Niche conservatism and geographical range expansion of Pomacea canaliculata and Pomacea maculata in non-native United States and China

Abstract

The most noxious apple snails (Pomacea canaliculata and P. maculata) native to South America, currently have two distinct invaded ranges in China and the United States. Whether the environmental niches of the two closely related species have changed or remained stable (niche conservatism hypothesis) during the invasion process has become an important issue in forecasting their potential geographic distributions. For each Pomacea snail, two ecological niche models (ENMs, employing BIOMOD2) were generated based on bioclimatic variables and occurrence records in: (1) the native range; (2) the different invaded range. Conservation of ecological niche between the native and invasive snail populations was then tested by principal component and niche dynamics analysis. According to all models, precipitation contributed most to distribution of P. maculata, whereas low temperature was another most influential factor for spread of P. canaliculata. Niche conservatism was indicated by niche similarity tests and high niche stability for both Pomacea snails during their invasions in two regions. Niche expansions of P. canaliculata were relatively larger than unfilling values, whereas niche expansions of P. maculata were lower than unfillings. High niche unfilling for P. maculata in the United States revealed a great potential for further expansion in this region. We discussed the possible roles of physiological tolerances, genetic variation, residence time and hybridization in shaping niche changes for Pomacea snails during their invasion processes. Findings of this work can improve the understanding of potential mechanisms for niche differentiation and provide a theoretical basis for forecasting the invasion potential of Pomacea snails.

Full text

Vol.: (0123456789)
1 3
Biol Invasions (2023) 25:3391–3405
https://doi.org/10.1007/s10530-023-03100-9
ORIGINAL PAPER
Niche conservatism andgeographical range expansion
ofPomacea canaliculata andPomacea maculata
innon‑native United States andChina
ZhongQin· JiaenZhang · FuchengYao· JiminLiu· ZhaojiShi·
BenliangZhao· JingGuo
Received: 19 December 2020 / Accepted: 25 May 2023 / Published online: 9 July 2023
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2023, corrected publication 2023
Abstract The most noxious apple snails (Poma-
cea canaliculata and P. maculata) native to South
America, currently have two distinct invaded ranges
in China and the United States. Whether the envi-
ronmental niches of the two closely related spe-
cies have changed or remained stable (niche con-
servatism hypothesis) during the invasion process
has become an important issue in forecasting their
potential geographic distributions. For each Poma-
cea snail, two ecological niche models (ENMs,
employingBIOMOD2) were generated based on bio-
climatic variables and occurrence records in: (1) the
native range; (2) the different invaded range. Con-
servation of ecological niche between the native and
invasive snail populations was then tested by princi-
pal component and niche dynamics analysis. Accord-
ing to all models, precipitation contributed most to
distribution of P. maculata, whereas low temperature
was another most influential factor for spread of P.
canaliculata. Niche conservatism was indicated by
niche similarity tests and high niche stability for both
Pomacea snails during their invasions in two regions.
Niche expansions of P. canaliculata were relatively
larger than unfilling values, whereas niche expan-
sions of P. maculata were lower than unfillings. High
niche unfilling for P. maculata in the United States
revealed a great potential for further expansion in this
region. We discussed the possible roles of physiologi-
cal tolerances, genetic variation, residence time and
hybridization in shaping niche changes for Pomacea
snails during their invasion processes. Findings of
this work can improve the understanding of potential
mechanisms for niche differentiation and provide a
theoretical basis for forecasting the invasion potential
of Pomacea snails.
Keywords Niche conservatism· Ecological niche
models· Niche dynamics· Pomacea canaliculata·
Pomacea maculate· Invasive species
Supplementary Information The online version
contains supplementary material available at https:// doi.
org/ 10. 1007/ s10530- 023- 03100-9.
Z.Qin· J.Zhang(*)· F.Yao· J.Liu· Z.Shi· B.Zhao
Key Laboratory ofAgro-Environment intheTropics,
Ministry ofAgriculture andRural Affairs, 483 Wushan
Road, Tianhe District, GuangzhouCity510642, China
e-mail: jeanzh@scau.edu.cn
Z.Qin· J.Zhang· F.Yao· J.Liu· Z.Shi· B.Zhao
The Department ofEcology, College ofNatural Resources
andEnvironment, South China Agricultural University,
Guangzhou, China
Z.Qin· J.Zhang· F.Yao· J.Liu· Z.Shi· B.Zhao
Guangdong Provincial Key Laboratory ofEco-Circular
Agriculture, Guangdong Engineering Technology
Research Centre ofModern Eco-Agriculture andCircular
Agriculture, Guangzhou, China
J.Guo
School ofBiology andAgriculture, Shaoguan University,
Shaoguan512005, China

3392
Z.Qin et al.
1 3
Vol:. (1234567890)
Introduction
The snail genus Pomacea contains a successful
group of aquatic invaders that have caused detri-
mental agricultural and environmental impacts and
substantial economic losses in many countries of the
world. Among the most troublesome and much con-
cerned are Pomacea canaliculata (Lamarck, 1819)
and Pomacea maculata (Perry, 1810) [formerly P.
insularum (d’Orbigny, 1835)] (Ceanogastropoda:
Ampullariidae). The two apple snails are native to
freshwater wetlands of South America, but have
been widely introduced into North America, Europe,
east and southeast Asia and the Pacific Islands since
1980s (Hayes etal. 2008; Lopez Robles etal. 2010).
Because of the problematic nature in differentiating P.
canaliculata and P. maculata on the basis of morpho-
logical characteristics, their taxonomy has been rather
confused for a long period (Cazzaniga 2002). It is
now acknowledged that the two Pomacea species are
substantially genetically differentiated (Hayes et al.
2008; Rawlings etal. 2007).
The rapid spread of invasive Pomacea snails across
the continents and the resulting threats and damages
have prompted the task of improved understanding of
potential geographic distributions for effective con-
trol strategies and measures. Ecological niche models
(ENMs), a powerful methodological tool to identify
potential suitable environments for species by asso-
ciating their occurrence with the prevailing envi-
ronmental factors (Guisan and Thuiller 2005; Röd-
der and Lötters 2010) have been intensively used in
modeling species invasions. These models are usually
calibrated on the realized niche of species and make
a primary assumption that species retain their niche
properties during invasion (i.e. niche conservatism).
Previous field surveys and reports have identified the
colonization areas of Pomacea snails and provided
some insights in identification of regions with a high
invasion potential using niche-based models (Byers
etal. 2013; Lei etal. 2017). However, the ability of
these models to detect potential invasive ranges was
impeded by the problems in: (1) using Pomacea spp.
records without distinguishable genetic information
or taxonomic accuracy, has confounded efforts to
evaluate the biogeographic distribution of these spe-
cies (Cazzaniga 2002); (2) Whether or to what extent
the niches of Pomacea spp. are conserved remains
unclear. If the realized climatic niche of Pomacea
spp. differs in native versus invasive regions, the pre-
dictive capacity of ENMs will decrease (Welk 2004),
and models risk misrepresenting the potential for
invasion (Peterson 2006).
Introduced species have a higher probability of
successfully establishing viable populations in areas
with a climate that is similar to the native region (Di
Febbraro et al. 2013). However, it is also possible
that a niche shifts during biological invasion events
(Guisan etal. 2014), mainly because of nonanalogous
climate conditions between native and invaded areas
(Ribas etal. 2018). Studies on plants (Petitpierre etal.
2012), birds (Strubbe et al. 2013), insects (Petersen
2013), vertebrates (freshwater fish, mammals and
amphibians) (Strubbe et al. 2015) and invertebrates
(aquatic snails) (Torres et al. 2018) have reported
many cases of climatic mismatches during invasion.
Such climatic mismatches did not coincide with the
fundamental assumption of ENMs and make it diffi-
cult to identify the potential invasive areas accurately.
Thus, knowledge of niche conservatism between
native and invaded areas is very important in forecast-
ing the spread of invasive species. A recently devel-
oped framework (COUE scheme: centroid shift, over-
lap, unfilling, and expansion; Petitpierre etal. 2012)
in ENMs provided the technical development for
analyses of niche dynamics and niche comparisons.
These analyses have been widely used to test niche
changes and to explore the consequences of invasive
potential (Hill etal. 2017; Torres etal. 2018). How-
ever, few studies have addressed the niche status of
Pomacea spp. during their invasion process and the
key influential factors.
Pomacea canaliculata and P. maculata were ini-
tially intentionally introduced to south and southeast
Asia as an aquaculture food source and more recently
into mainland US and Hawaii through the aquarium
pet trade and for biological aquatic weed control
(Brito and Joshi 2016; Liu etal. 2019). Introduction
and invasion histories of the two Pomacea species
in these regions have been elucidated (Cerutti 1998;
Rawlings et al. 2007; Yang et al. 2018), with sub-
stantial molecular evidence for differentiating them
(Hayes etal. 2008; Kannan et al. 2020; Pasquevich
and Heras 2020). Different invasion histories of snail
populations in these regions may have potentially sig-
nificant implications regarding their current realized
ecological niches, evidence of niche changes in the
invaded range, and the risk of future invasion of new

3393
Niche conservatism andgeographical range expansion ofPomacea canaliculata andPomace…
1 3
Vol.: (0123456789)
geographic areas (Eckert etal. 2020). Records of the
two Pomacea snails across the two separated conti-
nents also provide the opportunity to examine niche
conservatism with respect to environmental varia-
tions and quantify the extent of niche differentiation
in two closely related congenerics.
We examine whether or not niches of P. canalicu-
lata and P. maculata are conserved during their inva-
sion across the two regions (i.e. China and US) using
ecological niche models (ENMs), with the hypothesis
that both snails would respond similarly to environ-
mental variables in invasive regions, and there will be
conservation of niches between the native and intro-
duced ranges. The aim of this study is to (1) inves-
tigate how environmental variables explain distribu-
tions of the two Pomacea snails in two regions; (2)
assess the extent of niche differences and calculate
niche metrics for comparing niches of the two Poma-
cea snails.
Materials andmethods
Species data
Occurrence records of P. canaliculata and P. macu-
lata in native range of South America were derived
from the Global Biodiversity Information Facility
Data Portal (GBIF; www. gbif. org) and CABIInva-
sive Species Compendium (ISC; https:// ckan. cabi.
org) (Fig. 1). For US, records of the two Pomacea
species were compiled from USGS Nonindigenous
Aquatic Species database (USGS-NAS 2020). Dis-
tribution data of the two Pomacea species in China
weresourced from: (1) published records in journals,
books and reports (Lv etal. 2013; Song 2010; Yang
et al. 2018); (2) mitochondrial COI sequences iden-
tification of specimens observed during field survey
(mostly opportunistic sampling) by the authors, and
(3) species-level identification of Pomacea specimens
from different geographical origins in south China
(provided by China Jiliang University). Because P.
maculata has a number of synonyms including the
name P. insularum, we queried each name in species
distribution sources to assemble native and invasive
occurrence data.
Specimens with uncertain location or non-reli-
able identification were discarded. For Pomacea
occurrences in South America and US, specimens
collected and genetically verified from the year 2007
to the present were used for niche analysis. For Poma-
cea occurrences of China, specimens collected after
2010 were used to be in approximate concordance
with the clarification of the genetic distinctiveness
of the two congeneric Ampullariidae snail species.
Records of Pomacea species were adjusted to the ras-
ter of the climatic variables with a spatial resolution
of 2.5 arcmin. To minimize spatial autocorrelation,
records were randomly selected within each grid cell.
The resulting datasets of P. canaliculata contained 75
unique occurrence localities in native, 35 and 66 loca-
tions for invasive ranges of China and US. Datasets of
P. maculata contained 41 unique occurrence localities
in native, 485 and 20 locations for the two invasive
ranges respectively.
Environmental variable
Six out of 19 available bioclimatic variables were
derived from Worldclim database (www. world clim.
org; Version 2.0) for the 1970–2000 period at 2.5
Fig. 1 Occurrence of Pomacea canaliculata (red filled circles)
and P. maculata (black filled square) collected in native South
America range. Convex hulls in green and lilac color represent
the 10-km buffered minimum convex polygon for P. canalicu-
lata and P. maculata occurrence locations, respectively

3394
Z.Qin et al.
1 3
Vol:. (1234567890)
arcmin spatial resolution (c. 5km × 5km). Thesub-
sets comprised of two temperature variables including
maximum temperature of the warmest month (bio5)
and minimum temperature in the coldest month
(bio6), three precipitation variables including annual
precipitation (bio12), precipitation of the driest quar-
ter (bio17), and precipitation of the warmest quarter
(bio18) as well as geographic elevation. These biocli-
matic variables represent annual trends, seasonality
and extremes that are appropriate to explain species
survival (Hijmans etal. 2005) and were thus selected
based on their roles in limiting Pomacea species
establishment and survival, and therefore distribution
(Byers etal. 2013; Lei etal. 2017). Multicollinearity
among the six candidate variables and possible over-
fitting were assessed based on Pearson correlation
and variance inflation factor statistics (VIF) using
usdm package in R (Naimi etal. 2014). No sign of
strong multi-collinearity existed among the six can-
didate variables (Spearman rank correlation < 0.7 and
VIF < 10 (Dormann et al. 2013). This set of biocli-
matic variables has been considered as the most influ-
ential in predicting distribution of P. maculata in the
southeastern US (Byers etal. 2013).
Geographic backgrounds
The background environments for climate niche
comparison should only include areas that have been
accessible to the species (Barve et al. 2011). To
delimit the geographic background for niche com-
parison of the two Pomacea snails, a gridded Köp-
pen–Geiger climate classification for the period
1951–2000 at a resolution of 30 s derived from the
CliMond database (Kriticos etal. 2012); www. climo
nd. org), was used for sub‐setting the available envi-
ronment for the species within each range. The Köp-
pen–Geiger system was selected because it classifies
climate into 5 main classes and 30 sub-types based on
threshold values and seasonality of monthly air tem-
perature and precipitation, and has been widely used
to define the background based on species distribu-
tion records. The distribution maps for both Pomacea
snails spatially intersected with the Köppen–Geiger
climate zones at 2.5 arcmin resolution, respectively.
The resulting Köppen–Geiger polygons containing
one or more species records were included in the
background of each range.
In South America, the geographic background of
the two Pomacea snails was delimitated by consider-
ing their demographic history, field survey collections
and molecular phylogenetic data. The range of P. can-
aliculata is restricted to the Lower Paraná, Uruguay,
and La Plata basins, while P. maculata has a much
larger range, occurring throughout much of western
Brazil, from the border of Paraguay in the south to
the Amazon Basin in the north (Hayes etal. 2012). A
10-kmbuffer zone for the minimalconvexhull cov-
ering all occurrence locations was drawn to define
the background for each Pomacea snail (Fig.1). For
invaded China and US ranges, the Köppen–Geiger
polygons where the species occurred were consid-
ered to be potentially suitable for the species. Pseudo-
absences were then extract randomly from the native
and invasive backgrounds as described below.
Testing niche conservatism
An analytical framework developed by Broenni-
mann etal. (2012) was employed to assess the envi-
ronmental niche variation of the two Pomacea snail
between their native range and the introduced ranges
(China, US) separately. This method creates a global
environmental space that covers all the environ-
mental conditions where the species occurs, includ-
ing both native and invaded areas, and generates
occupancy values based on comparisons of the spe-
cies occurrence data with the global environmental
space (Broennimann etal. 2012). For each Pomacea
snail, principal component analysis (PCA) using the
selected six environmental variables, combined with
the species occurrence data and the species pseudo-
absence data was performed. The first two axes of
PCA were used to create an environmental grid of
100 × 100 cells (global environmental space), where
species occupancy was allocated (Broennimann etal.
2012). Species’ densities of occurrence within each
grid cell were calculated using a kernel function to
generate smooth distribution for native and invasive
datasets, respectively. The niche metrics accounting
for niche overlap, niche similarity, niche equivalency
were compared between native and non-native ranges
for the two Pomacea snails, respectively. To complete
this, the Schoener’s D metrics, an index ranging from
0 (no overlap) to 1 (complete overlap) were used to
measure niche overlap. Niche equivalency and niche
similarity test were performed based on the 95%

3395
Niche conservatism andgeographical range expansion ofPomacea canaliculata andPomace…
1 3
Vol.: (0123456789)
confidence interval to test the null hypothesis of the
similar and equivalent niches for both Pomacea snails
in their native and introduced regions, respectively
(Warren et al. 2008). The significance of similarity
and equivalency tests was assessed by 1000 permu-
tations. Three niche metrics namely: niche stability
(proportion of the native niche observed in the exotic
niche), unfilling (proportion of the native niche not
occupied in the exotic niche) and expansion (new
environmental requirements observed in the exotic
niche), were calculated to quantify the niche dynam-
ics of Pomacea species (Petitpierre et al. 2012). A
proportion of 95% of the intersection area between
the native and invaded gridded environmental space
was used to control for possible environmental outli-
ers. Additionally, the niche metrics were also calcu-
lated for comparisons between species in the same
geographical region. To complete this, the proportion
of environmental space occupied by P. canaliculata
or P. maculata solely as well as by both species was
quantified using the similar method mentioned above.
All niche comparisons were performed with the
ecospat package (Di Cola etal. 2017) in R environ-
ment (R Development Core Team).
Ecological niche modeling
Potential distribution of Pomacea snails in their native
and invasive ranges was predicted using the same cli-
matic variables, occurrence localities as described in
niche conservatism tests. Two different models, based
on Pomacea species occurrences of the native range
and invaded ranges only, were built for each species
separately. The resulting models were projected onto
the studied invasive range to test their performances.
The BIOMOD2 modelling framework imple-
mented in R 4.0.2 (R Development Core Team 2012)
was employed to project the potential distribution
of P. canaliculata or P. maculata. An ensemble of
three modeling techniques were performed: gener-
alized boosted models (GBM), artificial neural net-
works (ANN) and random forests (RF). The mean
of probabilities of the three selected algorithms was
used for ensemble building, which has been proved
to supply more robust predictions than other ensem-
ble building techniques (Marmion et al. 2009). The
equal number of pseudo-absences were generated by
randomly sampling from the grid cells of each study
range (Barbet-Massin et al. 2012). Default settings
for three modeling techniques were used, except that
3000 trees were used as fitting basis for GBMs and
500 trees were built for RF. For each Pomacea snail,
70% of occurrence records were randomly selected
for calibration and 30% reserved for evaluation by
cross-validation. Ten replicates run for each model.
To compare the magnitude of influence of predic-
tors in the models, variable importance values were
derived using the function provided in BIOMOD2.
Performance of each modeling technique was evalu-
ated by the true skill statistic (TSS). Only the best fit-
ted model runs above critical TSS values (> 0.4) and
AUC (> 0.7) were implemented in the final ensemble
model run. To constrain model uncertainty, consen-
sus prediction models were obtained using a TSS
weighted average method to account for the predic-
tive power of each algorithm. Binary suitable/non-
suitable maps for Pomacea snails were created using
the threshold that maximizes the true skill statistic
(TSS), which was known to improve the accuracy of
prediction (Jiménez-Valverde and Lobo 2007).
To examine analogy of the presumed introduction
locations with native range, the multivariate environ-
mental similarity surface (MESS) analysis was per-
formed for each Pomacea snails separately. In MESS
analysis, the environment of grid cells occupied by
the snail in a specific invasive range was compared
with that of the native range, with respect to the
set of selected environmental variables (Elith et al.
2010). The grid cells having positive value indicate
similar environment between two ranges whereas grid
cells with the dissimilar environment for at least one
variable receive negative values (Broennimann etal.
2014). MESS analysis was conducted in R with the
ecospat package (Di Cola et al. 2017). MESS out-
put values (climatic similarity) were reported at the
presence records in the range where the model was
projected.
Results
Niche dynamics and comparisons
The invasive niche centroid of Pomacea snails in
China shifted towards areas with higher annual pre-
cipitation (bio12) and elevation, and lower minimum
temperature in the coldest month (bio6) (Fig.2b; Fig.

3396
Z.Qin et al.
1 3
Vol:. (1234567890)
Fig. 2 Environmental niche pattern of native P. canaliculata
(a) and its changes in China (b) and US (d) based on princi-
pal component analysis. The first two axes of each PCA repre-
sent the density of species occurrences and the environmental
space. Solid and dashed lines indicate 100% and 90% of all the
available environments. The blue color represents the niche
overlap between native and introduced region. The green and
red color represent niche unfilling, niche expansion environ-
ments. The red arrows show the change in the niche centroid
between native and introduced range. c and e describe the
impacts of environmental variables to niche changes in China
and the United States, respectively. The correlation circles
represent the variable importance along the first two principle
axes

3397
Niche conservatism andgeographical range expansion ofPomacea canaliculata andPomace…
1 3
Vol.: (0123456789)
S1b and S2). The invasive niche centroid of P. cana-
liculata in the USshowed a general broader range of
precipitation conditions (i.e., bio12 and bio18) than
the native niche (Fig. 2d; Fig. S3b–c), whereas the
niche centroid of P. maculata moved towards areas
with narrower variability in precipitation conditions
and a higher maximum temperature of the warmest
month (bio5) (Fig. S1d and S3d–f).
The first two axes of the PCA used to compare the
environmental niches of the Pomacea snails between
the native and invaded China explained proportions
of variance (P. canaliculata: 80.65%; P. maculata:
81.73%). The first two axes of PCA-env explained
slightly lower proportions (P. canaliculata: 72.59%;
P. maculata: 78.32%) of variance across the occu-
pied native and invaded US environments. For
native-China ranges, all the environmental predic-
tors loaded strongly in the PC1 axis (around 62.0%
for the two Pomacea snails), with both temperature
(loading values from − 0.77 to − 0.85) and precipita-
tion (loading values from − 0.67 to − 0.89) variables
being negatively correlated and elevation positively
correlated (loading values around 0.80) (Fig.2c; Fig.
S1c). Annual precipitation (bio12) contributed most
to the PC1 axis, followed by minimum temperature
in the coldest month (bio6). The PC1 axis explained
maximum environmental variation (53.7–62.7%) for
native-US ranges, with both temperature (loading val-
ues from − 0.59 to − 0.84) and precipitation (loading
values from − 0.81 to − 0.92) variables being nega-
tively correlated and elevation positively correlated
(loading values around 0.62) (Fig.2e; Fig. S1e). Max-
imum temperature of the warmest month (bio5) posi-
tively contributed most to the PC2 axis, while bio12
contributed most negatively to the PC1 axis, followed
by precipitation of the warmest quarter (bio18).
Niche change indices of P. canaliculata and P.
maculata revealed moderate and low degree of niche
expansions in invaded ranges respectively, although
larger niche stabilities were detected for P. maculata
(Table1). Niche unfillings were no more than 0.1 for
P. maculata. In US range, P. canaliculata occupied
the largest area in the PCA space which signified the
wide range of environments experienced by the spe-
cies in this region. The exotic niche for P. maculata
in USwas unique because of highest degree of stable
and unfilled niches, and was concurrent with the low-
est Schoener’s D value.
For each Pomacea snail, the niche compari-
sons showed modest overlaps (Schoener’s D values
around 0.4) among environmental conditions occu-
pied in native-China ranges, and low overlaps (Sch-
oener’s D values of 0.28 and 0.18 for P. canalicu-
lata and P. maculata) in native-US ranges (Table1).
The hypothesis of niche equivalency was confirmed
for both Pomacea snails and invasive regions, given
that the observed overlaps between invaded and
native ranges were significantly higher than 95% of
simulated overlaps (Table1). All the observed niche
overlaps between the native and invasive niches were
significantly higher than the random niche overlap
(p > 0.05), indicating equivalency of the environmen-
tal nichesexisted.
Among the six environmental candidates, the
proportion of niche expansion of Pomacea snails
in invaded range of Chinawas highest for bio18 (P.
canaliculata: 0.2850; P. maculata: 0.1687), followed
by elevation (P. canaliculata: 0.2235; P. maculata:
0.1038) (Table2). Full niche stability of both Poma-
cea snails was detected along precipitation of the dri-
est quarter (bio17). Niche dynamics in invaded range
ofUS displayed different species-specific pattern. For
P. canaliculata, the proportion of niche expansion
was highest for bio18 (0.3388) and bio12 (0.1718),
with a considerable proportion of niche unfilling for
elevation (0.0679). For P. maculata, varying degrees
Table 1 Niche change
metrics of Pomacea
canaliculata and P.
maculata based on the
analogous (common)
environmental spaces
between the native and the
two invaded ranges
Species P. canaliculata P. maculata
Region Native-China Native-US Native-China Native-US
Overlap (O) 0.4358 0.2833 0.4885 0.1778
Equivalence (p value) 0.9802 1.0000 0.7326 1.0000
Similarity (p value) 0.0990 0.1782 0.0594 0.2178
Expansion (E) 0.2366 0.3918 0.0713 0.0262
Stability (S) 0.7634 0.6082 0.9287 0.9738
Unfilled (U) 0.0349 0.0973 0.2932 0.6621

3398
Z.Qin et al.
1 3
Vol:. (1234567890)
of niche unfillings were observed for each environ-
mental variable, with peak proportions along bio6
(0.5641) and bio12 (0.2152) (Table2).
Predicted distributions
The ensemble modeling of the current distribution
of Pomacea snails obtained high evaluation scores
(TSS values of all the ENMs were great than 0.85)
(Table3). Models computed from the native or inva-
sive ranges showed high predictive performance
generally. The native model identified environmen-
tal suitabilities (on average, ≥ 0.274 for P. canalicu-
lata and ≥ 0.473 for P. maculata) in southeast China
and included habitats being invaded during the early
phase of invasions (Fig. 3a and c). In US, projec-
tions from the native model yielded environmental
suitability (on average, ≥ 0.283 for P. canaliculata
and ≥ 0.494 for P. maculata) in southeastern US and
an area that extending from north Louisiana to cen-
tral portions of Missouri to south Illionis and adjacent
regions (Fig. S4a and c).
Both the native and China-range models for P.
maculata overpredicted the extent occupied in south-
eastern US. The species did not fully occupy all the
environmental space that is occupied in the native
range. Occurrences of Pomacea snails in invaded
range of China were better predicted with the inva-
sive models for US. Environmental suitabilities for
P. canaliculata were identified in south China and
Table 2 Niche dynamics of Pomacea canaliculata and P. maculata along the environmental variables
The initial set of six candidate variables were selected by assessment of their ecological relevance to Pomacea snails and collinearity
tests of VIF. Variables definition: bio5: maximum temperature of the warmest month (°C* 10); bio6: minimum temperature of cold-
est month (°C* 10); bio12: annual precipitation (mm); bio17: precipitation of driest quarter (mm); bio18: precipitation of warmest
quarter (mm)
Different upper letters indicate the snail species. “sp1” represents Pomacea canaliculata, whereas “sp2” represents P. maculata. The
highest values of niche expansion and theunfilled were set in bold
Region Niche dynamics bio5 bio6 bio12 bio17 bio18 Elevation
Native-Chinasp1 Expansion (E) 0.0152 0.0200 0.0351 0.0000 0.2850 0.2235
Stability (S) 0.9848 0.9800 0.9649 1.0000 0.7150 0.7765
Unfilled (U) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
Native-USsp1 Expansion (E) 0.0129 0.0129 0.1718 0.0000 0.3388 0.0000
Stability (S) 0.9871 0.9871 0.8282 1.0000 0.6612 1.0000
Unfilled (U) 0.0043 0.0000 0.0000 0.0000 0.0000 0.0679
Native-Chinasp2 Expansion (E) 0.0000 0.0000 0.0000 0.0000 0.1687 0.1038
Stability (S) 1.0000 1.0000 1.0000 1.0000 0.8313 0.8962
Unfilled (U) 0.0044 0.0615 0.0009 0.0000 0.0041 0.0000
Native-USsp2 Expansion (E) 0.0000 0.0000 0.0000 0.0000 0.0005 0.0000
Stability (S) 1.0000 1.0000 1.0000 1.0000 0.9995 1.0000
Unfilled (U) 0.5641 0.0034 0.2152 0.0062 0.0080 0.1011
Table 3 Performance of
the ensemble models for
Pomacea canaliculata and
P. maculata based on TSS
values TSS True kill statistic
*Models based on Pomacea
species occurrences from
the native range or invaded
ranges only, were projected
to the target invasive range
to test their performances
Species Models* Testing data Cut off Sensitivity Specificity
P. canaliculata Native-China 0.862 332 97.059 89.189
US-China 0.996 679 99.93 99.93
Native-US 0.836 393 94.118 89.527
China-US 0.940 489 98.333 95.703
P. maculata Native-China 0.938 473 99.92 93.750
US-China 0.974 322 99.79 97.612
Native-US 0.892 401 97.368 91.875
China-US 0.987 613 99.90 98.684

3399
Niche conservatism andgeographical range expansion ofPomacea canaliculata andPomace…
1 3
Vol.: (0123456789)
patches in southeast China (Fig. 3b). Parts of south
China, portions in southeast coastal areas, contigu-
ousareaof western Sichan and Chongqing were pre-
dicted to be suitable for P. maculata (Fig. 3d). Simi-
larly, presences of P. canaliculata or P. maculata in
invaded US range were better predicted using the
models calibrated on invaded China range (Fig. S4b
and d). Moreover, the niche overlaps between the two
invaded ranges were higher than those between the
native-invaded US ranges. These niche overlaps were
more equivalent than random.
The MESS analysis for P. canaliculata identified
non-analog small patches scatted at the margin of
eastern Taiwan, parts of southwest regions in China
(Fig. S5a–c). The margins of western US and in scat-
ted habitats of central US were non-analog to the
native range (Fig. S5d–f). For P. maculata, the non-
analog environments detected in China were almost
the same as those found for P. canaliculata except
that southwest regions in China displayed analog to
the native range (Fig. S6a–c). Regions at southwest
margins of US had non-analog environments to the
native range (Fig. S6d–f).
Discussion
Environmental factors affecting the potential
distribution
Regional distributions of P. canaliculata and P. mac-
ulata are currently spanning a wide range of environ-
ments between the native South America and inva-
sive ranges (Fig.1). Difference in local adaptation of
Pomacea populations across the ranges may connect
with the environmental variables, which contributed
the most to regional ENMs. When native Poma-
cea populations spread to invasive ranges of China,
the most influential factors (i.e., bio12 and bio6)
revealed the moisture and thermal restrictions to
northward dispersal of Pomacea species. The find-
ing aligned with numerous biological studies and
modeling work on Pomacea snails (Lv et al. 2006;
Matsukura and Wada 2007) reportedthat dry condi-
tions and cold weather in winter limited the expan-
sion of this species. In invaded US, the most influen-
tial environmental variables revealed the necessity of
moisture and thermal environments for Pomacea spe-
cies to establish permanently. Extreme temperature
conditions in invasive ranges therefore, highlighted
the constraints of thermal tolerance on establishment
of Pomacea species. Pomacea canaliculata is gener-
ally considered to high adaptability and variability
in thermal tolerance which allows them to colonize
areas with high peak temperatures (e.g. in the sum-
mer in Hawaii, Hainan island) as well as cold winter
conditions (e.g. in Ibaraki of Japan) (Wada and Mat-
sukura 2007). This species had acquired sufficient
cold tolerance to colonize East Asia before they were
introduced (Wada and Matsukura 2011; Yoshida etal.
2014). In east and south-east Asia, the apple snails
have invaded a wide range from tropical to temper-
ate regions after several introductions (Hayes et al.
2008), northern expansion of this species was deter-
mined by low temperature during winter (Ito 2002;
Syobu etal. 2001). Compared with P. canaliculata, P.
maculata has poorer tolerance of both cold and desic-
cation (Yang etal. 2020a, b), suggesting that P. macu-
lata was more adapted to temperate southern, western
coastal regions and Florida, the major habitats of this
invasive apple snails in US. The findings on differ-
ence in ecophysiological tolerances of the two Poma-
cea snails, have implications for understanding differ-
ential spread potentials in the invasive range.
Patterns of niche dynamics and possible explanations
Pomacea maculata has a much larger native range
than P. canaliculata (Hayes etal. 2012). Comparisons
of environmental variables in P. canaliculata and P.
maculata native occurrences showed statistically sig-
nificant differences for half of 6 variables used in this
study (Table 2). Elevation values for native occur-
rences of the two Pomacea snails revealed no sig-
nificant difference between species elevation profiles.
Instead, both species were reported to inhabit in low
elevation fresh water bodies, wetland areas and paddy
field in their native ranges (Hayes et al. 2015). This
indicated that the two Pomacea snails occur under
similar environmental conditions despite the differ-
ences in native range sizes.
During invasion in US and China regions, both P.
canaliculata and P. maculata retain signatures of their
native environmental niche, but also indicate changes
in environmental space following introduction. The
two species showed niche conservatism (niche sta-
bility) in two invaded regions as indicated by the
higher stability values of 0.76 and 0.97, respectively

3400
Z.Qin et al.
1 3
Vol:. (1234567890)
and lower unfilling and expansion values (Table 1).
Niche expansions of P. canaliculata were relatively
larger (0.2366 and 0.3918 in invaded China and US)
compared with the unfilling (no more than 0.1 for
two invasive ranges). Contrastly, niche expansions of
P. maculata (no more than 0.1) were relatively rare
compared with the unfilling (0.2932 and 0.6621 in
invaded China and US). Differences in niche dynam-
ics of the two Pomacea snails can be ascribed to
physiological tolerances, genetic variation, residence
time and dispersal limitations.
In invaded ranges of China, P. canaliculata popu-
lations were introduced more than once from mul-
tiple locations in Argentina, with higher haplotype
Fig. 3 Predicted distribution of Pomacea snails in invaded
China using the ensemble models calibrated with either the
native (P. canaliculata: (a); P. maculata: (c)) or invaded US
occurrences (P. canaliculata: (b); P. maculata: (d)). Grey
shading on the binary maps indicates areas with suitable envi-
ronmental requirements for Pomacea snails. Known localities
of P. canaliculata (blue filled squares) and P. maculata (red
filled circles) in invaded China were displayed in correspond-
ing small figures

3401
Niche conservatism andgeographical range expansion ofPomacea canaliculata andPomace…
1 3
Vol.: (0123456789)
diversity than in native populations (Hayes etal. 2008;
Lv et al. 2013). The multiple introduction events of
P. canaliculata enhanced genetic diversity, allowing
the species to occupy an environmental space that
coincides with the native niche (Ahmad etal. 2019),
and also creating the opportunity for niche expansion
when individual snail face novel environments. Dif-
ferent from P. canaliculata, P. maculata populations
had a much lower haplotype diversity in China than
that in their native populations (Yang et al. 2018),
with thata single lineage from Brazil was introduced
and established (Hayes etal. 2008). The comparative
higher haplotype diversity and wider distribution of P.
canaliculata populationsin China, may explain why
P. canaliculata have colonized a larger portion of the
suitable environments than P. maculata, despite simi-
lar introduction dates in this region.
Our findings of a high proportion of niche stabil-
ity support niche conservatism between US invasive
P. maculata populations and the native populations,
but also indicate that currently P. maculata does not
occupy all the suitable areas available in US. Further-
more, hybridization of P. canaliculata and P. macu-
lata was unlikely to have occurred because of the low
genetic variation found in Pomacea snails from Loui-
siana and Texas (Mueck 2017). Pomacea maculata
populations having large niche unfilling may attrib-
ute to the short time since introduction in this region
(nor more than 40 years). Niche unfilling seems to
be larger for species introduced recently and into
a small number of locations, compared with those
with ancient colonization history and introduced in
several points in space (Strubbe etal. 2015). In US,
niche unfilling was larger for P. maculata introduced
recently and into a small number of locations, when
compared with P. canaliculata being introduced a
few years earlier. Despite the short colonization span,
increasing human-mediated importation, acquisi-
tion and transportation makes P. maculata espe-
cially capable of invading new suitable habitats and
lead to explosive population growth. For instance, in
Southwest Louisiana, P. maculata was reported to be
expanding north at an average rate of approximately
6km per year. Human transportation through attach-
ment to boats was expected to be among the mecha-
nisms for the most recent introduction of this species
(Lucero 2021). Given the high level of niche unfill-
ing in US, the invasive potential of P. maculata in this
region might be extremely high. Therefore, effective
population control in areas already invaded is needed
to prevent further expansion of P. maculata into adja-
cent southeastern areas.
Both P. canaliculata and P. maculata can occupy
new environments in the invaded ranges because of
preadaptation to those environments in their native
range (Rawlings etal. 2007). For instance, the snails
can survive adverse conditions by retreating into
their shell and closing it firmly with the operculum
(Horgan etal. 2014). The presence of adaptivecapac-
ity could lead to a realized niche expansion in the
invaded range. However, challenges remain to dif-
ferentiate between an evolutionary change, pheno-
typic plasticity and ecological responses (Moran and
Alexander 2014). In Asia, P. canaliculata has been
reported to survive months of drought by digging
deep into the mud and closing their operculum, sur-
facing again after renewed flooding. In the same way
snails hibernate in winter at the northern border of
their distribution (Oya 1987). Such burrowing behav-
ior was not evidenced in P. maculata. But this spe-
cies has the capacity to move over land and sustain
an aestivated state for over ten months, which may
lead to an increase in geographic range and viability
(Mueck 2017). More detailed comparative studies
of the two Pomacea snails are expected to offer new
insights into the ecological or evolutionary processes
that allow species to colonize new environments.
Potential causes of niche changes
Hybridization or other genetic changes in popula-
tions may allow for greater changes in niche space,
which in turn may lead to increased invasiveness and
more rapid range shifts or better adaptation (Thorn-
ton and Murray 2014). Pomacea canaliculata and P.
maculata had supposed to have significant genetic
differentiation and limited gene flow among differ-
ent countries (Hayes et al. 2009). However, many
studies documented the existence of substantial
introgressive hybridization (Kannan etal. 2021; Mat-
sukura etal. 2013).Genetic exchange and the ongoing
hybridization between the two Pomacea populations
were supported by the sympatrical occurrence and
mating in the field (e.g., paddy field), both in native
areas (Argentina) (Glasheen et al. 2020) and inva-
sive Asia countries (e.g. Japan, South Korea, China,
and Malaysia) (Matsukura et al. 2013; Yang et al.
2020a, b). Hybridization or other genetic changes in

3402
Z.Qin et al.
1 3
Vol:. (1234567890)
populations may allow for greater changes in niche
space, which in turn may lead to increased invasive-
ness, more rapid range shifts or better adaptation and
tolerance of climate change (Matsukura etal. 2016).
In China, P. maculata populations possessed nuclear
genotypes of P. canaliculata to a certain extent and
have acquired improved dessication and cold toler-
ances, thus expanding their distribution into temper-
ate regions (Yang etal. 2020a, b). We speculate that
hybridization or other processes that promote genetic
changes in Pomacea populations may influence the
presence and degree of niche conservatism in expand-
ing populations. However, current knowledge on
potential influence of hybridization on niche stasis or
change remains very limited and was hindered by dif-
ficulties in extensive identification of pure and hybrid
snails in natural systems.
Uncertainties and challenges
We compared the realized environmental niches of
two Pomacea snails in their native (East Asia) and
invasive (US and China) ranges respectively. Niche
conservatism was indicated by niche similarity tests
and high niche stability, with the results showing that
the native and invaded niches of Pomacea snails were
more equivalent than random and were more similar
than expected by chance. The findings were compat-
ible with the reports on a variety of introduced non-
native species, which have emphasized the prevalence
and important role played by niche conservatism dur-
ing different stages of biological invasions (Broenni-
mann etal. 2014; Petitpierre etal. 2012). However, a
recent study on 22 invasive freshwater invertebrates
in New Zealand indicated that 90% of these species
(including P. canaliculata) showed a significant niche
change (Torres etal. 2018). Niche conservatism of P.
canaliculata and P. maculata populations in our study
seems to be special cases since niche shifts are more
common for freshwater invertebrates than those for
other organisms (Parravicini etal. 2015). Such a dif-
ference would reflect distinctive selection of variables
for defining the species niche, sample size, study
regions, ect. The selection of the environmental pre-
dictors for niche modeling is a source of uncertainty
in model predictions. We based the niche analysis on
the several environmental factors only for this study.
Additional variables (e.g., land cover, soil type, dis-
solved oxygen and pH) having effects on snail growth
and reproduction, may allow the precise characteri-
zation of aquatic species’ environmental niche but
not available in the bioclim dataset (Hah etal. 2022;
Seuffert and Martín 2021). Human or flood regime-
mediated dispersal and altered biotic processes (e.g.,
competition, predation) are effective in shaping fresh-
water species distributions (Gallardo etal. 2015; Loo
et al. 2007). Ignoring these factors may potentially
lead to substantial uncertainties in assessing niche
conservatism, because these factors typically become
important at the regional or local scale. Therefore,
more intensive research efforts are needed to combine
the biotic (e.g., ecophysiological tolerances, behavior
flexibility, adaptive genetic variation, competition)
and abiotic factors (e.g., land cover, habitat type) in
the distribution models for Pomacea snails to have a
more refined understanding of successful establish-
ment and niche dynamics during the invasion process.
Studies on genetics contributed to a better charac-
terization of the taxonomy and phylogeography of the
two Pomacea species. These case studies were car-
ried out on fragmentary sampling sites so they could
not reflect the whole spectrum of conditions inhab-
ited by the Pomacea species in the study ranges and
might partially explain the observed niche status. For
this, we calculated the density of occurrences of the
two Pomacea species respectively, using a smooth
kernel density function to correct for potential sam-
pling biases (Broennimann et al. 2012). For each
Pomacea species, we used the first two axes of a Prin-
cipal Component Analysis (PCA) including the six
most influential bioclimatic variables to create grid-
ded environmental gradients, thereby allowing the
comparison between the environmental spaces avail-
able for the species in the different biogeographic
regions. The niches were then visualized and com-
pared between native and the two non-native ranges
accounting for niche metrics (Broennimann et al.
2012; Petitpierre etal. 2012). This framework could
be helpful to prevent the bias that occurred in records
of studied ranges from unbalanced sampling and spe-
cies identificationefforts. However, it should be noted
that the assessment for each Pomacea species fit only
to that portion of the niche that was represented by
the observed records. The degree to which the Poma-
cea species is at equilibrium with current environ-
mental conditions should be evaluated more care-
fully with the variations of the available records.To
acquire adequate understanding of niche dynamics,

3403
Niche conservatism andgeographical range expansion ofPomacea canaliculata andPomace…
1 3
Vol.: (0123456789)
both standardized field survey along with the phy-
logenetic and phylogeographic data are needed to
obtain extensive occurrence records for providing
accurate pictures as to the environments colonized by
the two Pomacea species in the study regions.
Targeting geographical background specifically
involved in the invasion process can help captur-
ing detailed niche dynamics in invaded ranges.
Here we refined geographical backgrounds to two
invaded ranges (China and US) when assessing the
niche stasis of the two South America introduced
Pomacea snail populations. Because lack of reliable
identified species of apple snail information, pat-
terns of environmental niche dynamics across other
regions were not investigated in this study. Fur-
ther work in this regard would be (i) acquire more
detailed genetic data sampled across native and
invasive distributions; (ii) assess whether the Poma-
cea snail will be able to occupy additional ecologi-
cal niches beyond its native range and expand its
current geographic distribution globally. Although
more intensive research efforts are needed for a bet-
ter understanding of niche status across the globe,
our study is one of the few studies in which niche
conservatism of Pomacea snails was investigated
using available genetically confirmed data in sets
of geographic extents. We have shown the patterns
of realized niche expansion and unfilling across the
invaded ranges of P. canaliculata and P. maculata
in US and China. The results of niche conservatism
provide novel insights into the usefulness of ENMs
for predicting regional invasion potential of the two
Pomacea snails and can be useful in identify impor-
tant drivers associated with niche transferability
across regions.
Acknowledgements This work was supported by the
National Natural Science Foundation of China (41871034,
31870525, 31770484, 31901229, U1131006), Science and
Technology Planning Project of Guangdong Province of China
(2019B030301007), Guangdong Modern Agricultural Tech-
nology Innovation Team Construction Project (2023KJ134,
2023KJ105). We thank anonymous reviewers for providing
helpful comments on earlier versions of this manuscript.
Author contribution ZQ Data analysis, devised and drafted
the manuscript. JEZ Conceptualization, review and editing.
FCY and JMLAssistance for revision. BLZ andJG Data col-
lection and filtering.ZJSMap creation and interpretation.
Data availability Datasets in the article can be accessed at
Dryad Digital Repository with https:// doi. org/ 10. 5061/ dryad.
2fqz6 12n7
Declarations
Conflict of interest The authors declare that they have no
conflict of interest.
References
Ahmad R, Khuroo AA, Charles B etal (2019) Global distri-
bution modelling, invasion risk assessment and niche
dynamics of Leucanthemum vulgare (Ox-eye Daisy)
under climate change. Sci Rep-Uk 9:11395
Barbet-Massin M, Jiguet F, Albert CH, Thuiller W (2012)
Selecting pseudo-absences for species distribution mod-
els: how, where and how many? Methods Ecol Evol
3:327–338
Barve N, Barve V, Jiménez-Valverde A etal (2011) The cru-
cial role of the accessible area in ecological niche mod-
eling and species distribution modeling. Ecol Model
222:1810–1819
Brito F, Joshi R (2016) The golden apple snail Pomacea cana-
liculata: a review on invasion, dispersion and control.
Outlooks on Pest Management 27:157–163
Broennimann O, Fitzpatrick MC, Pearman PB et al (2012)
Measuring ecological niche overlap from occurrence
and spatial environmental data. Global Ecol Biogeogr
21:481–497
Broennimann O, Mráz P, Petitpierre B, Guisan A, Müller-
Schärer H (2014) Contrasting spatio-temporal climatic
niche dynamics during the eastern and western inva-
sions of spotted knapweed in North America. J Biogeogr
41:1126–1136
Byers JE, McDowell WG, Dodd SR etal (2013) Climate and
pH predict the potential range of the invasive apple snail
(Pomacea insularum) in the southeastern United States.
PLoS ONE 8:e56812
Cazzaniga NJ (2002) Old species and new concepts in the tax-
onomy of Pomacea (Gastropoda: Ampullariidae). Biocell
26:71–81
Cerutti R (1998) An infestation of Pomacea canaliculata
(Lamarck, 1804) in Lake Miramar, San Diego. California
Festivus 30(25–27):29
Di Cola V, Broennimann O, Petitpierre B etal (2017) ecospat:
an R package to support spatial analyses and modeling of
species niches and distributions. Ecography 40:774–787
Di Febbraro M, Lurz P, Genovesi P etal (2013) The use of cli-
matic niches in screening procedures for introduced spe-
cies to evaluate risk of spread: a case with the American
Eastern Grey Squirrel. PLoS ONE 8:e66559
Dormann CF, Elith J, Bacher S etal (2013) Collinearity: a
review of methods to deal with it and a simulation study
evaluating their performance. Ecography 36:027–046
Eckert S, Hamad A, Kilawe CJ etal (2020) Niche change
analysis as a tool to inform management of two invasive
species in Eastern Africa. Ecosphere 11:e02987

3404
Z.Qin et al.
1 3
Vol:. (1234567890)
Elith J, Kearney M, Phillips S (2010) The art of modelling
range-shifting species. Methods Ecol Evol 1:330–342
Gallardo B, Zieritz A, Aldridge D (2015) The importance of
the human footprint in shaping the global distribution of
terrestrial, freshwater and marine invaders. PLoS ONE
10:e0125801
Glasheen PM, Burks RL, Campos SR, Hayes KA (2020)
First evidence of introgressive hybridization of apple
snails (Pomacea spp.) in their native range. J Mollus
Stud 86:96–103
Guisan A, Thuiller W (2005) Predicting species distribution:
offering more than simple habitat models. Ecol Lett
8:993–1009
Guisan A, Petitpierre B, Broennimann O, Daehler C, Kueffer
C (2014) Unifying niche shift studies: insights from bio-
logical invasions. Trends Ecol Evol 29:260–269
Hah H, Liew T, Rama Rao S, Yow Y, Ratnayeke S (2022)
Distribution and environmental associations of invasive
freshwater Pomacea snails in Peninsular Malaysia. Biol
Invasions 24:189–204
Hayes KA, Joshi RC, Thiengo SC, Cowie RH (2008) Out of
South America: multiple origins of non-native apple
snails in Asia. Divers Distrib 14:701–712
Hayes KA, Cowie RH, Thiengo SC (2009) A global phylog-
eny of apple snails: Gondwanan origin, generic relation-
ships, and the influence of outgroup choice (Caenogas-
tropoda: Ampullariidae). Biol J Linn Soc 98:61–76
Hayes KA, Cowie RH, Thiengo SC, Strong EE (2012) Com-
paring apples with apples: clarifying the identities of two
highly invasive Neotropical Ampullariidae (Caenogas-
tropoda). Zool J Linn Soc-Lond 166:723–753
Hayes KA, Burks RL, Castro-Vazquez A etal (2015) Insights
from an integrated view of the biology of apple snails
(Caenogastropoda: Ampullariidae). Malacologia
58:245–303
Hijmans RJ, Cameron SE, Parra JL, Jones PG, Jarvis A (2005)
Very high resolution interpolated climate surfaces for
global land areas. Int J Climatol 25:1965–1978
Hill MP, Gallardo B, Terblanche JS (2017) A global assess-
ment of climatic niche shifts and human influence in
insect invasions. Global Ecol Biogeogr 26:679–689
Horgan FG, Stuart AM, Kudavidanage EP (2014) Impact of
invasive apple snails on the functioning and services
of natural and managed wetlands. Acta Oecologica
54:90–100
Ito K (2002) Environmental factors influencing overwintering
success of the golden apple snail, Pomacea canaliculata
(Gastropoda: Ampullariidae), in the northernmost popula-
tion of Japan. Appl Entomol Zool 37:655–661
Jiménez-Valverde A, Lobo JM (2007) Threshold criteria for
conversion of probability of species presence to either or
presence-absence. Acta Oecologica 31:361–369
Kannan A, Rao SR, Ratnayeke S, Yow Y (2020) The efficiency
of universal mitochondrial DNA barcodes for species dis-
crimination of Pomacea canaliculata and Pomacea macu-
lata. PeerJ 8:e8755
Kannan A, Ratnayeke S, Yow Y (2021) Molecular evidence of
hybridisation in two invasive species of Pomacea (Gas-
tropoda: Ampullariidae) in Peninsular Malaysia. Raffles B
Zool 69:570–585
Kriticos DJ, Webber BL, Leriche A et al (2012) CliMond:
global high-resolution historical and future scenario cli-
mate surfaces for bioclimatic modelling. Methods Ecol
Evol 3:53–64
Lei J, Chen L, Li H (2017) Using ensemble forecasting to
examine how climate change promotes worldwide inva-
sion of the golden apple snail (Pomacea canaliculata).
Environ Monit Assess 189:404
Liu X, Zhou Y, Ouyang S, Wu X (2019) Phylogeographic pat-
terns and demographic history of Pomacea canaliculata
and Pomacea maculata from different countries (Amp-
ullariidae, Gastropoda, Mollusca). Nature Conservation
36:71
Loo SE, Keller RP, Leung B (2007) Freshwater invasions:
using historical data to analyse spread. Divers Distrib
13:23–32
Lopez Robles MA, Altaba C, Andree K, Lopez Robles V
(2010) First invasion of the Apple snail Pomacea insu-
larum in Europe. Tentacle 18:26–28
Lucero JM (2021) Regional expansion and evaluation of poten-
tial chemical control for invasive apple snails (Pomacea
maculata) in Southwest Louisiana. Louisiana State Uni-
versity and Agricultural & Mechanical College.
Lv S, Zhou X, Zhang Y etal (2006) The effect of temperature
on the development of Angiostrongylus cantonensis (Chen
1935) in Pomacea canaliculata (Lamarck 1822). Parasitol
Res 99:583–587
Lv S, Zhang Y, Liu HX etal (2013) Phylogenetic evidence for
multiple and secondary introductions of invasive snails:
pomacea species in the People’s Republic of China.
Divers Distrib 19:147–156
Marmion M, Parviainen M, Luoto M, Heikkinen RK, Thuiller
W (2009) Evaluation of consensus methods in predictive
species distribution modelling. Divers Distrib 15:59–69
Matsukura K, Wada T (2007) Environmental factors affecting
the increase of cold hardiness in the apple snail Pomacea
canaliculata (Gastropoda: Ampullariidae). Appl Entomol
Zool 42:533–539
Matsukura K, Okuda M, Cazzaniga NJ, Wada T (2013) Genetic
exchange between two freshwater apple snails, Pomacea
canaliculata and Pomacea maculata invading East and
Southeast Asia. Biol Invasions 15:2039–2048
Matsukura K, Izumi Y, Yoshida K, Wada T (2016) Cold toler-
ance of invasive freshwater snails, Pomacea canaliculata,
P. maculata, and their hybrids helps explain their different
distributions. Freshwater Biol 61:80–87
Moran EV, Alexander JM (2014) Evolutionary responses to
global change: lessons from invasive species. Ecol Lett
17:637–649
Mueck K (2017) Physiology of the invasive apple Snail, Poma-
cea maculata (Perry, 1810), in Louisiana. University of
Louisiana at Lafayette, Louisiana, the United States
Naimi B, Hamm NAS, Groen TA, Skidmore AK, Toxopeus
AG (2014) Where is positional uncertainty a problem for
species distribution modelling? Ecography 37:191–203
Oya S (1987) Overwintering of the apple snail, Pomacea cana-
liculata (Lamarck) in north Kyushu. Jpn J Appl Entomol
Z 31:206–212
Parravicini V, Azzurro E, Kulbicki M, Belmaker J (2015)
Niche shift can impair the ability to predict invasion risk

3405
Niche conservatism andgeographical range expansion ofPomacea canaliculata andPomace…
1 3
Vol.: (0123456789)
in the marine realm: an illustration using Mediterranean
fish invaders. Ecol Lett 18:246–253
Pasquevich MY, Heras H (2020) Apple snail egg perivitellin
coloration, as a taxonomic character for invasive Poma-
cea maculata and P. canaliculata, determined by a simple
method. Biol Invasions 22:2299–2307
Petersen MJ (2013) Evidence of a climatic niche shift follow-
ing North American introductions of two crane flies (Dip-
tera; genus Tipula). Biol Invasions 15:885–897
Peterson AT (2006) Uses and requirements of ecological niche
models and related distributional models
Petitpierre B, Kueffer C, Broennimann O etal (2012) Climatic
niche shifts are rare among terrestrial plant invaders. Sci-
ence 335:1344–1348
Rawlings TA, Hayes KA, Cowie RH, Collins TM (2007) The
identity, distribution, and impacts of non-native apple
snails in the continental United States. Bmc Evol Biol
7:97
Ribas LGDS, de Cássia-Silva C, Petsch DK, Silveira MJ,
Lima-Ribeiro MS (2018) The potential invasiveness of
an aquatic macrophyte reflects founder effects from native
niche. Biol Invas 20:3347–3355
Rödder D, Lötters S (2010) Explanative power of variables
used in species distribution modelling: an issue of general
model transferability or niche shift in the invasive Green-
house frog (Eleutherodactylus planirostris). Naturwissen-
schaften 97:781–796
Seuffert ME, Martín PR (2021) Exceeding its own limits: range
expansion in Argentina of the globally invasive apple snail
Pomacea canaliculata. Hydrobiologia 848:385–401
Song HM (2010) Sequencing cytochrome oxidase subunit I
of mitochondrial DNA and the taxonomic status of apple
snails. Chinese J Zool 45:1–7
Strubbe D, Broennimann O, Chiron F, Matthysen E (2013)
Niche conservatism in non-native birds in Europe: niche
unfilling rather than niche expansion. Global Ecol Bioge-
ogr 22:962–970
Strubbe D, Beauchard O, Matthysen E (2015) Niche conserva-
tism among non-native vertebrates in Europe and North
America. Ecography 38:321–329
Syobu S, Mikuriya H, Yamaguchi J et al (2001) Estimating
the overwintering mortality of the apple snail, Pomacea
canaliculata (Lamarck)(Gastropoda: Ampullariidae) in a
paddy field of Southern Japan using temperature data. Jpn
J Appl Entomol Z 45:203–207
Thornton DH, Murray DL (2014) Influence of hybridization on
niche shifts in expanding coyote populations. Divers Dis-
trib 20:1355–1364
Torres U, Godsoe W, Buckley HL etal (2018) Using niche con-
servatism information to prioritize hotspots of invasion
by non-native freshwater invertebrates in New Zealand.
Divers Distrib 24:1802–1815
Wada T, Matsukura K (2007) Seasonal changes in cold hardi-
ness of the invasive freshwater apple snail, Pomacea can-
aliculata (Lamarck) (Gastropoda: Ampullariidae). Mala-
cologia 49:383–392
Wada T, Matsukura K (2011) Linkage of cold hardiness with
desiccation tolerance in the invasive freshwater apple
snail, Pomacea canaliculata (Caenogastropoda: Ampul-
lariidae). J Mollus Stud 77:149–153
Warren DL, Glor RE, Turelli M (2008) Environmental niche
equivalency versus conservatism: quantitative approaches
to niche evolution. Evolution 62:2868–2883
Welk E (2004) Constraints in range predictions of invasive
plant species due to non-equilibrium distribution patterns:
purple loosestrife ( Lythrum salicaria) in North America.
Ecol Model 179:551–567
Yang QQ, Liu SW, He C, Yu XP (2018) Distribution and the
origin of invasive apple snails, Pomacea canaliculata and
P maculata (Gastropoda: Ampullariidae) in China. Sci
Rep 8:1185
Yang Q, He C, Liu G etal (2020a) Introgressive hybridization
between two non-native apple snails in China: widespread
hybridization and homogenization in egg morphology.
Pest Manag Sci 12:4231–4239
Yang QQ, He C, Liu GF etal (2020b) Introgressive hybridiza-
tion between two non-native apple snails in China: wide-
spread hybridization and homogenization in egg morphol-
ogy. Pest Manag Sci 76:4231–4239
Yoshida K, Matsukura K, Cazzaniga NJ, Wada T (2014) Toler-
ance to low temperature and desiccation in two invasive
apple snails, Pomacea canaliculata and P. maculata (Cae-
nogastropoda: Ampullariidae), collected in their original
distribution area (northern and central Argentina). J Mol-
lus Stud 80:62–66
Publisher’s Note Springer Nature remains neutral with regard
to jurisdictional claims in published maps and institutional
affiliations.
Springer Nature or its licensor (e.g. a society or other partner)
holds exclusive rights to this article under a publishing
agreement with the author(s) or other rightsholder(s); author
self-archiving of the accepted manuscript version of this article
is solely governed by the terms of such publishing agreement
and applicable law.