Revista de Biología Tropical, ISSN: 2215-2075, Vol. 72: e53860, enero-diciembre 2024 (Publicado Ene. 29, 2024)

Environmental sympatry through time: spatio-temporal distribution

and conservation status of two sympatric anuran species

(Leptodactylidae) in South America

Rebeca Acosta1; https://orcid.org/0000-0002-9052-3248

Facundo Alvarez*2; https://orcid.org/0000-0002-7095-9570

Betto Figueira3; https://orcid.org/0000-0001-6892-9922

Sofia Castro Cavicchini1; https://orcid.org/0000-0003-0280-9981

Rolando Vera1; https://orcid.org/0000-0003-3072-8740

Daryl D. Cruz4; https://orcid.org/0000-0002-7714-2459

Alejandro Nuñez1; https://orcid.org/0000-0003-4316-9423

1. Consejo de Investigación, Universidad Nacional de Salta, Salta, Argentina; normarebecaacosta@gmail.com;

soficastro2594@gmail.com; rolandovera824@gmail.com; alenunez.1964@gmail.com

2. Programa de Pós-graduação em Ecologia e Conservação, Campus Nova Xavantina, Nova Xavantina, Universidade do

Estado de Mato Grosso, Mato Grosso, Brasil; facualva87@gmail.com (*Correspondence)

3. Laboratório de Biologia Celular e Helmintologia, Instituto de Ciências Biológicas, Universidade Federal do Pará,

Belém, Pará, Brasil; herpetologostm@gmail.com

4. Centro de Investigación en Biodiversidad y Conservación, Universidad Autónoma del Estado de Morelos, México;

daryldavidcf@gmail.com

Received 25-I-2023. Corrected 07-VIII-2023. Accepted 18-I-2024.

ABSTRACT

Introduction: Leptodactylus latinasus and Physalaemus cuqui are sympatric anuran species with similar envi-

ronmental requirements and contrasting reproductive modes. Climatic configuration determines distribution

patterns and promotes sympatry of environmental niches, but specificity/selectivity determines the success of

reproductive modes. Species distribution models (SDM) are a valuable tool to predict spatio-temporal distribu-

tions based on the extrapolation of environmental predictors.

Objectives: To determine the spatio-temporal distribution of environmental niches and assess whether the

protected areas of the World Database of Protected Areas (WDPA) allow the conservation of these species in the

current scenario and future.

Methods: We applied different algorithms to predict the distribution and spatio-temporal overlap of environ-

mental niches of L. latinasus and P. c u q u i within South America in the last glacial maximum (LGM), middle-

Holocene, current and future scenarios. We assess the conservation status of both species with the WDPA

conservation units.

Results: All applied algorithms showed high performance for both species ( TSS = 0.87, AUC = 0.95). The L. lati-

nasus predictions showed wide environmental niches from LGM to the current scenario (49 % stable niches, 37 %

gained niches, and 13 % lost niches), suggesting historical fidelity to stable climatic-environmental regions. In the

current-future transition, L. latinasus would increase the number of stable (70 %) and lost (20 %) niches, suggest-

ing fidelity to lowland regions and a possible trend toward microendemism. P. cu q ui loses environmental niches

from the LGM to the current scenario (25 %) and in the current-future transition (63 %), increasing the envi-

ronmental sympathy between both species; 31 % spatial overlap in the current scenario and 70 % in the future.

https://doi.org/10.15517/rev.biol.trop..v72i1.53860

CONSERVATION

2Revista de Biología Tropical, ISSN: 2215-2075 Vol. 72: e53860, enero-diciembre 2024 (Publicado Ene. 29, 2024)

INTRODUCTION

Climate change, driven by anthropogenic

actions, has the potential to modify the distri-

bution patterns of biodiversity and the compo-

sition of ecological communities (Bellard et al.,

2012; Kohler et al., 2010; Menéndez-Guerrero

et al., 2020). Its effects include alterations in the

spatio-temporal distribution patterns, increases

in the probability of extinction, and declines

in species richness (Clavel et al., 2011; Pounds

et al., 2006; Stuart et al., 2004; Wake, 1991).

Therefore, climate change could alter the beta

diversity of anurans at the landscape level;

declines in anuran species richness are pro-

jected, limiting ecosystem functions (Anderson

et al., 2011; Magurran et al., 2019; Menéndez-

Guerrero et al., 2020). Climate change favors the

turnover of specialist species, with geographi-

cally restricted distribution patterns, by gen-

eralist species with wide spatial distributions,

all without generating concomitant changes in

species richness (Anderson et al., 2011; Bellard

et al., 2012; Clavel et al., 2011; Magurran et al.,

Conclusion: Extreme drought events and rainfall variations, derived from climate change, suggest the loss of

environmental niches for these species that are not currently threatened but are not adequately protected by con-

servation units. The loss of environmental niches increases spatial sympatry which represents a new challenge for

anurans and the conservation of their populations.

Key words: species distribution models; environmental sympatry; niche overlap; climate change; Leptodactylus

latinasus; Physalaemus cuqui.

RESUMEN

Simpatría ambiental a través del tiempo: distribución espacio-temporal y estado de conservación de dos

especies de anuros simpátricos (Leptodactylidae) en América del Sur

Introducción: Leptodactylus latinasus y Physalaemus cuqui son especies de anuros simpátricos con requerimien-

tos ambientales similares y modos reproductivos contrastantes. La configuración climática determina los patrones

de distribución y promueve la simpatría de los nichos ambientales, pero la especificidad/selectividad determina el

éxito de los modos reproductivos. Los modelos de distribución de especies (MDE) son una herramienta valiosa

para predecir distribuciones espacio-temporales basadas en la extrapolación de predictores ambientales.

Objetivos: Determinar la distribución espacio-temporal de los nichos ambientales y evaluar si las áreas protegi-

das de la base de Datos Mundial de Áreas Protegidas (DMAP) permiten la conservación de estas especies en el

escenario actual y futuro.

Métodos: Aplicamos diferentes algoritmos para predecir la distribución y superposición espacio-temporal de

nichos ambientales de L. latinasus y P. cu qu i dentro de América del Sur en el último máximo glacial (UGM),

Holoceno medio, actual y futuro. Evaluamos el estado de conservación de ambas especies con las unidades de

conservación de la DMAP.

Resultados: Todos los algoritmos aplicados mostraron un alto rendimiento para ambas especies ( TSS = 0.87,

AUC = 0.95). Las predicciones de L. latinasus mostraron amplios nichos ambientales desde LGM hasta el esce-

nario actual (49 % de nichos estables, 37 % de nichos ganados y 13 % de nichos perdidos), sugiriendo fidelidad

histórica por regiones climático-ambientales estables. En la transición actual-futura L. latinasus incrementaría

la cantidad de nichos estables (70 %) y perdidos (20 %), sugiriendo fidelidad por regiones de tierras bajas y la

posible tendencia hacia el microendemismo. P. c u q u i pierde nichos ambientales desde el LGM al escenario actual

(25 %) y en la transición actual-futura (63 %), incrementando la simpatría ambiental entre ambas especies; 31 %

de superposición espacial en el escenario actual y 70 % en el futuro.

Conclusión: Los eventos de sequía extrema y las variaciones de precipitaciones, derivados del cambio climático,

sugieren la pérdida de nichos ambientales para estas especies, actualmente no se encuentran amenazadas, pero

no están adecuadamente protegidas por las unidades de conservación. La pérdida de nichos ambientales aumenta

la simpatría espacial que representa un nuevo desafío para estos anuros y la conservación de sus poblaciones.

Palabras clave: modelos de distribución de especies; simpatría ambiental; superposición de nicho; cambio climá-

tico; Leptodactylus latinasus; Physalaemus cuqui.

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 72: e53860, enero-diciembre 2024 (Publicado Ene. 29, 2024)

2019). Modeling the spatio-temporal distribu-

tion patterns of species with different reproduc-

tive modes and climate dependencies will help

understand species-specific responses to global

warming, acting as a framework for the devel-

opment of conservation strategies.

The last glacial maximum (LGM) and

the middle-Holocene (MH) are synonyms of

climatic dynamics and environmental hetero-

geneity characterized by connection/discon-

nection processes between large tropical forests

(Kohler et al., 2010; Sobral-Souza et al., 2015).

These highly dynamic scenarios represented a

challenge for anuran biodiversity that experi-

enced variations in composition, abundance,

and spatio-temporal distribution patterns

(Pounds et al., 2006; Stuart et al., 2004; Wake,

1991). Although the tropical region exhibits

high abundance, richness, and endemism of

anurans, South America is the region where

this taxon is most threatened (Fouquet et al.,

2007; Menéndez-Guerrero et al., 2020; Stuart

et al., 2004; Stuart et al., 2008). Specifically, the

Neotropical region supports the greatest diver-

sity of anurans in the world, with approximately

3 000 registered species (Frost, 2022). Anurans

inhabit terrestrial and aquatic ecosystems, mak-

ing this taxon particularly sensitive to the

negative effects of climate change (Blaustein

et al., 2010; Bruzzi-Lion et al., 2019; Medina et

al., 2020; Menéndez-Guerrero et al., 2020). In

fact, in recent decades Neotropical amphibian

populations have decreased (Carey & Alexan-

der, 2003; Clavel et al., 2011; Pounds et al., 2006;

Stuart et al., 2004; Wake, 1991), with climate

change being a determining factor on differ-

ent spatio-temporal scales (Bellard et al., 2012;

Menéndez-Guerrero et al., 2020).

Niche models calculate the n-dimensional

hypervolume that defines the species’ ecologi-

cal niches from the environmental dimensions

of their conditions and resources (Hutchin-

son, 1957). Similarly, species distribution mod-

els (SDMs) allow inference of environmental

niches in past, current, and future climate

scenarios from the extrapolation of environ-

mental predictors, and the occurrence records

(Elith et al., 2006; Soberon, 2007). The family

Leptodactylidae is a broad taxon and in Argen-

tina is represented by 39 of the 232 species

described (Frost, 2022). Leptodactylus latina-

sus (Jiménez de la Espada, 1875), has a wide

geographic distribution including Argentina,

Bolivia, Paraguay, Uruguay, and southern Brazil

(Frost, 2022; Medina et al., 2020). Reproductive

activity of L. latinasus begins with the rainy

season (January-March) when temporary water

bodies are still dry (Ponssa et al., 2019). Like

many congeneric species, L. latinasus, lays its

eggs in underground chambers within foam

nests which prevent the larvae from drying

out from desiccation (Downie & Smith, 2003;

Ponssa & Barrionuevo, 2008). Larvaes hatch

inside the chamber and reach bodies of water

to complete their development (Bruzzi-Lion

et al., 2019; Ponssa & Barrionuevo, 2008). Phy-

salaemus cuqui (Lobo, 1993), is distributed in

Bolivia, Paraguay, and Argentina (Iglesias &

Natale, 2013). The reproductive activity of P.

cuqui depends strictly on rain: this species pres-

ents a marked seasonality in oviposition, which

occurs in floating foam nests inside temporary

water bodies (Ferrari & Vaira, 2001).

Although the environmental and climatic

requirements of both species are similar, the

reproductive modes have contrasting ecophysi-

ological constraints: How future climate change

could modify the spatial distribution patterns

of these species? We expect future climate

change to decrease the environmental niche

of both species, forcing the spatial distribution

patterns to stable climatic latitudes (Bruzzi-

Lion et al., 2019; Ferrari & Vaira, 2001; Medina

et al., 2020). The objectives of this research

were: 1) to determine the spatio-temporal dis-

tribution patterns of these sympatric species

with contrasting reproductive modes and, 2)

to evaluate whether the protected areas of the

World Database on Protected Areas (WDPA),

intended for the protection of biodiversity,

allow the conservation of these species in the

current scenario and in the future climatic

changes scenario.

4Revista de Biología Tropical, ISSN: 2215-2075 Vol. 72: e53860, enero-diciembre 2024 (Publicado Ene. 29, 2024)

MATERIALS AND METHODS

Study site: We delimited the study area

from the occurrence records of both species,

restricted to South America (Fig. 1). The topo-

graphic landscape is represented by floodplains

and highland ridges with air temperatures

ranging between 20-26 ºC and annual rain-

fall between 2 200 and 3 100 mm, classified

according to Köppen as Af and Am climates

(Peel et al., 2007). The study area exhibits high

spatial and environmental heterogeneity, incor-

porating different mosaics such as high-alti-

tude mountain ranges, floodplains, mangroves,

swamps, coastal regions, deserts, and savan-

nahs interconnected by hydrographic basins

and phytophysiognomic formations (Alvarez

et al., 2022).

Occurrence records: We selected the

Leptodactylus latinasus and Physalaemus

cuqui species, from the Leptodactylidae fam-

ily, because they are sympatric and although

both species present similar environmental

and climatic requirements, their reproduc-

tive modes respond differentially to climatic-

environmental conditions. The target species

are classified as “Least-concern”: L. latinasus

Fig. 1. Digital elevation model with the geographic distribution of Leptodactylus latinasus and Physalaemus cuqui in South

America. In grey scale, Kernel density map with areas of highest (white) and lowest (black) abundance density of records of

occurrence.

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 72: e53860, enero-diciembre 2024 (Publicado Ene. 29, 2024)

(Lavilla et al., 2004) and P. c u q u i (IUCN SSC

Amphibian Specialist Group, 2022). Their envi-

ronmental niches overlap, mainly, with the

Chaco region, exposed not only to anthropic

environmental simplification but also to the

cultural degradation of their traditional com-

munities (De Marzo et al., 2022). For both spe-

cies, we corroborated the existence of synonymy

and possible taxonomic identification errors. In

Fig. 2 we summarized the logic and methodol-

ogy applied in this research, from obtaining the

input data to the SDMs evaluation.

We obtained records of occurrence (OOCs)

from speciesLink (speciesLink, 2023a; species-

Link, 2023b) and the Global Biodiversity Infor-

mation Facility (GBIF) platforms. Although the

number of records does not interfere with the

algorithm’s performance (De Almeida et al.,

2010) we used all the OCCs available in GBIF

[L. latinasus (GBIF, 2023a) and P. c u q u i (GBIF,

2023b)] without applying filters of sampling

methods, sex, size, developmental stage, or spe-

cific time period. We removed all OCCs with

null or duplicate geospatial data, with errors, or

outside the South American geographic bound-

aries using QGIS software version 3.14 (QGIS

Development Team., 2022). The final OCCs for

each species totaled 164 for L. latinasus and 12

for P. c u q u i (Fig. 2A).

We executed Epanechnikov’s Kernel func-

tion, using the ‘Heatmap’ extension of the

QGIS software, with 10 000 km of bandwidth

(Fig. 1). The Kernel density estimation (KDE)

is a non-parametric weighting function that

shows the spatial heterogeneity of the occur-

rence records, indicating areas of greater or

lesser sampling effort (Silverman, 1986). We

evaluated the spatial autocorrelation (SAC) of

the occurrence records with the Moran Index.

To reduce possible side effects derived from

spatially biased sampling efforts (logistical ease

of collection, historical sampling points, preci-

sion errors, among others), we used the ‘spThin’

package (Aiello-Lammens et al., 2015). The

‘spThin’ package reduces redundant informa-

tion; removes occurrence records that could

Fig. 2. Logical scheme of the SDMs methodology applied to Leptodactylus latinasus and Physalaemus cuqui in South

America. The four main axes are A. Occurrence records (OCCs), B. Environmental predictors (EPs), C. Modeling procedure

and D. Models evaluation.

6Revista de Biología Tropical, ISSN: 2215-2075 Vol. 72: e53860, enero-diciembre 2024 (Publicado Ene. 29, 2024)

contribute to SAC, increasing SDMs perfor-

mance and prediction quality (Aiello-Lammens

et al., 2015; Guélat & Kéry, 2018). Although

both species are sympatric, the OCCs did not

show SAC effects (values less than |0.2|), so we

did not lose data.

Environmental predictors and protected

areas: We used the 19 climate variables from

the WorldClim 2.0 platform (Fick & Hijmans,

2017; Hijmans et al., 2005) as environmental

predictors (EPs). The selected EPs represent

the climatic dynamics for 1) the current cli-

mate scenario (1960-1990), 2) the last gla-

cial maximum (LGM: ~22 000 years), 3) the

middle-Holocene (MH: ~6 000 years), and 4)

the future scenario (2041–2060). We used the

CCSM4, MPI-ESM-P, and MIROC-ESM global

circulation models due to the high quality and

availability of the 19 EPs in all the climate sce-

narios evaluated (Hijmans et al., 2005; Varela

et al., 2015). Combining different climate vari-

ables, including greenhouse gas concentration

trajectories, global circulation models simulate

the atmospheric and oceanic circulation of

past and future scenarios (Hijmans et al., 2005;

Schwalm et al., 2020). Based on intensified

anthropogenic actions and the global context

of greenhouse gas emissions (Schwalm et al.,

2020), we selected the most extreme green-

house gas emission scenarios to represent the

future scenario (Representative Concentration

Paths, RCP = 8.5).

We selected all the EPs with five arcminutes

of spatial resolution (~10 km at the equator)

since this does not compromise computational

capacity and allows a correct spatio-temporal

representation of the configuration and envi-

ronmental dynamics (Barve et al., 2011). This

resolution also facilitates the EPs adjustment to

possible spatial biases of the occurrence records

and reduces the effects of bionomic factors

(Guélat & Kéry, 2018; Soberon, 2007). To carry

out the EPs standardization (z~Normal distri-

bution: mean of zero and standard deviation

of |1|) and avoid unequal weights, we applied

the packages ‘maps’ (Becker & Wilks, 1993),

‘rgdal’ (Keitt, 2010) and, ‘raster’ (Hijmans et al.,

2015) (Fig. 2B). To incorporate all the potential

environmental regions for the target species

(Barve et al., 2011), we selected the geographic

limits of South America as the modeling area.

With a clipping mask with a South American

geographic extent and using the ‘raster’ and

‘SDMTools’ packages (Vanderwal et al., 2014),

we clipped all EPs with R software (R Core

Team, 2019; Fig. 2B). We applied a princi-

pal component analysis (PCA) to reduce the

multicollinearity effects of the environmental

variables (De Marco & Nóbrega, 2018). Before

applying the PCA, for each climate scenario, we

projected the linear coefficients of the current

scenario to the past (MH and LGM) and future

climate scenarios. In each temporal scenario,

we selected the first eight axes that explained

more than 95 % of the original environmental

variation in the PEs (Fig. 2B). The presence/

absence predictions overlap the (henceforth:

binary predictions) between the climate scenar-

ios (LGM-current and current-future) allowed

us to estimate the loss, stability, and gain of

niches of both species.

To assess the conservation status of species

concerning the environmental niches obtained

for the current and future scenarios, we super-

imposed our binary predictions on the protect-

ed areas from the World Database on Protected

Areas (WDPA) (UNEP-WCMC & IUCN,

2020). To adjust the conservation units to the

target species’ ecology and distribution, we

selected only protected areas classified as strict

reserves of IUCN categories I-IV. We adjusted

the spatial resolution and geographic extent of

the WDPA mask using the same parameters

selected for the EPs. Finally, we used the binary

predictions of each climate scenario to calculate

the overlap of the environmental niches with

the protected areas (Fig. 2B, Fig. 2C).

Species distribution model: Based on the

algorithm’s performance and their flexibility

concerning the input data (Elith et al., 2006), we

applied an ensemble model approach with the

Maxent algorithms (Maximum Entropy) (Phil-

lips et al., 2006), Random Forest (RF) (James

et al., 2013), and SVM algorithms (Support

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 72: e53860, enero-diciembre 2024 (Publicado Ene. 29, 2024)

Vector Machine) (Salcedo-Sanz et al., 2014).

This approach allows a better environmental

niches approximation, summarizing in a single

output matrix the overfitting/underfitting pre-

dictions of the algorithms, and representing

the maximum/minimum thresholds of envi-

ronmental suitability (Barve et al., 2011). Start-

ing from the region of lower environmental

suitability predicted by the Bioclim algorithm,

we geographically delimited the randomly dis-

tributed pseudo-absences without spatial coin-

cidence with the OCCs (De Andrade et al.,

2020). We selected as a partition method the

OCCs division into 80 % to be used as test data

and 20 % to be used as training data for the

algorithms (Fig. 2C).

Following the parsimony principle, we set

the algorithms by default so that the predic-

tions respond to the input variables and not

to the parametric settings (Alvarez et al., 2022;

De Andrade et al., 2020). In this sense, we

configured ME using linear characteristics, and

the parameters were selected by default, with

one thousand interactions, 10 thousand back-

ground points, and logistic type output (Elith et

al., 2006). We adjusted the RF settings automat-

ically using 500 trees (Hastie et al., 2009). In the

same way, we configured SVM to default using

Kernel’s linear fitting function, with probabilis-

tic output (Fig. 2C). Following De Andrade et

al. (2020) we ran the different SDMs from the

R software using the packages ‘dismo’ (Hijmans

et al., 2017), ‘randomForest’ (Liaw & Wiener,

2002) and ‘kernlab’ (Karatzoglou et al., 2004).

We defined the randomization process to 30

replicates for each algorithm to reduce potential

correlation errors and avoid biased evaluations.

We generated the binary predictions by

cutting the environmental niche predictions for

each algorithm by the lowest presence thresh-

old (LPT) (Pearson et al., 2007). The obtained

binary predictions for each algorithm represent

the areas of presence/absence of the target spe-

cies based on the environmental niche predic-

tion matrices (Barve et al., 2011; Grenouillet

et al., 2011). Following the ensemble model

approach, we combined and concatenated the

binary predictions of all the algorithms in a sin-

gle output matrix, both for the current scenario

and for the past and future projections of both

species. For this, we used the weighted average

consensus: considering all predictions with true

skill statistics (TSS) values above the average

TSS value (Grenouillet et al., 2011).

We evaluated the performance of SDMs

using the metrics TSS and the Area Under the

Receiver-Operator Curve (AUC) (Fielding &

Bell, 1997) which attribute different weights

to prediction errors (Fig. 2D). The values of

TSS and AUC vary between -1 and 1, values

greater than 0.5 show that the predictions fit

better than the random models (Allouche et al.,

2006; Fielding & Bell, 1997). We selected the

AUC statistic because, being independent of

fixed thresholds, it allows us to assess precision

across environmental niches, discriminating

areas of omission from areas of known presence

(Fielding & Bell, 1997).

Historical niche variation: To calculate

the n-dimensional hypervolume variation of

environmental niches on a temporal scale, we

generated an environmental background with-

in South America using QGIS software. This

selection allowed us to consider regions with

suitable environmental conditions for the spe-

cies and minimize the inclusion of areas where

they might not be found due to the presence of

physical barriers or biotic interactions (Kubiak,

et al., 2017). We generated the environmental

background by randomly distributing 1 000

points for each scenario (past, current, and

future). We used the minimum-volume ellip-

soids to visualize the ecological niches in a

three-dimensional space built from the first

three principal components and with a convex

polyhedron (Qiao et al., 2016; Soberon & Naka-

mura, 2009; Van Aelst & Rousseeuw, 2009). We

conducted the analyses with NicheA, a software

for exploring and analyzing the environmental

and geographic spaces of both virtual and real

species (Qiao et al., 2016). These analyses cal-

culate the overlap of environmental niches and

allow us to evaluate possible past and future

8Revista de Biología Tropical, ISSN: 2215-2075 Vol. 72: e53860, enero-diciembre 2024 (Publicado Ene. 29, 2024)

variations concerning the current configuration

of ecological niches (Qiao et al., 2016).

RESULTS

Occurrence records and model evalua-

tion: The KDE evidence niche overlap and spa-

tial heterogeneity in sampling efforts (Fig. 1).

After applying the spatial filter from the ‘spThin’

package, we possibly reduce the residual effects

of the SAC on the OCCs, increasing the overall

performance of all applied algorithms. In fact,

all the algorithms showed good performance

(SMG 1) with high mean values in the statistics

evaluated for L. latinasus [ TSS = 0.84 ± 0.04,

AUC = 0.94 ± 0.02 (mean ± standard devia-

tion)] and for P. c u q u i ( TSS = 0.90 ± 0.12,

AUC = 0.96 ± 0.05). Although all the algo-

rithms had high performance, SVM obtained

the highest values of both AUC and TSS.

Environmental niches and binary predic-

tions: To identify environmental niches and

presence/absence areas, we visually combined

the potential environmental suitability with the

binary predictions of each temporal scenario

and for each species (Fig. 3). The environmen-

tal niches showed different spatial complexi-

ties and connectivity between both species in

each climatic scenario. However, despite the

differences in the spatio-temporal distribution

patterns, both species show geographic fidelity

associated with the central region of their his-

torical distribution (Fig. 3).

The environmental niches of L. latinasus

were geographically delimited between the par-

allels 20 °S and 40 °S of South America, more

precisely to the low altitude region of southern

Bolivia, southeastern Paraguay and Brazil, cen-

tral Argentina, and northern from Uruguay. In

the LGM (Fig. 3A) and MH scenarios (Fig. 3B),

Fig. 3. Spatio-temporal distribution patterns obtained for the species Leptodactylus latinasus (A., B., C. and D.) and

Physalaemus cuqui (E., F., G. and H.) in different temporal settings. Potential environmental niches on color scales (blue and

green) and binary predictions with absence (black color) and presence (red color).

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 72: e53860, enero-diciembre 2024 (Publicado Ene. 29, 2024)

the environmental suitability predictions of L.

latinasus suggest aggregate distribution patterns

of low spatial complexity and high connectivity

between patches. While the binary predictions

were restricted to the alluvial plains of central

Argentina and Uruguay, which, connected by a

“diagonal axis”, constitute the reported distribu-

tion center for both species. During the LGM

scenario, approximately 33 451 cells (3.4 % of

the total) were predicted for L. latinasus, and

the number of cells increased to 44 115 (4.5 %

of the total) in the MH scenario (Fig. 3). The

environmental niche of L. latinasus in the cur-

rent scenario (Fig. 3C) experienced a second

increase, reaching 46 456 cells (4.7 % of the

total) and showing distribution patterns with

high spatial connectivity and low complexity.

The diagonal axis that joins Argentina-Uruguay

could be maintained in the future scenario (Fig.

3D) with a more complex spatial pattern and

less connectivity between the patches, espe-

cially in the central region. In fact, with 40 930

cells (4.1 % of the total) in the current-future

transition, predictions for L. latinasus suggest

that this species could decrease the width of its

environmental niche (Fig. 4B).

For P. c u q ui , the environmental niches of

the LGM scenario, with 25 739 cells (2.6 % of

the total), suggest a high spatial complexity:

diffuse and disconnected patches (Fig. 3E).

Fig. 4. Spatio-temporal distribution of lost (red), gained (green), and stable (blue) environmental niches in the LGM-current

and present-future transitions for Leptodactylus latinasus (A. and B.) and Physalaemus cuqui (C. and D.). Spatial overlap of

environmental niches in the current and future scenarios for L. latinasus (E. and F.) and P. c u q u i (G. and H.) in relation to

the WDPA conservation units.

10 Revista de Biología Tropical, ISSN: 2215-2075 Vol. 72: e53860, enero-diciembre 2024 (Publicado Ene. 29, 2024)

However, the binary predictions of the LGM

predict the presence of environmental niches

along a North-South axis that perhaps crossed

the central region of Argentina latitudinally

(Fig. 3E). Although P. c u q u i maintained its

distribution in the North-South axis during the

MH, it could have experienced environmental

niche fragmentation and reduction, with iso-

lated patches totaling 25 537 cells (2.5 % of the

total). For the current (Fig. 3G) and future (Fig.

3H) scenarios, P. c u q u i continues to experi-

ence fragmentation and loss of environmental

niches. In fact, for the current scenario, with 21

726 cells (2.2 % of the total), the predictions of

environmental suitability of P. c u qui show high

spatial complexity with isolated patches that

increase the fragmentation and vulnerability of

their populations. However, unlike other sce-

narios, the binary predictions showed few iso-

lated patches, with high spatial connection and

continuity along the North-South axis. In the

future scenario, binary predictions suggest that

P. c u q u i would experience the greatest loss of its

environmental niches, decreasing to 9 228 cells

(0.9 % of the total). In this sense, climate change

will reduce the environmental niches of P. c u q u i

by 42.5 %; this spatial retraction represents 12

498 cells (57.5 % of the total).

Temporal scenarios and protected areas:

The visualization of the environmental niche

using the n-dimensional hypervolume showed

that the current environmental requirements

of L. latinasus have not varied considerably

about the LGM (SMG 2A, SMG 2B). Similarly,

from the minimum-volume ellipsoids, some

similarity is observed when comparing current

environmental conditions with future ones,

suggesting stability and environmental fidel-

ity (SMG 2C). In contrast, the n-dimensional

hypervolume of P. c u q u i showed historical fluc-

tuations between past and current environmen-

tal requirements, mainly evident during the

MH scenario (SMG 2D, SMG 2E). Although

the loss of the environmental niche is progres-

sive and continuous from the LGM to the cur-

rent (Fig. 3), the minimum-volume ellipsoids

confirm environmental niche loss for P. c u q u i

when comparing current and future scenarios

(SMG 2F).

The environmental niches in both species’

current and future scenarios (Fig. 4) show high

overlap in the minimum-volume ellipsoids

(Niche spatial overlap = 31 %). This overlap

increased considerably in the future scenario

(niche spatial overlap = 70 %), possibly due

to the increase in environmental sympatry

between both species derived from the envi-

ronmental niche reduction of P. c u q u i (Fig. 5).

For both species, the temporal transitions

LGM-current and current-future suggest dif-

ferent proportions of lost, gained, and stable

niches (Fig. 4). Temporal transitions of L. lati-

nasus showed 49 % stable niches in the LGM-

current scenario and 70 % in the current-future

scenario, suggesting high spatio-temporal fidel-

ity in these environmental regions (Fig. 4A, Fig.

4B). In the LGM-current transition (Fig. 4A),

L. latinasus lost 13 % of its niches, while in the

current-future transition it would lose 20 %

(Fig. 4B). On the contrary, for P. c u q u i , all the

temporal scenarios were negative; in all tempo-

ral transitions, the species suffered fragmenta-

tion and loss of environmental niches (Fig. 4C).

In the LGM-current transition P. c u q ui loses 25

% of its environmental niches, while in the cur-

rent-future transition, it loses 63 % (Fig. 4D).

The binary predictions overlap with the

WDPA-protected areas shows that these areas

cover a small proportion of the environmental

niches of both species, both in the current and

future scenarios (Fig. 4). For L. latinasus the

predictions obtained for the current and future

scenarios show 0.8 % overlap with protected

areas (Fig. 4E, Fig. 4F). For P. c u q u i the spatial

overlap of its predictions for the current sce-

nario was 1.1 % (Fig. 4G) and for the future, it

was 1.4 % (Fig. 4H).

DISCUSSION

The KDE function visually represented

the geographic distribution and the sampling

efforts, identifying subsampled areas of eco-

logical interest for these species and their crit-

ical niches (Hortal et al., 2015). The high

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 72: e53860, enero-diciembre 2024 (Publicado Ene. 29, 2024)

performances obtained could derive from the

input data pre-processing; the reduction of

the SAC effects on the occurrence records and

the multicollinearity decrease of the predic-

tors (Fig. 3). Even so, since the AUC statistic

is conditioned by the calibration area and

represents a source of overfitting, we comple-

ment the performance evaluation with the TSS

statistic that depends on the input data quality

(Allouche et al., 2006; Leroy et al., 2018). In the

LGM-current scenario, the predictions suggest

that L. latinasus maintained high environmen-

tal fidelity (49 %), with 37 % expansion and 13

% loss of environmental niches (Fig. 4, SMG 2).

L. latinasus could be coupled to climatic varia-

tions: 70 % of the stable niches suggest high

climatic-environmental fidelity in the current-

future transition. However, L. latinasus loses

more niches than it could gain in a climate

change scenario, suggesting restricted distribu-

tions and a tendency towards microendemism.

We estimated that from the LGM to the current

scenario, P. c u q ui gained stable environmental

niches (64 %), perhaps due to the environ-

mental expansion of its populations (Fig. 4,

SMG 2). Climate change would reduce 40 %

of the stable environmental niches of P. c u q u i ,

forcing their populations to move northward

(Fig. 4, SMG 2). Regardless of spatial sympatry

or reproductive modes, climate change could

promote microendemism and the extinction of

species currently classified as “Least-concern”,

with stable populations (Burger et al., 2019;

IUCN SSC Amphibian Specialist Group, 2022;

Lavilla et al., 2004). Spatial distribution patterns

and their temporal fluctuations vary; therefore,

it is important to consider the different tempo-

ral scenarios to manage and apply appropriate

conservation actions to each species (Alvarez et

al., 2021; Alvarez et al., 2022).

The predictions obtained for the current

scenario allow us to identify environmental

niche regions not only for these species but also

for other species of the family (Medina et al.,

2020; Soberon, 2007). These results facilitate

the objective planning of future samplings and

allow for defining management and mitiga-

tion actions in regions exposed to intensified

anthropic actions in agricultural deforestation

and mining exploitation (De Marzo et al.,

2022). Consistent with our predictions, the

family Leptodactylidae has been reported from

Fig. 5. Spatial overlap of the environmental niches obtained for Leptodactylus latinasus and Physalaemus cuqui in current

and future scenarios. The n-dimensional hypervolume of the environmental niches, defined by the environmental

variables (background points, gray color), is represented by the ellipsoids in the spatial axes (red lines) and responds to the

Jacckard index.

12 Revista de Biología Tropical, ISSN: 2215-2075 Vol. 72: e53860, enero-diciembre 2024 (Publicado Ene. 29, 2024)

the southern tip of Texas (United States), South-

ern Sonora (Mexico), and the northern Antilles

to southern regions of Brazil, Argentina, and

Chile (Frost, 2022; Medina et al., 2020). For

L. latinasus we estimate wide environmental

niches and, it is expected that in the face of

future climate change, the species will main-

tain the proportion of potential environmental

niches. This could be adjusted to the high spe-

cies’ environmental plasticity and the ability

to adapt to ecosystems with different levels of

anthropic intervention (Ponssa & Barrionuevo,

2008). Binary predictions for P. c u q u i showed

progressive declines in spatial distributions

(Fig. 3), maintaining spatial fidelity along an

imaginary north-south axis in the Argentine

lowlands. For the Yungas Biome, in the north-

ern portion of Argentina, it was estimated

that P. c u q ui would lose more than 50 % of its

environmental niches because of its specificity

and selectivity of habitats (Andrade-Díaz et al.,

2021). Although in the LGM-current transi-

tion, the number of stable niches is high, in the

current-future transition the sum between the

niches gained and stable does not exceed the

number of niches lost by P. c u q u i . Indeed, we

predicted losses in environmental niches for

the HM scenario, but the largest niche losses for

this species were predicted for the future sce-

nario (Fig. 4, SMG 2). This suggests that climate

fluctuations, such as those reported during the

HM (Kohler et al., 2010; Sobral-Souza et al.,

2015) and those predicted for the future (Bel-

lard et al., 2012; Medina et al., 2020; Menéndez-

Guerrero et al., 2020), determined and define

the spatial and demographic patterns of these

species (Andrade-Díaz et al., 2021; Blaustein et

al., 2010; Carey & Alexander, 2003).

Three points to highlight in the spatio-

temporal distribution patterns of the pre-

dicted environmental niches: first, in the

current-future transition, the availability of

environmental niches decreases. Second, the

environmental sympatry between both spe-

cies increases, possibly due to the decrease in

the environmental niche of P. c u q u i (Fig. 5).

Third, the spatial predictions were associated

with lowland regions of central Argentina (Fig.

4). Menéndez-Guerrero et al., (2020) suggest

that the regional topography is a determinis-

tic variable of changes in the anurans α and β

diversities. Tropical mountainous areas har-

bor high anuran biodiversity with restricted

distributions: rugged topography favors local

endemism (Anderson et al., 2011; Magurran et

al., 2019; Menéndez-Guerrero et al., 2020). In

contrast, the lowlands are inhabited by general-

ist species, with wide spatial distributions since

the regional topography does not represent

physical barriers to the dispersal and accumula-

tion of water (Menéndez-Guerrero et al., 2020).

In this sense, higher altitude regions favor a

decrease in β diversity and an increase in α

diversity, while lower altitude regions respond

oppositely (Anderson et al., 2011; Magurran et

al., 2019; Menéndez-Guerrero et al., 2020). Our

target species could respond to these landscape

characteristics, associating with lowlands with

historical climatic and hydrological stability

(Andrade-Díaz et al., 2021; Menéndez-Guer-

rero et al., 2020). Climate change in synergy

with anthropic actions could cause alarming

declines not only in the environmental niches

of these species but also in anuran biodiversity

and their critical habitats (Alvarez et al., 2021;

Alvarez et al., 2022; Exbrayat, 2018; Menéndez-

Guerrero et al., 2020).

In the current-future transition, both spe-

cies lose environmental niches, in parallel, the

percentages of areas designated for their protec-

tion also decrease. According to the IUCN, both

L. latinasus and P. c u q u i (IUCN SSC Amphib-

ian Specialist Group, 2022; Lavilla et al., 2004)

are classified as “Least-concern”. However, the

environmental niches for both species show a

low spatial overlap with areas designated for

conservation in the current and future scenario,

where climate change would act as an aggra-

vating factor (Fig. 4). Despite deterministic

climate change effects, the greatest risk lies in

the conservation units: few, small and discon-

nected. In the current and future scenarios L.

latinasus (with 70 % stable niches), conserves

0.8 % of its niches, while P. c u q u i (with 63

% lost niches) conserves less than 2 % of its

environmental niches (Fig. 4E, Fig. 4F). The

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 72: e53860, enero-diciembre 2024 (Publicado Ene. 29, 2024)

accumulated and intensified effects of climate

change will decrease water availability in the

coming decades, conditioning the spatio-tem-

poral distribution patterns of anurans (Medina

et al., 2020). Global warming and variations

in rainfall regimes and hydrological networks

limit the reproductive behaviors and oviposi-

tion of different species of anurans (Blaustein

et al., 2010; Bruzzi-Lion et al., 2019; Clavel et

al., 2011; Martins, 1988; Pounds et al., 2006).

This added to the regionally intensified land

use, could modify the spatial and demographic

patterns of currently stable populations (De

Marzo et al., 2022). The reproductive modes of

anurans could emerge as determinants of their

reproductive success; for most species, repro-

duction occurs during the warm season, coin-

ciding with the rainy season (Exbrayat, 2018).

With spatial hierarchy effect, global cli-

matic variables define the regional environ-

mental conditions and local biotic interactions,

establishing the environmental niches of the

species (Domisch et al., 2015). In this sense,

spatial sympatry partially responds to climatic

and environmental characteristics of shared

niches: both species couple their reproductive

activities and synchronize oviposition with the

rainy season (Ferrari & Vaira, 2001; Iglesias &

Natale, 2013; Ponssa & Barrionuevo, 2008). L.

latinasus, like other species of the genus, initi-

ates courtship activities and the reproductive

period even when the rains are not abundant

and when temporary bodies of water are dry

(Downie & Smith, 2003; Martins, 1988; Ponssa

& Barrionuevo, 2008). In turn, L. latinasus

oviposits inside foam nests that it builds in

underground chambers; this thermal insulation

increases the chances of reproductive success

(Martins, 1988, Ponssa & Barrionuevo, 2008).

The underground chambers and the foam pre-

vent larvae dehydration in case of late rains and

severe dry conditions (Downie & Smith, 2003;

Ponssa & Barrionuevo, 2008). With the rain,

the larvae hatch inside the chamber and reach

the bodies of water to complete their develop-

ment, showing tendencies towards terrestrial

life forms (Martins, 1988; Ponssa & Barrion-

uevo, 2008). The environmental and behavioral

plasticity of L. latinasus would partially argue

the stability/width of its historical environmen-

tal niches, explaining its better performance

against climate change (Bardier et al., 2014). P.

cuqui depends on frequent and intense rains,

forming vocal choruses at the beginning of

the rainy season, from inside water bodies of

variable depth and with margins covered by

herbaceous vegetation (Ferrari & Vaira, 2001;

Iglesias & Natale, 2013). The oviposition of P.

cuqui has a pattern similar to that of other spe-

cies of the genus that build foam nests on the

margins of water bodies (Ferrari & Vaira, 2001;

Acosta’s observation). The specificity/selectiv-

ity of P. c u q u i for niches with specific climatic,

edaphic, and vegetation conditions restrict the

amplitudes of its environmental niches. There

are different knowledge gaps not only about the

reproductive modes of anurans but also about

the performance of these modes against the

effects of climate change. Abrupt climatic varia-

tions could lead habitat specialist species, such

as P. c u q u i , to lose demographic and biogeo-

graphic stability, increasing their vulnerability

to extinction (Andrade-Díaz et al., 2021; Burger

et al., 2019).

The target species are currently not at

risk of extinction and, their populations are

probably stable, but climatic conditions result-

ing from climate change suggest that their

environmental niches will decline, following

the fate of other anurans (Andrade-Díaz et al.,

2021; Burger et al., 2019; Menéndez-Guerrero

et al., 2020; Pounds et al., 2006; Stuart et al.,

2004). Extreme drought events coupled with

hydrological variations could compromise their

physiological tolerance thresholds and decrease

their reproductive performances (Hijmans et

al., 2005; Menéndez-Guerrero et al., 2020). Our

results suggest that the loss of anuran environ-

mental niches could increase environmental

sympatry, pressuring spatial responses coupled

with competition; new ecophysiological barri-

ers derived from climatic and anthropic barri-

ers (Bruzzi-Lion et al., 2019; Exbrayat, 2018).

In this sense, since the period of reproductive

activity of anurans is associated with season-

ality, changes in climatic and environmental

14 Revista de Biología Tropical, ISSN: 2215-2075 Vol. 72: e53860, enero-diciembre 2024 (Publicado Ene. 29, 2024)

dynamics could modify the reproductive activi-

ties of these species (Medina et al., 2020). The

predictions identify stable historical niches;

From the LGM to the current scenario, envi-

ronmental niches are showing where species

conservation actions should be focused.

Ethical statement: the authors declare that

they all agree with this publication and made

significant contributions; that there is no con-

flict of interest of any kind; and that we fol-

lowed all pertinent ethical and legal procedures

and requirements. All financial sources are fully

and clearly stated in the acknowledgments sec-

tion. A signed document has been filed in the

journal archives.

See supplementary material

a06v72n1-MS1

ACKNOWLEDGMENTS

We thank two independent and anony-

mous reviewers who contributed significantly

to the improvement of this manuscript. We

also thank the “Universidade do Estado de

Mato Grosso” (UNEMAT - Nova Xavantina,

Brazil), the postgraduate course and the “Labo-

ratorio de Ecologia Vegetal” (LABEV) for the

support provided during the realization of

this research. Part of this research was funded

by the “Coordenação de Aperfeiçoamento de

Pessoal de NívelSuperior-Brazil” (CAPES) -

Finances Code 001.

REFERENCES

Aiello-Lammens, M. E., Boria, R. A., Radosavljevic, A.,

Vilela, B., & Anderson, R. P. (2015). spThin: an R

package for spatial thinning of species occurrence

records for use in ecological niche models. Ecography,

38(5), 541–545. https://doi.org/10.1111/ecog.01132

Allouche, O., Tsoar, A., & Kadmon, R. (2006). Assessing the

accuracy of species distribution models: prevalence,

kappa and the true skill statistic (TSS). Journal of

Applied Ecology, 43, 1223–1232. https://doi.org/10.11

11/j.1365-2664.2006.01214

Alvarez, F., da Silva, S. G., Guevara-Chumacero, L. M.,

Fernandes Ferreira, F., Alvarez Borla, L., de Sousa,

R. F., & Silva D. P. (2021). The ghost vampire: spatio-

temporal distribution and conservation status of the

largest bat in the Americas. Biodiversity and Conser-

vation, 30(14), 4359–4377. https://doi.org/10.1007/

s10531-021-02311-7

Alvarez, F., Morandi, P. S., Marimon-Junior, B. H., Exa-

vier, R., Araújo, I., Mariano, L. H., Muller, A. O.,

Feldpausch, T. R., & Marimon, B. S. (2022). Cli-

mate defined but not soil-restricted: the distribu-

tion of a Neotropical tree through space and time.

Plant and Soil, 471, 175–191. https://doi.org/10.1007/

s11104-021-05202-6

Anderson, M. J., Crist, T. O., Chase, J. M., Vellend, M.,

Inouye, B. D., Freestone, A. L., & Harrison, S. P.

(2011). Navigating the multiple meanings of β diver-

sity: a roadmap for the practicing ecologist. Ecology

Letters, 14(1), 19–28.

Andrade-Díaz, M. S., Giraudo, A. R., Marás, G. A., Didier,

K., Sarquis, J. A., Díaz-Gómez, J. M., & Prieto-Torres,

D. A. (2021). Austral Yungas under future climate and

land-use changes scenarios: the importance of pro-

tected areas for long-term amphibian conservation.

Biodiversity and Conservation, 30(12), 3335–3357.

https://doi.org/10.1007/s10531-021-02250-3

Bardier, C., Canavero, A., & Maneyro, R. (2014). Temporal

and spatial activity patterns of three species in the Lep-

todactylus fuscus group (Amphibia, Leptodactylidae).

South American Journal of Herpetology, 9(2), 106–113.

https://doi.org/10.2994/SAJH-D-13-00036.1

Barve, N., Barve, V., Jiménez-Valverde, A., Lira-Noriega, A.,

Maher, S. P., Peterson, A. T., & Villalobos, F. (2011).

The crucial role of the accessible area in ecological

niche modeling and species distribution modeling.

Ecological Modelling, 222(11), 1810–1819.

Becker, R. A., & Wilks, A. R. (1993). Maps in S. AT&T Bell

Laboratories Statistics Research Report (93.2).

Bellard, C., Bertelsmeier, C., Leadley, P., Thuiller, W., &

Courchamp, F. (2012). Impacts of climate change

on the future of biodiversity. Ecology Letters, 15(4),

365–377.

Blaustein, A. R., Walls, S. C., Bancroft, B. A., Lawler, J.

J., Searle, C. L., & Gervasi, S. S. (2010). Direct and

indirect effects of climate change on amphibian popu-

lations. Diversity, 2(2), 281–313.

Bruzzi-Lion, M., Garda, A., Lee, T. M., Bickford, D.,

Correa-Costa, G., & Fonseca, C. (2019). Global pat-

terns of terrestriality in amphibian reproduction.

Global Ecology and Biogeography, 28(6), 744–756.

Burger, J. R., Anderson, R. P., Balk, M. A., & Fristoe, T. S.

(2019). A constraint-based model of dynamic island

biogeography: environmental history and species

traits predict hysteresis in populations and communi-

ties. Frontiers of Biogeography, 11(3), e44383.

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 72: e53860, enero-diciembre 2024 (Publicado Ene. 29, 2024)

Carey, C., & Alexander, M. A. (2003). Climate change and

amphibian declines: is there a link? Diversity and Dis-

tributions, 9(2), 111–121.

Clavel, J., Julliard, R., & Devictor, V. (2011). Worldwide

decline of specialist species: toward a global functio-

nal homogenization? Frontiers in Ecology and the

Environment, 9(4), 222–228.

De Almeida, M. C., Cortes, L. G., & De Marco Junior, P. (2010).

New records and a niche model for the distribution of

two Neotropical damselflies: Schistolobos boliviensis

and Tuberculobasis inversa (Odonata: Coenagrioni-

dae). Insect Conservation and Diversity, 3(4), 252–256.

https://doi.org/10.1111/j.1752-4598.2010.00096.x

De Andrade, A. F. A., Velazco, S. J. E., & De Marco, P.

J. (2020). ENMTML: An R package for a straight-

forward construction of complex ecological niche

models. Environmental Modelling & Software, 125,

104615.

De Marco, P. J., & Nóbrega, C. C. (2018). Evaluating colli-

nearity effects on species distribution models: An

approach based on virtual species simulation. PloS

ONE, 13(9), e0202403. https://doi.org/10.1371/jour-

nal.pone.0202403

De Marzo, T., Gasparri, N. I., Lambin, E. F., & Kuemmerle,

T. (2022). Agents of forest disturbance in the Argen-

tine dry Chaco. Remote Sensing, 14(7), 1758. https://

doi.org/10.3390/rs14071758

Domisch, S., Jähnig, S. C., Simaika, J. P., Kuemmerlen, M.,

& Stoll, S. (2015). Application of species distribution

models in stream ecosystems: the challenges of spatial

and temporal scale, environmental predictors and

species occurrence data. Fundamental and Applied

Limnology, 186(1-2), 45–61. https://doi.org/0.1127/

fal/2015/0627

Downie, J. R., & Smith, J. (2003). Survival of larval Lepto-

dactylus fuscus (Anura: Leptodactylidae) out of water:

developmental differences and interspecific compari-

sons. Journal of Herpetology, 37(1), 107–115.

Elith, J., Graham, C. H., Anderson, R. P., Dudík, M., Ferrier,

S., & Guisan, A. (2006). Novel methods improve

prediction of species’ distributions from occurrence

data. Ecography, 29(2),129–151. https://doi.org/10.11

11/j.2006.0906-7590.04596

Exbrayat, J. M. (2018). From oviparity to marsupialism:

strange modes of reproduction in amphibians. Inter-

national Journal of Zoology and Animal Biology, 1(2),

000108.

Ferrari, L., & Vaira, M. (2001). Advertisement call and

breeding activity of Physalaemus cuqui (Lobo, 1993).

Herpetological Bulletin, 77, 20–22.

Fick, S. E., & Hijmans, R. J. (2017). WorldClim 2: new 1-km

spatial resolution climate surfaces for global land

areas. International Journal of Climatology, 37(12),

4302–4315.

Fielding, A. H., & Bell, J. F. (1997). A review of methods

for the assessment of prediction errors in conser-

vation presence/absence models. Environmental

Conservation, 24(1), 38–49. https://doi.org/10.1017/

S0376892997000088

Fouquet, A., Gilles, A., Vences, M., Marty, C., & Blanc,

M. (2007). Underestimation of species richness in

neotropical frogs revealed by mtDNA analyses. PloS

ONE 2(10), e1109. https://doi.org/10.1371/journal.

pone.0001109

Frost, D. R. (2022). Amphibian Species of the World: an

online reference (Version 6.1). https://amphibian-

softheworld.amnh.org/

GBIF. (2023a). GBIF Occurrence Download. https://doi.

org/10.15468/dl.94zqjf

GBIF. (2023b). GBIF Occurrence Download. https://doi.

org/10.15468/dl.xuj6h9

Grenouillet, G., Buisson, L., Casajus, N., & Lek, S. (2011).

Ensemble modelling of species distribution: the

effects of geographical and environmental ranges.

Ecography, 34(1), 9–17. https://doi.org/10.1111

/j.1600-0587.2010.06152

Guélat, J., & Kéry, M. (2018). Effects of spatial autocorrela-

tion and imperfect detection on species distribution

models. Methods in Ecology and Evolution, 9(6),

1614–1625.

Hastie, T., Tibshirani, R. & Friedman, J. (2009). The ele-

ments of statistical learning (2nd Ed). Springer.

Hijmans, R. J., Cameron, S. E., Parra, J. L., Jones, P. G., &

Jarvis, A. (2005). Very high resolution interpolated

climate surfaces for global land areas. International

Journal of Climatology, 25(15), 1965–1978. http://doi.

wiley.com/10.1002/joc.1276

Hijmans, R. J., Phillips, S., Leathwick, J., Elith, J., & Hij-

mans, M. R. J. (2017). Package ‘dismo’. Circles, 9(1),

1–68.

Hijmans, R. J., Van Etten, J., Cheng, J., Mattiuzzi, M., Sum-

ner, M., Greenberg, J. A., & Hijmans, M. R. J. (2015).

Package ‘raster’. R Package, 734.

Hortal, J., de Bello, F., Diniz-Filho, J. A. F., Lewinsohn,

T. M., Lobo, J. M., & Ladle, R. J. (2015). Seven

shortfalls that beset large-scale knowledge of bio-

diversity. Annual Review of Ecology, Evolution, and

Systematics, 46, 523–549. https://doi.org/10.1146/

annurev-ecolsys-112414-054400

Hutchinson, G. E. (1957) Concluding Remarks. Cold

Spring Harbor Symposium on Quantitative Biology,

22, 415–427.

16 Revista de Biología Tropical, ISSN: 2215-2075 Vol. 72: e53860, enero-diciembre 2024 (Publicado Ene. 29, 2024)

Iglesias, J. M. P., & Natale, G. S. (2013). Physalaemus cuqui

(Lobo, 1993). Ampliación de su distribución y confir-

mación del registro para la provincia de Santiago del

Estero, Argentina. Cuadernos de Herpetología, 27(2),

173–176.

IUCN SSC Amphibian Specialist Group. (2022). Phy-

salaemus cuqui. The IUCN Red List of Threate-

ned Species 2022: e.T57249A101431302. https://

dx.doi.org/10.2305/IUCN.UK.2020-1.RLTS.

T57249A101431302.

James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An

Introduction to Statistical Learning. Springer.

Jiménez de La Espada, M. J. (1875). Vertebrados del viaje al

Pacífico verificado de (1862) a (1865) por una comi-

sión de naturalistas enviada por el gobierno español.

Ginesta.

Karatzoglou, A., Smola, A., Hornik, K., & Zeileis, A. (2004).

Kernlab-an S4 package for kernel methods in R. Jour-

nal of Statistical Software, 11(9), 1–20.

Keitt, T. H. (2010). Rgdal: Bindings for the Geospatial Data

Abstraction Library. (Version 0.6-28, “R Package”).

http://cran.r-project.org/package=rgdal

Kohler, P., Bintanja, R., Fischer, H., Joos, F., Knutti, R.,

& Lomhann, G. (2010). What caused Earth’s tem-

perature variations during the last 800,000 years?

Data-based evidence on radiative forcing and cons-

traints on climate sensitivity. Quaternary Science

Reviews, 29(1-2), 129–45. https://doi.org/10.1016/j.

quascirev.2009.09.026

Kubiak, B. B., Gutiérrez, E. E., Galiano, D., Maestri, R., & de

Freitas, T. R. (2017). Can niche modeling and geome-

tric morphometrics document competitive exclusion

in a pair of subterranean rodents (genus Ctenomys)

with tiny parapatric distributions? Scientific Reports,

7(1), 1–13.

Lavilla, E., Heyer, R., Kwet, A., & Langone, J. (2004).

Leptodactylus latinasus. The IUCN Red List of

Threatened Species 2004: e. T57139A11590252.

https://dx.doi.org/10.2305/IUCN.UK.2004.RLTS.

T57139A11590252

Leroy, B., Delsol, R., Hugueny, B., Meynard, C. N., Bar-

houmi, C., Barbet-Massin, M., & Bellard, C. (2018).

Without quality presence–absence data, discrimina-

tion metrics such as TSS can be misleading measures

of model performance. Journal of Biogeography, 45(9),

1994–2002. https://doi.org/10.1111/jbi.13402

Liaw, A., & Wiener, M. (2002). Classification and regression

by random forest. R News, 2, 18–22.

Lobo, F. (1993). Descripción de una nueva especie del

género Physalaemus (Anura: Leptodactylidae) del

Noroeste Argentino. Revista Española de Herpetolo-

gía, 7, 13–20.

Magurran, A. E., Dornelas, M., Moyes, F., & Henderson, P.

A. (2019). Temporal β diversity. A macroecological

perspective. Global Ecology and Biogeography, 28(12),

1949–1960.

Martins, M. (1988). Biologia reprodutiva de Leptodactylus

fuscus em Boa Vista, Raraima (Amphibia: Anura).

Revista Brasileira de Biología, 48(4), 969–977.

Medina, R. G., Lira-Noriega, A., Aráoz, E., & Ponssa, M. L.

(2020). Potential effects of climate change on a Neo-

tropical frog genus: changes in the spatial diversity

patterns of Leptodactylus (Anura, Leptodactylidae)

and implications for their conservation. Climatic

Change, 161(4), 535–553.

Menéndez-Guerrero, P. A., Green, D. M., & Davies, T. J.

(2020). Climate change and the future restructuring

of Neotropical anuran biodiversity. Ecography, 43(2),

222–235.

Pearson, R. G., Raxworthy, C. J., Nakamura, M., & Town-

send, P. A. (2007). Predicting species distributions

from small numbers of occurrence records: a test

case using cryptic geckos in Madagascar. Journal of

Biogeography, 34(1), 102–117.

Peel, M. C., Finlayson, B. L., & McMahon, T. A. (2007).

Updated world map of the Köppen-Geiger climate

classification. Hydrology and Earth System Sciences,

11(5), 1633–1644.

Phillips, S. J., Anderson, R. P., & Schapire, R. E. (2006).

Maximum entropy modeling of species geographic

distributions. Ecological Modeling, 190(3), 231–259.

https://doi.org/10.1016/j.ecolm odel.2005.03.026

Ponssa, M. L., & Barrionuevo, J. S. (2008). Foam-genera-

ting behaviour in tadpoles of Leptodactylus latinasus

(Amphibia, Leptodactylidae): significance in systema-

tics. Zootaxa, 1884(1), 51–59.

Ponssa, M. L., Medina, R. G., & Vera Candiotti, M.

F. (2019). Ranita silbadora Leptodactylus latinasus.

Universo Tucumano: cómo, cuándo y dónde de la

naturaleza tucumana, contada por los lilloanos. En

G. J. Scrocchi, & C. Szumik (Eds.). Fundación Miguel

Lilo & CONICET.

Pounds, J. A., Bustamante, M. R., Coloma, L. A., Con-

suegra, J. A., Fogden, M. P., Foster, P. N., & Ron, S.

R. (2006). Widespread amphibian extinctions from

epidemic disease driven by global warming. Nature,

439(7073), 161–167.

QGIS Development Team. (2022). QGIS Geographic Infor-

mation System. Open Source Geospatial Foundation

Project. http://qgis.osgeo.org

Qiao, H., Peterson, T., Campbell, L. P., Soberon, J., Ji, L., &

Escobar, L. E. (2016). NicheA: Creating virtual species

and ecological niches in multivariate environmental

scenarios. Ecography, 39, 805–813.

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 72: e53860, enero-diciembre 2024 (Publicado Ene. 29, 2024)

R Core Team. (2019). R: A language and environment

for statistical computing (Software). R Foundation

for Statistical Computing, Vienna, Austria. http://R-

project.org

Salcedo-Sanz, S., Rojo-Álvarez, J. L., Martínez-Ramón, M.,

& Camps-Valls, G. (2014). Support vector machines

in engineering: an overview. Wiley Interdisciplinary

Reviews: Data Mining and Knowledge Discovery, 4(3),

234–267.

Schwalm, C. R., Glendon, S., & Duffy, P. B. (2020). RCP8.

5 tracks cumulative CO2 emissions. Proceedings of the

National Academy of Sciences, 117(33), 19656–19657.

Silverman, B. W. (1986). Density estimation for statistics and

data analysis. Chapman and Hall.

Soberon, J. (2007). Grinnellian and Eltonian niches

and geographic distributions of species. Eco-

logy Letters, 10(12), 1115–1123. https://doi.

org/10.1111/j.1461-0248.2007.01107.x

Soberon, J., & Nakamura, M. (2009). Niches and distribu-

tional areas: concepts, methods, and assumptions.

Proceedings of the National Academy of Sciences, 106,

19644–19650.

Sobral-Souza, T., Lima-Ribeiro, M. S., & Solferini, V. N.

(2015). Biogeography of Neotropical Rainforests: past

connections between Amazon and Atlantic Forest

detected by ecological niche modeling. Evolutionary

Ecology, 29(5), 643–655.

speciesLink. (2023a). Leptodactylus latinasus. speciesLink

Occurrence Download https://specieslink.net/search/

speciesLink. (2023b). Physalaemus cuqui. speciesLink

Occurrence Download https://specieslink.net/search/

index/

Stuart, S. N., Chanson, J. S., Cox, N. A., Young, B. E., Rodri-

gues, A. S., Fischman, D. L., & Waller, R. W. (2004).

Status and trends of amphibian declines and extinc-

tions worldwide. Science, 306(5702), 1783–1786.

Stuart, S. N., Hoffmann, M., Chanson, J. S., Cox, N. A.,

Berridge, R. J., Ramani, P., & Young, B. E. (2008).

Threatened Amphibians of the World. Lynx Editions,

IUCN & Conservation International.

UNEP-WCMC (United Nations Environment Programme

World Conservation Monitoring Centre), & IUCN

(International Union for Conservation of Nature).

(2020). Protected Planet: The World Database on Pro-

tected Areas (WDPA). UNEP-WCMC & IUCN. www.

protectedplanet.net

Van Aelst, S., & Rousseeuw, P. (2009). Minimum volume

ellipsoid. Wiley Interdisciplinary Reviews: Computa-

tional Statistics, 1, 71–82.

Vanderwal, J., Falconi, L., Januchowski, S., Shoo, L., &

Storlie, C. (2014). SDMTools: Species Distribution

Modelling Tools: Tools for processing data associated

with species distribution modelling exercises (Version

1, 1-221, “R Package”).

Varela, S., Lima-Ribeiro, M. S., & Terribile, L. C. (2015).

A short guide to the climatic variables of the gla-

cial maximum for biogeographers. PLoS ONE,

10(6), e0129037. https://doi.org/10.1371/journal.

pone.0129037

Wake, D. B. (1991). Declining amphibian populations.

Science, 253(5022), 860–861.