 November 2017
|
Revised: 19 March 2018
|
Accepted: 7 May 2018
DOI: 10.1111/gcb.14334
Glob Change Biol. 2018;1–13. wileyonlinelibrary.com/journal/gcb
|
1
1
|
INTRODUCTION
Human activities are altering aquatic and terrestrial ecosystems
across the planet through a multitude of processes including pollu-
tion (Bunzel, Kattwinkel, & Liess, 2013), climate change (Walther,
2010), introduction of exotic species (Vitousek, D’Antonio, Loope, &
Westbrooks, 1996), and land use change (Foley, 2005). The latter
causes natural ecosystems to undergo rapid and dramatic shifts in
temperature, humidity and light regimes, and to become character-
ized by non-native and managed vegetation, impervious surfaces,
and man-made structures (Pickett et al., 2001). Hence, human-im-
pacted environments apply novel selection pressures and environ-
mental filters on the organisms that inhabit them (Shochat, Warren,
Faeth, McIntyre, & Hope, 2006).
It is widely recognized that human land conversion directly
affects species diversity, but the direction and magnitude of these
effects differ across taxa and ecosystems (Saari et al., 2016; Simons
et al., 2017). For example, richness of vertebrates is usually higher in
areas of low human impact, whereas plant richness is highest in
intermediately impacted sites (McKinney, 2008). Moreover, native
and exotic species can differ in their responses to land use change
(Hansen et al., 2005; McKinney, 2006). Land conversion has been
associated with native species declines, potentially because those
often specialized species lack suitable traits to cope with rapid envi-
ronmental change (McKinney, 2002; Shea & Chesson, 2002). In con-
trast, exotic species can benefit from land use change, as the points
of entry for exotic species are often located in human-inhabited
areas (Kowarik, 2011; McKinney, 2006), and exotic species may ben-
efit disproportionally from the niche opportunities in novel, anthro-
pogenic habitats compared to locally adapted native species (Sax &
Brown, 2000). In their novel range, exotic species can be released
from biotic interactions, while native species may be subject to sub-
stantial numbers of coevolved competitors, predators, parasites, and
diseases (Shea & Chesson, 2002). Although exotic establishment
causes a small, local contribution to species richness, unknown inter-
actions between native and exotic species can either lead to an
increase (Traveset & Richardson, 2014) or decrease in overall species
richness over time (Jauni & Ramula, 2015; Sugiura, 2016).
To better understand if exotic species follow the rules of local
community assembly, it is important to consider changes in functional
diversity: the suite of organismal traits in a given community (D
ıaz
et al., 2007). When we consider human land use as an environmental
filter that only allows species with suitable traits to persist in anthro-
pogenic habitats, we expect species that do not possess such traits to
go extinct over time (Aronson et al., 2016). For instance, populations
of several thermo-sensitive neotropical lizard species are at risk of
extirpation as a result of increasing temperatures in human-modified
habitats, possibly owing to physiological limits placed on survival,
activity, and foraging efficiency (Nowakowski et al., 2018). Alterna-
tively, preadapted exotic species with favorable traits for survival in
anthropogenic habitats may invade, such as highly fecund reptile spe-
cies with short maturation times (Van Wilgen & Richardson, 2012) and
broad environmental tolerances (Mahoney et al., 2015). These
changes in trait value and frequency either decrease functional diver-
sity of a community due to trait filtering (e.g., Flynn et al., 2009), or
increase it when species with novel, more suitable traits enter the sys-
tem (e.g., Whittaker et al., 2014). Studying trait diversity can thus pro-
vide additional insight into the processes of exotic establishment and
species decline in human-impacted environments.
Here, we investigate effects of human land use on native and
exotic species abundance and diversity of the terrestrial reptile com-
munities on two neighboring, Caribbean islands. Island biota are par-
ticularly sensitive to habitat change because of the restricted area,
the often limited species pool, and the relatively high number of
endemic species on islands (Paulay, 1994; Sugiura, 2016). Further-
more, under scenarios of intensifying human activities, islands are
expected reach higher human population densities, undergo rela-
tively more land use change (Kier et al., 2009; Venter et al., 2016),
and experience higher rates of species introductions compared to
mainland areas (Van Kleunen et al., 2015). Many Caribbean islands
have undergone extensive land conversion since Western coloniza-
tion, but it is largely unknown how human land use has affected
within-island native and exotic species distributions, and how this is
related to functional diversity of native and exotic species assem-
blages (Henderson & Powell, 2001).
We surveyed insular reptile communities across various human-
impacted and nonimpacted sites to assess how different aspects of
land use (i.e., habitat structural change and the addition of man-made
substrates and constructions) affect reptile abundances and diversity,
and specifically test for differences between exotic and native species
responses. We expected that, similar to findings in other taxa, exotic
reptiles would be predominantly associated with human-impacted
environments, and taxonomic diversity would be higher in human-im-
pacted habitats due to local species establishment. We expected
native species richness and population sizes to decline with human
land use. Furthermore, we explored the occurrence of environmental
filtering of functional traits as a result of human land use, and effects
thereof on overall functional diversity of reptile communities. We
expected the dominant trait values within species communities to shift
in response to the altered thermal and physical conditions in human-
impacted environments. Lastly, we anticipated that overall functional
trait diversity would decrease if sensitive native species were lost, or
increase if species with novel, unique traits had invaded the system.
2
|
MATERIALS AND METHODS
2.1
|
Study area and reptile community
The Caribbean islands of St. Eustatius (21 km
2
) and St. Martin
(87 km
2
) (Figure 1a,b) are part of the Northern Lesser Antilles. On
both islands, vegetation consists predominantly of xeric shrub and
woodland, with a smaller area of tropical rainforest (De Freitas,
Rojer, Nijhof, & Debrot, 2012). Saint Martin has a human population
size of approximately 75,000 permanent residents and receives
roughly 2.5 million visitors per year (CIA, 2016a, 2016b; Department
of Statistics Sint Maarten, 2015). A large proportion of the island
2
|
JESSE ET AL.
area is used for housing and roads (i.e., developed area of approxi-
mately 60% estimated from OpenStreetMap contributors, 2017; Fig-
ure 1b). St. Eustatius’population size is approximately 3,200 people
with 12,000 annual visitors (CBS, 2016), and its developed area is
currently around 30% (Figure 1a). Only a small proportion of the
islands are used for agriculture (approximately 5% and 3% for St.
Eustatius and St. Martin, respectively; estimated from OpenStreet-
Map contributors, 2017). All reptile species that we encountered in
our survey belong to the order of Squamata (snakes and lizards). The
currently assumed extant squamate community (of both native and
exotic origin) consists of 12 and 11 species on St. Martin and St.
Eustatius, respectively (Figure 3a,b) (Powell & Henderson, 2012;
Powell, Henderson, & Parmerlee, 2015).
2.2
|
Habitat categorization
Prior to sampling, we defined three dominant habitat categories
based on dominant vegetation type and habitat structure: Forest,
Shrub, and Urban. Forests were characterized by tall primary or sec-
ondary growth vegetation (mostly trees of
x= 9.27 m, s= 1.00 m)
with stems of at least 10 cm in diameter and a high percentage of
canopy cover (
x= 78.92%, s= 1.75%). Shrub areas were dry natural
environments dominated by xeric multistem shrubs (a diameter
≤10 cm), a relatively high percentage of understory cover
(
x= 31.79%, s= 2.15%), and a relatively low percentage of canopy
cover compared to forest areas (
x= 38.46%, s= 3.19%). Urban habi-
tats were located in or near towns and characterized by man-made
impervious substrates (e.g., asphalt and concrete), buildings, land-
scaping, and ornamental tree and shrub species. To ensure an equal
sampling effort across different levels of human impact, these three
habitat categories were subdivided into seven distinct habitat types
according to their exposure to human disturbance. First, Forest and
Shrub plots were classified based on the presence or absence of
human impact, resulting in natural plots (Forest Natural and Shrub
Natural) devoid of any impervious surfaces and man-made construc-
tions, and impacted plots (Forest Impact and Shrub Impact) with
approximately 25% of the plot surface developed by humans. Sec-
ond, Urban habitat plots were classified according to the level and
type of human impact, resulting in Urban Built Sites with at least
60% of the surface area covered by impervious surfaces, Urban
Developed Gardens with approximately 25% of the surface area cov-
ered by impervious surfaces and characterized by ornamental,
S
Legend
Canopy cover (%) Sampled habitat Developed area
Roads and buildings
1 - 25
26 - 75
76 - 85
86 - 100
Urban Built Sites
Urban Mixed Vegetation
Urban Developed Gardens
Forest Impact
Forest Natural
Shrub Impact
Shrub Natural
(a) (b)
FIGURE 1 Distribution of 114 sample sites on St. Eustatius (a) and St. Martin (b). The map includes a layer of remotely sensed estimates of
canopy cover
1
, a composite layer of developed area
2
, and a symbolized layer of seven sampled habitat types. Symbols have a similar shape to
symbols depicted in Figure 2 and thus represent the same habitat types.
1
Source: Hansen/UMD/Google/USGS/NASA; Hansen et al., 2013.
2
Source: ©OpenStreetMap contributors, 2017; openstreetmap.org
JESSE ET AL.
|
3
managed vegetation, and Urban Mixed Vegetation with a dominant
cover of unmanaged ruderal plants due to earlier severe human dis-
turbance (e.g., dumping of rubble or clearing vegetation for construc-
tion). In terms of human impact we considered Forest Natural and
Shrub Natural plots least impacted followed by Forest Impact and
Shrub Impact, and further increasing toward Urban Mixed Vegeta-
tion, Urban Developed Gardens, and Urban Built Sites. Multiple plots
of each habitat type were sampled on both islands (see Supporting
Information S1 for the exact number of replicates).
2.3
|
Sampling methods
A total of 114 plots were sampled on St. Eustatius (n= 71) and on St.
Martin (n= 43) between 2 and 31 July 2015 and 1 August and 28
August 2015, respectively. Plots of 80 m
2
and varying dimensions
(10 98 m unless there were extenuating circumstances) were
selected based on dominant vegetation type and level of human
impact, and assigned to one of seven predefined habitat types (see
Section 2.2 above). Plots were separated by at least 100 m to ensure
that communities represent independent samples, which is reasonable
given that all surveyed species likely disperse <100 m/day (Calsbeek,
2009; Daltry, McCauley, & Morton, 2002; Kelehear, Brown, & Shine,
2013; Knapp, Alvarez-Clare, & Perez-Heydrich, 2010), and many rep-
tiles in our survey are territorial, often with home ranges of far <1 hec-
tare (Moura, Cavalcanti, Leite-Filho, Mesquita, & McConkey, 2015;
Powell et al., 2015). Plots were surveyed between 8.30 a.m. and
18.00 p.m. for a constant search time of 80 person-minutes (following
a standard reptile surveying technique from McDiarmid, 2012). In the
first half of the search period, the observers recorded highly mobile
species by walking parallel to each other along the length of the plot.
In the second half of the search, the observers would turn around and
look for more sedentary species under rocks, tree bark, and through
the litter layer. In order to limit the chance of repeated observations,
mobile species were not recorded during the second half of the search
with the exception of new species to that specific plot and any other
animals that were clearly missed before. No animals were handled or
hurt in the process of surveying, and we strived to minimize distur-
bance to animals during surveying following the guidelines of the
National Centre for the Replacement, Refinement and Reduction of
Animals in Research (NC2RS, 2017). Subsequent to the search period,
the following environmental data were collected in each plot: temper-
ature (°C) at five perch locations of detected reptiles (e.g., tree bark,
bare soil, in litter layer, etc.), leaf litter depth (cm) at four locations, and
estimates of canopy cover (%), maximum vegetation height (m), and
understory cover (%) within the plot surface area. Human impact was
described in detail and summarized in terms of the impact of roads,
upright structures, debris, and the use of irrigation (see Supporting
Information S1 for detailed descriptions).
2.4
|
Taxonomic and functional diversity variables
We calculated total abundance of all animals, species richness, and
Pielou’s evenness index for each plot using the vegan R package
(Oksanen et al., 2016). In addition, we calculated native and exotic
species richness and abundance for every plot. The observed species
richness did not differ from two measures of rarefied species rich-
ness based on Chao’s and Hurlbert’s rarefaction curves (implemented
in the vegan package; see Supporting Information S1), so we used
observed richness as dependent variable in our statistical analyses.
Evenness was calculated for all plots with a minimum of three
recorded individuals and/or a minimum of two recorded species
(n= 91; 35 Urban, 27 Shrub, and 29 Forest plots). This is the mini-
mum number of observations per plot that would allow analyzing
evenness as a continuous variable. For communities with one or two
individuals distributed over just one species, the outcome would be
either fixed (even community in all cases) or dichotomous (either
even or uneven), resulting in a bias in evenness values for small com-
munities. We did not calculate evenness for native and exotic spe-
cies separately because exotic species richness and abundance were
often too low to calculate evenness values with.
We obtained functional trait values from the literature for all
observed species (see Supporting Information S2 for data and refer-
ences) focusing on traits associated with habitat use, resource acqui-
sition and climatic tolerance, as we expected such traits to be
differentially associated with various levels of land use change.
Included traits were: mean snout-vent length (SVL; cm), head length
(cm), preferred temperature (°C), night activity (index from diurnal to
completely nocturnal), sun perching (i.e., the likelihood of finding the
species perched in sun-lit conditions, expressed as an index from
exclusively shaded to exclusively sunny perches), omnivory (pres-
ence/absence of omnivorous lifestyle), and sexual size dimorphism
(SSD; mean ♂/mean ♀SVL). Sexual size dimorphism represents
male-biased dimorphism at values >1 and larger bodied females at
values <1. Head length values were corrected for mean SVL using
linear regression and taking the residuals for further analysis.
For every surveyed plot, we calculated functional dispersion
(Fdis) as a measure of functional diversity by including all functional
traits of the observed community into a distance-based functional
diversity function (FD package; Lalibert
e, Legendre, & Shipley, 2014).
Our databases included variables of different types and units, and
were therefore scaled using Gower’s standardization for mixed vari-
ables (Gower, 1971), ensuring that variables contributed equally to
Fdis. Functional diversity was weighted according to relative abun-
dance of species within each plot and calculated for the full commu-
nity (i.e., all individuals found in a plot) and for native species
separately (i.e., Fdis of only the native species in a plot), but not for
exotic species separately. Often only a single exotic species was
detected in a sample plot, resulting in an exotic functional diversity
of zero. This result is unrealistic, as the exclusion of exotic species
to calculate native functional diversity clearly changed Fdis-values.
We also calculated Community-Weighted Mean trait values
(CWM) per plot by weighing the individual traits (e.g., snout-vent
length) according to species abundances within the community. Such
a CWM reflects the dominant trait state among all individuals within
a given community (i.e., plot). Different CWM values along environ-
mental gradients, therefore, reflect environmental filtering of
4
|
JESSE ET AL.
functional traits toward favorable levels for that environment (Boukili
& Chazdon, 2017).
2.5
|
Statistical analyses
The nine environmental variables we recorded at each site (see Sec-
tion 2.3) were scaled and summarized by a Principal Component Anal-
ysis (PCA). We included both the mean and variance of litter depth
and temperature as separate variables, bringing the number of vari-
ables in the PCA to 11 (see Supporting Information S1 for a detailed
description of all variables). Due to a few incomplete surveys we
missed important values for vegetation height and temperature.
Therefore, seven values of vegetation height, and one temperature
mean and temperature variance value were calculated through non-
parametric imputation (implemented in the missForest package), filling
in the missing values from (non)linear relationships with the other
environmental variables (Stekhoven & B€
uhlmann, 2012). All 114 sites
could thereafter be included in the PCA. The first two principal com-
ponents were selected for further analyses based on their high inter-
pretability and clear association with human land use (Table 1; see
Supporting Information S1 for loadings of PCA axes 3–11).
Least-squares (linear) regression models for all diversity and trait-
based indices were fitted with PC1, PC2, and Island as independent
variables. Initially we also included the interactions between Island
and PC1 and Island and PC2 in the models. None of the models
showed significant interactions assuming a= 0.05 (results not
shown), therefore, interactions were excluded from the model. In all
cases, the fit of the reduced model including only the main factors
PC1, PC2, and Island was equivalent to that of the full model. Statis-
tical assumptions for normality and independence of model residuals
were checked and dependent variables were square root or power
transformed when necessary. The analyses with exotic species abun-
dance and exotic species richness as dependent variables were ana-
lyzed with a generalized linear model with poisson error distribution,
which proved sufficient in managing the large number of zero-obser-
vations in the exotic datasets. To ensure independence of model
residuals, we tested for spatial autocorrelation with a Moran’s I test
(spdep package; Bivand, Hauke, & Kossowski, 2013). No spatial auto-
correlation of model residuals was detected for any of our models,
verifying that interplot distances do not bias the value of the
response variables. As response variables were not independent
from each other (e.g., native and exotic abundances together form
total abundance) and tested repeatedly against the same indepen-
dent factors, all modeled p-values were adjusted by controlling for
the false discovery rate (Benjamini & Hochberg, 1995) (see S1 for a
detailed description of our methods). All analyses and plotting proce-
dures were executed in R 3.2.5 (R core team, 2016).
3
|
RESULTS
The first two Principal Components (PC1 and PC2) together
explained 46% of the environmental variation among 114 samples
(Table 1, Figure 2a,b). Both PC1 and PC2 differed significantly
between the sampled habitat types (MANOVA: df = 6, approx.
F= 25.51, p< 0.0001, Figure 2b). PC1 (29% of environmental varia-
tion) was positively associated with litter depth (mean and variance),
canopy cover, and vegetation height, and negatively associated with
temperature and understory cover (Table 1, Figure 2a). Therefore,
PC1 distinguished between forested (FI and FN) and nonforested
habitats (UBS, UDG, UMV, SN, and SI) with densely forested plots
scoring high on PC1 and more open vegetation having low PC1
scores (Figure 2b). Shrub and Urban habitat types had comparably
low PC1-scores (ANOVA: df =1, F= 0.77, p= 0.38). The second
principal component (PC2; 17% of variation) was correlated with
human-associated factors, including irrigation practices and the addi-
tion of man-made structures, debris, and roads. PC2 is interpreted to
represent human impact level, ranging from natural environments at
low PC2 scores to human-impacted habitats at high PC2 scores.
Hence, using these PC-axes as independent variables in the subse-
quent analyses enabled us to separate the effects of the addition of
man-made substrates and associated activities (i.e., irrigation, traffic)
from habitat structural change, which can also be human induced.
None of the eleven environmental variables that were included
in the PCA differed significantly between islands (MANOVA: df =1,
approx. F= 0.94, p= 0.51). We did detect a minor significant effect
of island on PC1 scores (df =1,F= 4.13, p= 0.04), which likely is
due to a few primary forest plots on St. Eustatius with exceptionally
high levels of canopy cover, vegetation height and litter depth in
comparison to forests on St. Martin (Supporting Information S1).
Human impact levels expressed as PC2 were comparable within the
sample plots on both islands (df =1,F= 2.07, p= 0.15).
3.1
|
General observations from surveys
The species pools on St. Eustatius and St. Martin are functionally
and phylogenetically similar (Figure 3a,b). Several species occur on
both islands and all native species are congeners. Furthermore, all of
TABLE 1 The loadings of the first two component axes from a
principal component analysis on the environmental variation among
114 sample sites on two Caribbean islands
Variable PC1 PC2
Roads 0.514 0.532
Upright structures 0.149 0.466
Debris 0.260 0.341
Irrigation 0.110 0.292
Vegetation height 0.343 0.119
Litter depth (variance) 0.211 0.104
Litter depth (mean) 0.403 0.090
Canopy cover 0.479 0.020
Understory cover 0.221 0.261
Temperature (variance) 0.298 0.345
Temperature (mean) 0.428 0.279
Explained variation 29.2%16.8%
Cumulative 29.2%46.0%
JESSE ET AL.
|
5
the exotic species on St. Eustatius have also established on St. Martin,
although St. Martin has had three additional exotic species establish
viable populations. A further difference between species pools is that
three native species are considered extirpated from St. Martin, com-
pared to zero species on St. Eustatius. On both islands the largest pro-
portion of observed animals in our surveys were species within the
genus Anolis, a group of medium-sized tree-dwelling lizards (mean
SVL: 42.8–53.9 mm), which made up 79% of the 1,453 observed rep-
tiles. Only a small proportion of observed animals were exotic (St. Eus-
tatius (0.8%) and St. Martin (6.2%) of all observed animals).
3.2
|
Responses of exotic and native reptiles to
human impact and habitat structure
Total (native + exotic) abundance of individuals per plot was posi-
tively correlated with both PC1 (0.28 0.07, t= 4.37, adjusted
p< 0.0001) and PC2 (0.20 0.08, t= 2.31, adj. p= 0.03), with
highest abundances in plots with forest vegetation and plots with
additional man-made substrates and structures (Figure 4). This is
consistent with the fact that “Forest Impact”plots had the highest
mean total abundance of all habitat types (
x= 19.73, s= 14.37; Sup-
porting Information S1). However, native and exotic abundances
were differentially affected by the two environmental axes (Fig-
ure 4). Exotic abundances were negatively correlated with PC1 (i.e.,
habitat structure) to the point that no exotic species were detected
in any of the forest plots on either island (0.30 0.14, t=2.07,
adj. p= 0.04), while native abundances were positively influenced by
this axis (0.30 0.07, t= 4.61, adj. p< 0.0001). Unlike native abun-
dances (0.16 0.09, t= 1.87, adj. p= 0.06), exotic species abun-
dances were significantly associated with PC2 (i.e., human impact)
(0.63 0.14, t= 4.52, adj. p< 0.0001). The interactions between
island and PCs were not statistically significant in any of the models
(results not shown), indicating that the effects of PC1 and PC2 on
reptile diversity were consistent across islands. Total abundances
and native abundances were similar across islands (0.28 0.24,
t= 1.16, adj. p= 0.37 and 0.17 0.24, t= 0.73, adj. p= 0.47,
–4
–2
0
2
−2 0 2 4
PC1 (29.2%)
PC2 (16.8%)
Habitat types
Urban Mixed Vegetation (UMV)
Urban Build Sites (UBS)
Urban Developed Gardens (UDG)
Forest Impact (FI)
Forest Natural (FN)
Shrub Impact (SI)
Shrub Natural (SN)
Roads
Upright Structures
Debris
Irrigation
Temperature
mean Temperature
variation
Understory cover
Vegetation height
Canopy cover
Litter mean
Litter variation
nonforested forested
natural
impact
−2
−1
0
1
2
3
PC1
–2
–1
0
1
2
PC2
UBS UDG UMV SI SN FI FN
Habitat type
aaaaa
b
b
a
ab
bc
bc
cc
d
(a) (b)
FIGURE 2 Alignment of seven habitat types with the first two component axes of a principal component analysis (PCA) of the
environmental variation among 114 sample plots. Eleven environmental variables measured in every plot on St. Martin and St. Eustatius were
included in PCA (panel a; gray arrows). The first two PCA axes together explained 46% of environmental variation among plots, and separated
forested from nonforested plots along PC1, and natural form impacted plots along PC2 (a). Ellipses depict 95% confidence intervals around the
mean, assuming a normal distribution. Symbols depict plots of seven distinct habitat types and are similar in shape and meaning to the symbols
in Figure 1. PC scores are significantly different across habitat types (a= 0.05) (b). The mean and standard error values of the PC scores are
depicted for every habitat type, and lowercase letters indicate post hoc results
6
|
JESSE ET AL.
respectively), whereas abundances of exotic species were higher on
St. Martin (1.51 0.43, t= 3.50, adj. p= 0.001; Figure 3).
Total (native + exotic) species richness was not significantly asso-
ciated with PC1 (0.01 0.06, t=0.20, adj. p= 0.84) or PC2
(0.14 0.08, t= 1.71, adj. p= 0.13; Figure 4), and did not differ
between islands (0.12 0.22, t= 0.52, adj. p= 0.60). Likewise,
native richness was associated with neither PC1 (0.04 0.05,
t= 0.69, adj. p= 0.74) nor PC2 (0.04 0.07, t= 0.62, adj.
p= 0.54), and was consistent across islands (0.16 0.20,
t=0.82, adj. p= 0.60). Also for exotic species, no statistically sig-
nificant association was detected for PC1 (0.32 0.17, z=1.86,
adj. p= 0.19), however, a borderline significant relationship with
PC2 was discovered for exotic species richness (0.37 0.16,
t= 2.36, adj. p= 0.05). Exotic richness was generally higher on St.
Martin (1.40 0.52, t= 2.69, adj. p= 0.02).
Abundances were more evenly distributed over species in non-
forested environments compared to forested environments, as we
detected a negative association of total (native + exotic) evenness
values with PC1 (0.02 0.01, t=2.82, p= 0.01; Figure 4).
Evenness of species communities was not related to PC2
(0.00 0.01, t=0.01, p= 0.94). Low evenness occurred in
forested habitats where the same native sister species consistently
dominated the communities on both islands (i.e., Anolis schwartzi on
St. Eustatius and A. pogus on St. Martin and dwarf geckos of genus
Sphaerodactylus on both islands).
Functional diversity was negatively correlated with PC1 for total
species (0.43 0.13, t=3.25, adj. p= 0.003) and native-only
analyses (0.34 0.13, t=2.71, adj. p= 0.008), indicating lower
functional diversity in forested habitats. Neither of the two func-
tional diversity estimates was related to PC2 (0.17 0.17, t= 0.97,
ExoticExotic Native
0
5
10
15
20
Species
Count (square root transformed)
Exotic Native
Hem.mab Igu.igu Ind.bra Als.ruf Ano.bim Ano.sch Igu.del Pho.ery Sph.sab Sph.spu The.spp Ano.bimAno.cri Ano.sag Gym.undHem.mab Igu.iguInd.bra Als.rij Ano.ginAno.pog Igu.del Pho.pleSph.par Sph.spu Spo.mar The.spp
0
5
10
15
20
Species
St. Eustatius St. Martin
Not detected
Not detected in plots
Likely extirpated
Not detected in plots
Likely extirpated
Likely extirpated
Likely extirpated
Abbreviation Species name
Als.ruf
Alsophis rufiventris
Als.rij
Alsophis rijgersmaei
Ano.bim
Anolis bimaculatus
Ano.cri
Anolis cristatellus
Ano.gin
Anolis gingivinus
Ano.pog
Anolis pogus
Ano.sag
Anolis sagrei
Ano.sch
Anolis schwartzi
Gym.und
Gymnophthalmus underwoodi
Hem.mab Hemidactylus mabouia
Igu.del Iguana delicatissima
Igu.igu
Iguana iguana
Ind.bra
Indotyphlops braminus
Pho.ery Pholidoscelis erythrocephala
Pho.ple Pholidoscelis plei analifera
Sph.par Sphaerodactylus parvus
Sph.sab Sphaerodactylus sabanus
Sph.spu Sphaerodactylus sputator
Spo.mar Spondylurus martinae
The.spp. Thecadactylus spp.*
(a) (b)
* There is uncertainty about the species status and occurrence of Thecadactylus (The.spp.).
On St. Eustatius only Thecadactylus rapicauda has been described, however on St. Martin the
morphologically similar T. rapicauda and T. oskrobapreinorum could simultaneously occur.
Functional group
Blind snake
Dwarf gecko
Gecko
Ground lizard
Iguana
Skink
Snake
Tegu
Tree lizard
FIGURE 3 Surveyed species counts of all currently known exotic and native species within 114 sample plots on St. Eustatius (a) and St.
Martin (b). Species that were not recorded are either considered extirpated or seen in very low numbers outside plot boundaries. Underlined
species abbreviations indicate that species occur on both islands. Species within the same functional group are generally closely related and
highly similar in terms of size, body plan, behavior and ecology
JESSE ET AL.
|
7
adj. p= 0.67 and 0.04 0.16, t= 0.29, adj. p= 0.78, respectively;
Figure 4). Total species functional diversity was similar on both
islands (St. Eustatius:
x= 24.90, s= 25.92; St. Martin:
x= 20.61,
s= 23.25; 0.72 0.49, t=1.48, adj. p= 0.14), whereas native
functional diversity was significantly lower on St. Martin (
x= 14.44,
s= 11.98) than on St. Eustatius (
x= 24.70, s= 25.98)
(1.22 0.47, t=2.61, adj. p= 0.02). This suggests that exotic
species augment functional diversity of St. Martin’s communities by
introducing novel traits.
3.3
|
Environmental filtering of functional traits
Community-weighted snout-vent length (0.31 0.04, t=7.79,
adj. p<0.0001), preferred temperature (0.21 0.06, t=3.39,
0
2
4
6
8
0
2
4
6
8
0
2
4
6
0
2
4
6
0
5
10
–2 0 2 4
0
5
10
–4 –2 0 2
Abundance (square root)
Species richness
Evenness
Functional diversity (square root)
PC1 (habitat structure) PC2 (human impact)
native + exotic
native
exotic
Legend
P ≤ 0.05
nonforest forest natural impact
0.4
0.6
0.8
1.0
0.4
0.6
0.8
1.0
FIGURE 4 Contrasting responses of reptile diversity indices to habitat structure (PC1) and human impact (PC2) on two Caribbean islands.
The diversity of total (exotic + native; black), native (blue) and exotic (red) species is plotted against PC-axes. Lines indicate relationships that
are statistically significant (a= 0.05). See text and Table 1 for the loadings of PC1 and PC2
8
|
JESSE ET AL.
adj. p= 0.001), sun perching (0.05 0.02, t=3.40, adj.
p= 0.001), head length (0.56 0.12, t=4.60, adj. p< 0.0001),
and sexual size dimorphism (0.02 0.00, t=5.07, adj.
p< 0.0001) were all negatively correlated with PC1 (Figure 5), that
is, communities have higher abundance-weighted trait values in non-
forested environments. The CWM of preferred temperature
(0.84 0.23, t=3.74, adj. p< 0.001), sexual size dimorphism
(0.09 0.02, t=5.96, adj. p< 0.0001), and head length
(2.05 0.46, t=4.51, adj. p< 0.0001) differed between islands
with significantly lower values on St. Martin. Hence, species commu-
nities on St. Martin are characterized by species with smaller heads
than expected from their body size, low preferred temperatures, and
low species-specific sexual size differentiation compared to St. Eus-
tatius. None of the relationships between community-weighted
mean traits and PC2 approached significance (all adjusted p-values
≥0.70; Figure 5).
4
|
DISCUSSION
In this study, habitat structure and human impact independently and
consistently affected the abundance and diversity of reptiles on two
Caribbean islands. However, the direction of these effects differed
between native and exotic species. Exotic individuals occupied pri-
marily impacted habitats and were found exclusively in nonforested
environments, while native species reached extremely high abun-
dances in forests and were not significantly affected by our estimate
of human impact. Nonforested environments harbored communities
with the highest functional diversity, which was further augmented
on St. Martin with the establishment of exotic species. In addition,
habitat structure appeared to be an agent for environmental filtering
of traits, producing patterns of divergent functional trait values
across forested and nonforested environments.
In our study, reptile abundance rather than richness seems to be
sensitive to differences in habitat structure and human impact.
Reports on differential effects of human land use on different diver-
sity indices are not uncommon in the literature (Saari et al., 2016).
Similar results have been found for grasslands (Simons et al., 2017)
and rainforest ecosystems (Laurance et al., 2006), in which species
abundances significantly changed with human impact while species
richness remained stable. Diversity of birds (Blair, 1996), butterflies
(Blair & Launer, 1997), and reptiles (Ackley, Carter, Henderson, Pow-
ell, & Muelleman, 2009; Germaine & Wakeling, 2001) have previ-
ously been reported to peak at intermediate levels of human
activities. However, we detected highest abundances at our most
extreme levels of human impact, and no effect of human impact on
richness or evenness. A reason for this difference could be the rela-
tively low levels of urban development on the Caribbean islands in
this study where fully developed cities are absent. Therefore, the
impact levels on our sampled islands might actually be intermediate
compared to other studies.
As expected, exotic and native species diversity differed in their
response to human impact and habitat structure. The most heavily
developed sites with limited forest cover supported more exotic
individuals and species than nonimpacted sites. Our results are con-
sistent with previous studies that detected associations of exotic
species with human development across various taxa and geographic
areas (Arag
on & Morales, 2003; Blair, 1996; Chace & Walsh, 2006;
Dawson et al., 2017; Qian & Ricklefs, 2006), and imply that to estab-
lish, exotic reptiles exploit not only urban sites but also human-im-
pacted areas in general. The highest native abundances were found
in forested areas where we observed increasing dominance of few
species, that is, extreme abundances of sister species A. schwarzi on
St. Eustatius and A. pogus on St. Martin, accompanied by increasing
abundance of dwarf geckos of the genus Sphaerodactylus on both
islands. Human impact level in itself did not appear to influence
native species diversity, given the absence of a negative relationship
between native diversity and human impact, and the fact that most
observed native species had an island-wide range. This could be sur-
prising, as urban development is generally assumed to be one of the
major causes of population decline and species extinction (Sala et al.,
2000). The fact that native species occur in nonforested and human-
impacted habitats could be due to spillover into less suitable habitats
from source populations in forests (Borges, Ugland, Dinis, & Gaspar,
2008). In addition, native reptiles might be particularly resistant to
human land use. For instance, urban anole populations can even
develop a better body condition than conspecific forest populations
(Chejanovski, Avil
es-Rodr
ıguez, Lapiedra, Preisser, & Kolbe, 2017).
However, given the highly significant positive association of native
abundances with forest habitat, forest-living likely presents optimal
conditions for native species, and it is deforestation that reduces
population numbers rather than purely the addition of man-made
substrates (Shochat et al., 2006).
Functional diversity was relatively low in forested sites compared
to nonforested sites on both islands. This result was unexpected as
tropical forests generally harbor a functionally diverse community
(Ernst, Linsenmair, & R€
odel, 2006), and habitat conversion has
caused functional diversity declines among New World mammals
and birds (Flynn et al., 2009). The most prominent difference
between these studies and ours is that functional loss was associ-
ated with species loss in impacted environments, whereas we found
equal species richness across all habitat structures and human impact
levels. Instead, our results are comparable to outcomes of a study
performed in temperate environments, in which agricultural and
urban land uses enhanced divergence of several functional traits in
birds and plants (Concepci
on et al., 2017). Functional diversity signif-
icantly increased on St. Martin following the introduction of exotic
species, which is similar to results for birds (Sobral, Lees, & Ciancia-
ruso, 2016) and arthropods (Whittaker et al., 2014) on other oceanic
islands.
Several community-weighted mean trait values varied with habi-
tat structure, but not with human impact. Large-bodied, large-
headed, thermo-tolerant and sexually dimorphic species occurred
disproportionately often in nonforested environments, which sug-
gests that habitat structural properties, such as vegetation height,
canopy cover, litter depth, and temperature act as environmental fil-
ter on the functional composition of reptile communities.
JESSE ET AL.
|
9
Particularly, traits that enhance survival in hot and open areas of
potentially variable food availability appeared to be favored in non-
forested environments including shrub and urban habitats. For
instance, body size is positively correlated with desiccation-tolerance
and thermo-regulatory efficiency in reptiles (Dzialowski & O’Connor,
1999; Herczeg, T€
or€
ok, & Kors
os, 2007). Preferred temperature is sig-
nificantly related to heat tolerance (Llewelyn, Macdonald, Hatcher,
Moritz, & Phillips, 2016), so it is likely that species that perform well
at high temperatures can also survive higher temperatures for longer
periods of time. This is a beneficial trait for survival in human-con-
verted habitats, considering nonforested plots were on average 5.22
and 6.71°C hotter than forested plots on St. Eustatius and St. Mar-
tin, respectively, and recent studies have demonstrated that thermal
tolerance largely explains reptile responses to human land use
(Brusch, Taylor, & Whitfield, 2016; Frishkoff, Hadly, & Daily, 2015;
Nowakowski et al., 2018). High sexual size dimorphism has previ-
ously been associated with low levels of intraspecific competition for
food and resources (Latella, Poe, & Giermakowski, 2011), and larger
headed reptiles are able to consume a wider range of prey sizes
(Schoener & Gorman, 1968), suggesting that nonforested sites have
a more variable prey community. Little research has gone into this
topic, but several studies have demonstrated decreasing insect abun-
dances (Blair & Launer, 1997; McIntyre, 2000), changing insect body
size and condition (Venn, Kotze, Lassila, & Niemel€
a, 2013), and
increasing insect body shape variation (Weller & Ganzhorn, 2004) in
shrub-like vegetation and along urban gradients. Interestingly, most
of these reptile traits have previously been associated with exotic
establishment success (Latella et al., 2011), as tolerances to desicca-
tion and food scarcity likely also improve survival of translocation
events. However, trait values of the exotic species on St. Martin and
St. Eustatius were distributed over the entire trait ranges (Supporting
Information S1). Therefore, rather than having distinctive traits in
this study, exotic species might occupy distinct functional niche
space over a range of trait values. This hypothesis is supported by
the increase in functional diversity when exotic species were
included in our analysis.
The results of our analyses were similar on the two islands we
sampled. Although we detected between-island differences for exo-
tic abundance, exotic richness and for native functional diversity,
these were likely caused by regional rather than habitat-specific fac-
tors. The introduction of exotic reptiles is positively associated with
shipping traffic, and St. Eustatius and St. Martin are on opposite
ends of this economic spectrum (Helmus, Mahler, & Losos, 2014).
Therefore, St. Martin likely received a relatively high number of exo-
tic introductions, increasing the probability of exotic establishment
and persistence thereafter. Furthermore, we found that native spe-
cies communities on St. Martin are functionally impoverished com-
pared to St. Eustatius’communities, which is likely due to species
6
8
10
Snout-vent length
(√ cm)
Sun perching (index)
1.0
1.1
1.2
1.3
1.4
Sexual size
dimorphism (♂/♀)
Night activity (index)
0
5
Head length
(residual values)
Omnivory (index)
26
28
30
32
Temperature (°C)
Legend
p ≤ 0.05
–2 4–4 –2
PC1 (habitat structure)
nonforest forest
PC2 (human impact)
natural impact
native + exotic
02 02
−2 4−4 −2
PC1 (habitat structure)
nonforest forest
PC2 (human impact)
natural impact
02 02
2.0
2.1
3.0
3.5
4.0
1.00
1.25
1.50
1.75
2.00
0.00
0.25
0.50
0.75
1.00
FIGURE 5 Effects of habitat structure (PC1) and human impact (PC2) on the community-weighted mean functional traits of 114 reptile
communities on two Caribbean islands. Lines indicate relationships that are statistically significant (a= 0.05). See text and Table 1 for
definition and loadings of PC1 and PC2
10
|
JESSE ET AL.
extirpations from St. Martin. Species of similar size and behavior to
these extirpated species are still present on St. Eustatius, resulting in
a higher native functional diversity on this island.
Our results suggest that exotic and native species favor different
environments and that diversity shifts can be attributed to either habi-
tat structural change or human impact on St. Martin and St. Eustatius.
Native populations seem to suffer from the reduction in forested habi-
tat, rather than the direct impact of additional man-made impervious
surfaces. However, native reptiles are able to persist in nonforested
environments, albeit at lower abundances. The functional composition
of native assemblages in forests is unique, and removing the remaining
forest vegetation on these islands could have severe effects on local
taxonomic and functional community composition. Even though the
proportion of exotic individuals in the communities is low, they occupy
a distinct environment in which they can increase functional diversity
in functionally impoverished communities. Our results could poten-
tially help to predict diversity trajectories under various land use sce-
narios on other oceanic islands.
ACKNOWLEDGEMENTS
We are grateful for the Academy Ecology Fund and the Netherlands
Organization for Scientific Research for their financial support. Many
thanks to Mark Yokoyama, St. Eustatius National Parks Foundation
(STENAPA), Ecological Professionals Foundation, the Caribbean
Netherlands Science Institute (CNSI), the St. Maarten Nature Founda-
tion, and R
eserve Nationale Naturelle de Saint-martin for supplying us
with permits, information, and a helping hand in the field. We thank
Dr. Johanna Wegener and Dr. Luke Mahler for their permission to use
trait data for our analyses. A special thanks goes out to the best field
assistants: Rotem Zilber, Tobia de Scisciolo, and Jasper Bekema, and
to Elizabeth Haber and Maarten Eppinga for their continuous support.
CONFLICT OF INTEREST
No potential conflict of interest was reported by any of the authors
or other contributors to this piece of research.
ORCID
Wendy A. M. Jesse http://orcid.org/0000-0002-2910-8352
REFERENCES
Ackley, J., Carter, R., Henderson, R., Powell, R., & Muelleman, P. (2009).
A rapid assessment of herpetofaunal diversity in variously altered
habitats on Dominica. Applied Herpetology,6, 171–184. https://doi.
org/10.1163/157075408X394124
Arag
on, R., & Morales, J. M. (2003). Species composition and invasions in
NW Argentinian secondary forests, Effects of land use history, envi-
ronment and landscape. Journal of Vegetation Science,14, 195–204.
https://doi.org/10.1111/j.1654-1103.2003.tb02144.x
Aronson, M. F. J., Nilon, C. H., Lepczyk, C. A., Parker, T. S., Warren, P. S.,
Cilliers, S. S., ... Zipperer, W. (2016). Hierarchical filters determine
community assembly of urban species pools. Ecology,97(11), 2952–
2963. https://doi.org/10.1002/ecy.1535
Benjamini, Y., & Hochberg, Y. (1995). Controlling the false discovery rate:
A practical and powerful approach to multiple testing. Journal of the
Royal Statistical Society Series B,57, 289–300.
Bivand, R. S., Hauke, J., & Kossowski, T. (2013). Computing the Jacobian
in Gaussian spatial autoregressive models: An illustrated comparison
of available methods. Geographical Analysis,45(2), 150–179. https://d
oi.org/10.1111/gean.12008
Blair, R. B. (1996). Land use and avian species diversity along an urban
gradient. Ecological Applications,6(2), 506–519. https://doi.org/10.
2307/2269387
Blair, Robert B., & Launer, A. E. (1997). Butterfly diversity and human
land use: Species assemblages along an urban gradient. Biological
Conservation,80, 113–125. https://doi.org/10.1016/S0006-3207(96)
00056-0
Borges, P. A. V., Ugland, K. I., Dinis, F. O., & Gaspar, C. (2008). Insect
and spider rarity in an oceanic island (Terceira, Azores): True rare and
pseudo-rare species. In S. Fattorini (Ed.), Insect ecology and conserva-
tion (pp. 47–70). Thiruvananthapuram, Kerala, India: Research Sign-
post.
Boukili, V. K., & Chazdon, R. L. (2017). Environmental filtering, local site
factors and landscape context drive changes in functional trait com-
position during tropical forest succession. Perspectives in Plant Ecol-
ogy, Evolution and Systematics,24,37–47. https://doi.org/10.1016/j.
ppees.2016.11.003
Brusch, G. A., Taylor, E. N., & Whitfield, S. M. (2016). Turn up the heat:
Thermal tolerances of lizards at La Selva, Costa Rica. Oecologia,180
(2), 325–334. https://doi.org/10.1007/s00442-015-3467-3
Bunzel, K., Kattwinkel, M., & Liess, M. (2013). Effects of organic pollu-
tants from wastewater treatment plants on aquatic invertebrate com-
munities. Water Research,47, 597–606. https://doi.org/10.1016/j.
watres.2012.10.031
Calsbeek, R. (2009). Sex-specific adult dispersal and its selective conse-
quences in the brown anole, Anolis sagrei.Journal of Animal Ecology,
78(3), 617–624. https://doi.org/10.1111/j.1365-2656.2009.01527.x
CBS (Centraal Bureau voor de Statistiek) (2016). Population Caribbean
Netherlands stable. Retrieved from https://www.cbs.nl/en-gb/news/
2016/29/population-caribbean-netherlands-stable
Chace, J. F., & Walsh, J. J. (2006). Urban effects on native avifauna: A
review. Landscape and Urban Planning,74,46–69. https://doi.org/10.
1016/j.landurbplan.2004.08.007
Chejanovski, Z. A., Avil
es-Rodr
ıguez, K. J., Lapiedra, O., Preisser, E. L., &
Kolbe, J. J. (2017). An experimental evaluation of foraging decisions
in urban and natural forest populations of Anolis lizards.Urban Ecosys-
tems,20(5), 1011–1018. https://doi.org/10.1007/s11252-017-0654-
5
Central Intelligence Agency (CIA) (2016a). The World Factbook –Saint
Martin. Retrieved from https://www.cia.gov/library/publications/
resources/the-world-factbook/geos/print_rn.html
Central Intelligence Agency (CIA) (2016b). The World Factbook –Sint
Maarten. Retrieved from https://www.cia.gov/library/publications/
the-world-factbook/geos/print_sk.html
Concepci
on, E. D., G€
otzenberger, L., Nobis, M. P., Bello, F., Obrist, M. K.,
& Moretti, M. (2017). Contrasting trait assembly patterns in plant and
bird communities along environmental and human-induced land-use
gradients. Ecography,40(6), 753–763. https://doi.org/10.1111/ecog.
02121
Daltry, J. C., McCauley, C., & Morton, M. (2002). Re-introduction of the
Antiguan Racer Alsophis antiguae, to Green Island: Interim Report.
Antiguan Racer Conservation Project, St John's, Antigua and Barbuda.
Dawson, W., Moser, D., van Kleunen, M., Kreft, H., Pergl, J., Pysek, P., ...
Dyer, E. E. (2017). Global hotspots and correlates of alien species
richness across taxonomic groups. Nature Ecology and Evolution,1,
0186. https://doi.org/10.1038/s41559-017-0186
JESSE ET AL.
|
11
De Freitas, J. A., Rojer, A. C., Nijhof, B. S. J., & Debrot, A. O. (2012). A
landscape Ecological vegetation map of Sint Eustatius (Lesser Antilles)
(Vol. No. C053/1).
Department of Statistics Sint Maarten (2015). Tourism. Retrieved from
http://stat.gov.sx/
D
ıaz, S., Lavorel, S., de Bello, F., Qu
etier, F., Grigulis, K., & Robson, T. M.
(2007). Incorporating plant functional diversity effects in ecosystem
service assessments. Proceedings of the National Academy of Sciences
of the United States of America,104, 20684–20689. https://doi.org/
10.1073/pnas.0704716104
Dzialowski, E. M., & O'Connor, M. P. (1999). Utility of blood flow to the
appendages in physiological control of heat exchange in reptiles. Jour-
nal of Thermal Biology,24,21–32. https://doi.org/10.1016/S0306-
4565(98)00034-5
Ernst, R., Linsenmair, K. E., & R€
odel, M.-O. (2006). Diversity erosion
beyond the species level: Dramatic loss of functional diversity after
selective logging in two tropical amphibian communities. Biological
Conservation,133, 143–155. https://doi.org/10.1016/j.biocon.2006.
05.028
Flynn, D. F. B., Gogol-Prokurat, M., Nogeire, T., Molinari, N., Richers, B.
T., Lin, B. B., ... DeClerck, F. (2009). Loss of functional diversity
under land use intensification across multiple taxa. Ecology Letters,
12,22–33. https://doi.org/10.1111/j.1461-0248.2008.01255.x
Foley, J. A. (2005). Global consequences of land use. Science,309, 570–
574. https://doi.org/10.1126/science.1111772
Frishkoff, L. O., Hadly, E. A., & Daily, G. C. (2015). Thermal niche predicts
tolerance to habitat conversion in tropical amphibians and reptiles.
Global Change Biology,21(11), 3901–3916. https://doi.org/10.1111/
gcb.13016
Germaine, S. S., & Wakeling, B. F. (2001). Lizard species distributions and
habitat occupation along an urban gradient in Tucson, Arizona, USA.
Biological Conservation,972,29–237.
Gower, J. C. (1971). A general coefficient of similarity and some of its
properties. Biometrics,27, 857–871. https://doi.org/10.2307/
2528823
Hansen, A. J., Knight, R. L., Marzluff, J. M., Powell, S., Brown, K., Gude,
P. H., & Jones, K. (2005). Effects of exurban development on biodi-
versity: Patterns, mechanisms, and research needs. Ecological Applica-
tions,15(6), 1893–1905. https://doi.org/10.1890/05-5221
Hansen, M. C., Potapov, P. V., Moore, R., Hancher, M., Turubanova, S. A.,
Tyukavina, A., ... Townshend, J. R. G. (2013). High-resolution global
maps of 21st-century forest cover change. Science,342, 850–853.
Retrieved from http://earthenginepartners.appspot.com/science-
2013-global-forest
Helmus, M. R., Mahler, D. L., & Losos, J. B. (2014). Island biogeography
of the Anthropocene. Nature,513, 543–546. https://doi.org/10.
1038/nature13739
Henderson, R. W., & Powell, R. (2001). Response by the West Indian her-
petofauna to human-influenced resource. Caribbean Journal of
Science,37,41–54.
Herczeg, G., T€
or€
ok, J., & Kors
os, Z. (2007). Size-dependent heating rates
determine the spatial and temporal distribution of small-bodied
lizards. Amphibia-Reptilia,28, 347–356. https://doi.org/10.1163/
156853807781374674
Jauni, M., & Ramula, S. (2015). Meta-analysis on the effects of exotic
plants on the fitness of native plants. Perspectives in Plant Ecology,
Evolution and Systematics,17, 412–420. https://doi.org/10.1016/j.
ppees.2015.06.002
Kelehear, C., Brown, G. P., & Shine, R. (2013). Invasive parasites in multi-
ple invasive hosts: The arrival of a new host revives a stalled prior
parasite invasion. Oikos,122(9), 1317–1324. https://doi.org/10.
1111/j.1600-0706.2013.00292.x
Kier, G., Kreft, H., Lee, T. M., Jetz, W., Ibisch, P. L., Nowicki, C., ... Barth-
lott, W. (2009). A global assessment of endemism and species rich-
ness across island and mainland regions. Proceedings of the National
Academy of Sciences of the United States of America,106, 9322–9327.
https://doi.org/10.1073/pnas.0810306106
Knapp, C. R., Alvarez-Clare, S., & Perez-Heydrich, C. (2010). The influ-
ence of landscape heterogeneity and dispersal on survival of neonate
insular iguanas. Copeia,2010(1), 62–70. https://doi.org/10.1643/CE-
09-014
Kowarik, I. (2011). Novel urban ecosystems, biodiversity, and conserva-
tion. Environmental Pollution,159(8), 1974–1983. https://doi.org/10.
1016/j.envpol.2011.02.022
Lalibert
e, E., Legendre, P., & Shipley, B. (2014). FD: Measuring functional
diversity from multiple traits, and other tools for functional ecology.
R package version 1.0-12.
Latella, I. M., Poe, S., & Giermakowski, J. T. (2011). Traits associated with
naturalization in Anolis lizards: Comparison of morphological, distribu-
tional, anthropogenic, and phylogenetic models. Biological Invasions,
13, 845–856. https://doi.org/10.1007/s10530-010-9873-x
Laurance, W. F., Nascimento, H. E. M., Laurance, S. G., Andrade, A.,
Ribeiro, J. E. L. S., Giraldo, J. P., ... D'Angelo, S. (2006). Rapid decay
of tree-community composition in Amazonian forest fragments. Pro-
ceedings of the National Academy of Sciences of the United States of
America,103, 19010–19014. https://doi.org/10.1073/pnas.
0609048103
Llewelyn, J., Macdonald, S., Hatcher, A., Moritz, C., & Phillips, B. L.
(2016). Thermoregulatory behavior explains countergradient variation
in the upper thermal limit of a rainforest skink. Oikos,126(5), 748–
757.
Mahoney, P. J., Beard, K. H., Durso, A. M., Tallian, A. G., Long, A. L., Kin-
dermann, R. J., ... Mohn, H. E. (2015). Introduction effort, climate
matching and species traits as predictors of global establishment suc-
cess in non-native reptiles. Diversity and Distributions,21(1), 64–74.
https://doi.org/10.1111/ddi.12240
McDiarmid, R. W. (Ed.) (2012). Reptile biodiversity: Standard methods for
inventory and monitoring. Berkeley, CA: University of California Press.
McIntyre, N. E. (2000). Ecology of urban arthropods: A review and a call
to action. Annals of the Entomological Society of America,93(4), 825–
835. https://doi.org/10.1603/0013-8746(2000)093[0825:EOUAAR]2.
0.CO;2
McKinney, M. L. (2002). Urbanization, biodiversity, and conservation.
BioScience,52(10), 883–890. https://doi.org/10.1641/0006-3568
(2002)052[0883:UBAC]2.0.CO;2
McKinney, M. L. (2006). Urbanization as a major cause of biotic homoge-
nization. Biological Conservation,127, 247–260. https://doi.org/10.
1016/j.biocon.2005.09.005
McKinney, M. L. (2008). Effects of urbanization on species richness: A
review of plants and animals. Urban Ecosystems,11, 161–176.
https://doi.org/10.1007/s11252-007-0045-4
Moura, A. C. de A., Cavalcanti, L., Leite-Filho, E., Mesquita, D. O., &
McConkey, K. R. (2015). Can green iguanas compensate for vanishing
seed dispersers in the Atlantic forest fragments of north-east Brazil?
Journal of Zoology,295(3), 189–196. https://doi.org/10.1111/jzo.
12186
NC2RS –National Center for the Replacement, Refinement & Reduction
of Animals in Research (2017). Retrieved from https://www.nc3rs.
org.uk/wildlife-research
Nowakowski, A. J., Watling, J. I., Thompson, M. E., Brusch, G. A., Cate-
nazzi, A., Whitfield, S. M., ... Todd, B. D. (2018). Thermal biology
mediates responses of amphibians and reptiles to habitat modifica-
tion. Ecology Letters,21, 345–355. https://doi.org/10.1111/ele.12901
Oksanen, J., Blanchet, F.G., Friendly, M., Kindt, R., Legendre, P., McGlinn,
D., ... Wagner, H. (2016). vegan: Community Ecology Package, R
package.
OpenStreetMap contributors (2017). Retrieved from https://www/open
streetmap.org
Paulay, G. (1994). Biodiversity on oceanic islands: Its origin and extinc-
tion. Integrative and Comparative Biology,34, 134–144.
12
|
JESSE ET AL.
Pickett, S. T. A., Cadenasso, M. L., Grove, J. M., Nilon, C. H., Pouyat, R.
V., Zipperer, W. C., & Costanza, R. (2001). Urban ecological systems:
Linking terrestrial ecological, physical, and socioeconomic of
metropolitan areas. Annual Review of Ecology and Systematics,32,
127–157. https://doi.org/10.1146/annurev.ecolsys.32.081501.114
012
Powell, R., & Henderson, R. W., Eds. (2012). Island lists of West Indian
amphibians and reptiles. Florida Museum of Natural History, Univer-
sity of Florida.
Powell, R., Henderson, R. W., & Parmerlee, J. S. (2015). The Reptiles and
Amphibians of the Dutch Caribbean –Saba, St. Eustatius, and St.
Martin. 2nd ed., revised and expanded. Nature Guide Series No. 004.
Dutch Caribbean Nature Alliance, Kralendijk, Bonaire.
Qian, H., & Ricklefs, R. E. (2006). The role of exotic species in homoge-
nizing the North American flora. Ecology Letters,9, 1293–1298.
https://doi.org/10.1111/j.1461-0248.2006.00982.x
R Core Team (2016). R: A language and environment for statistical comput-
ing. Vienna, Austria: R Foundation for Statistical Computing.
Retrieved from https://www.R-project.org/
Saari, S., Richter, S., Higgins, M., Oberhofer, M., Jennings, A., & Faeth, S.
H. (2016). Urbanization is not associated with increased abundance
or decreased richness of terrestrial animals –dissecting the literature
through meta-analysis. Urban Ecosystems,19, 1251–1264. https://d
oi.org/10.1007/s11252-016-0549-x
Sala, O. E., Chapin, F. S. III, Armesto, J. J., Berlow, E., Bloomfield, J.,
Dirzo, R., ... Wall, D. H. (2000). Global biodiversity scenarios for the
year 2100. Science,287, 1770–1774. https://doi.org/10.1126/scie
nce.287.5459.1770
Sax, D. F., & Brown, J. H. (2000). The paradox of invasion. Global Ecology
and Biogeography,9(5), 363–371. https://doi.org/10.1046/j.1365-
2699.2000.00217.x
Schoener, T. W., & Gorman, G. C. (1968). Some niche differences on
three Lesser Antillean lizards of the genus Anolis.Ecology,49, 819–
830. https://doi.org/10.2307/1936533
Shea, K., & Chesson, P. (2002). Community ecology theory as a frame-
work for biological invasions. Trends in Ecology & Evolution,17(4),
170–176. https://doi.org/10.1016/S0169-5347(02)02495-3
Shochat, E., Warren, P. S., Faeth, S. H., McIntyre, N. E., & Hope, D.
(2006). From patterns to emerging processes in mechanistic urban
ecology. Trends in Ecology and Evolution,21, 186–191. https://doi.
org/10.1016/j.tree.2005.11.019
Simons, N. K., Lewinsohn, T., Bl€
uthgen, N., Buscot, F., Boch, S., Daniel,
R., ... Prati, D. (2017). Contrasting effects of grassland management
modes on species-abundance distributions of multiple groups. Agricul-
ture, Ecosystems & Environment,237, 143–153. https://doi.org/10.
1016/j.agee.2016.12.022
Sobral, F. L., Lees, A. C., & Cianciaruso, M. V. (2016). Introductions do
not compensate for functional and phylogenetic losses following
extinctions in insular bird assemblages. Ecology Letters,19(9), 1091–
1100. https://doi.org/10.1111/ele.12646
Stekhoven, D. J., & B€
uhlmann, P. (2012). MissForest—non-parametric
missing value imputation for mixed-type data. Bioinformatics,28(1),
112–118. https://doi.org/10.1093/bioinformatics/btr597
Sugiura, S. (2016). Impacts of introduced species on the biota of an ocea-
nic archipelago: The relative importance of competitive and trophic
interactions. Ecological Research,31, 155–164. https://doi.org/10.
1007/s11284-016-1336-0
Traveset, A., & Richardson, D. M. (2014). Mutualistic interactions and
biological invasions. Annual Review of Ecology, Evolution, and System-
atics,45,89–113. https://doi.org/10.1146/annurev-ecolsys-120213-
091857
Van Kleunen, M., Dawson, W., Essl, F., Pergl, J., Winter, M., Weber, E.,
... Antonova, L. A. (2015). Global exchange and accumulation of
non-native plants. Nature,525(7567), 100–103. https://doi.org/10.
1038/nature14910
Van Wilgen, N. J., & Richardson, D. M. (2012). The roles of climate, phy-
logenetic relatedness, introduction effort, and reproductive traits in
the establishment of non-native reptiles and amphibians. Conservation
Biology,26(2), 267–277. https://doi.org/10.1111/j.1523-1739.2011.
01804.x
Venn, S. J., Kotze, D. J., Lassila, T., & Niemel€
a, J. K. (2013). Urban dry
meadows provide valuable habitat for granivorous and xerophylic
carabid beetles. Journal of Insect Conservation,17(4), 747–764.
https://doi.org/10.1007/s10841-013-9558-8
Venter, O., Sanderson, E. W., Magrach, A., Allan, J. R., Beher, J., Jones, K.
R., ... Levy, M. A. (2016). Sixteen years of change in the global ter-
restrial human footprint and implications for biodiversity conserva-
tion. Nature Communications,7, 12558. https://doi.org/10.1038/nc
omms12558
Vitousek, P., D'Antonio, C., Loope, L., & Westbrooks, R. (1996). Biological
invasions as global environmental change. American Scientist,84,
468–478.
Walther, G. R. (2010). Community and ecosystem responses to recent cli-
mate change. Philosophical Transactions of the Royal Society B: Biologi-
cal Sciences,365, 2019–2024. https://doi.org/10.1098/rstb.2010.
0021
Weller, B., & Ganzhorn, J. U. (2004). Carabid beetle community composi-
tion, body size, and fluctuating asymmetry along an urban-rural gradi-
ent. Basic and Applied Ecology,5(2), 193–201. https://doi.org/10.
1078/1439-1791-00220
Whittaker, R., Rigal, F., Borges, P. A. V., Cardoso, P., Terzopoulou, S.,
Casanoves, F., ... Triantis, K. A. (2014). Functional biogeography of
oceanic islands and the scaling of functional diversity in the Azores.
Proceedings of the National Academy of Sciences of the United States
of America,111(38), 13709–13714. https://doi.org/10.1073/pnas.
1218036111
SUPPORTING INFORMATION
Additional supporting information may be found online in the
Supporting Information section at the end of the article.
How to cite this article: Jesse WAM, Behm JE, Helmus MR,
Ellers J. Human land use promotes the abundance and
diversity of exotic species on Caribbean islands. Glob Change
Biol. 2018;00:1–13. https://doi.org/10.1111/gcb.14334
JESSE ET AL.
|
13
