Received: 4 September 2024 Revised: 27 January 2025 Accepted: 28 January 2025

DOI: 10.1111/1440-1703.12551

ORIGINAL ARTICLE

Effects of elevation on taxonomic and functional diversity

of frog communities in a tropical forest

Carla C. Siqueira 1 | Davor Vrcibradic 2 | Carlos Frederico D. Rocha 1

1
Departamento de Ecologia, Universidade
do Estado do Rio de Janeiro (UERJ), Rio Abstract
de Janeiro, Brazil The assessment of taxonomic diversity (TD) and functional diversity (FD) in
2
Departamento de Zoologia, Universidade natural gradients, including elevation, has grown over the past two decades,
Federal do Estado do Rio de Janeiro
but is still relatively little studied. By studying amphibians in the Brazilian for-
(UNIRIO), Rio de Janeiro, Brazil
est areas, this research aims to address critical knowledge gaps about biodiver-
Correspondence sity distribution, while highlighting the need to conserve this fragile and
Carla C. Siqueira, Departamento de
Ecologia, Universidade do Estado do Rio
unique hotspot. Herein, we assessed whether and how environmental predic-
de Janeiro (UERJ), Rua São Francisco tors explain taxonomic and functional alpha and beta diversities of frogs in a
Xavier 524 Maracanã, Rio de Janeiro, RJ Brazilian tropical forest. Elevation significantly and negatively affected TD and
20550-013, Brazil.
Email: [email protected] FD, with a monotonic linear decrease of both variables with increasing alti-
tude. We did not detect a significant influence of habitat amount (as measured
Funding information
by proportion of forest vegetation cover) on either TD or FD, probably because
Edital Espécies Ameaçadas”, Fundação
Biodiversitas/CEPAN and RAN/ICMBio, we observed minimal variation in habitat amount along the elevational gradi-
Grant/Award Number: 0158A/012006; ent. We found a significant influence of elevation on taxonomic and functional
Conselho Nacional de Desenvolvimento
dissimilarity values, predominantly explained by the turnover rather than the
Científico e Tecnol
ogico, Grant/Award
Numbers: 302974/2015-6, 424473/2016-0, nestedness component. Our results highlight the importance of functional
304375/2020-9; Fundação Carlos Chagas traits for understanding frog distribution patterns along environmental gradi-
Filho de Amparo à Pesquisa do Estado do
Rio de Janeiro, Grant/Award Numbers: E-
ents. Community-level studies on tropical frogs, especially montane species
26/202.803/2018, E-26/201.083/2022; vulnerable to environmental and climatic impacts, remain limited, posing
Prociência and Programa de Apoio à challenges for conservation and management efforts. The high turnover of spe-
Docência (PAPD)/Universidade do Estado
do Rio de Janeiro (UERJ) cies composition and functional roles emphasizes the importance of conserv-
ing the entire mountain habitat, including the upper elevations.

KEYWORDS
amphibia, Atlantic Forest, beta diversity, conservation, turnover

1 | INTRODUCTION physiological, morphological, and ecological traits defining

a species' ecological role in a community (Magnago
Studies on functional diversity (FD) have grown exponen- et al., 2014; Petchey & Gaston, 2002). These studies have
tially over the past two decades (Mammola et al., 2021), received interest because they can provide a more mecha-
expanding our understanding of diversity and its changes nistic understanding of the relationship between diversity
in space and time (Dehling & Dehling, 2021; Magnago and ecosystem functioning (Christmann et al., 2021;
et al., 2014; Nunes et al., 2017; Pitogo et al., 2021). FD rep- Dehling et al., 2020; Thakur & Chawla, 2019).
resents the diversity of species traits influencing ecosystem Taxonomic and functional regional diversity can be
functioning (Tilman, 2001), typically focusing on the partitioned into its spatial components, local diversity

Ecological Research. 2025;1–13. wileyonlinelibrary.com/journal/ere © 2025 The Ecological Society of Japan. 1

2 SIQUEIRA ET AL.

(α-diversity) and diversity among communities (β-diver- environmental gradients have been studied (Almeida-
sity) (Almeida-Gomes et al., 2019; Kaltsas et al., 2018; Gomes & Rocha, 2015; Ferreira et al., 2016; Paz
Nunes et al., 2016, 2017). Furthermore, β-diversity can be et al., 2022). In altered landscapes, habitat amount
decoupled into turnover, that is, replacement of species (HA) negatively affects FD and functional traits of frogs,
and functional traits, and nestedness, that is, the set of and certain species and their associated traits may disap-
species and functional traits in functionally poor commu- pear from areas lacking specific habitats such as lotic
nities that are subsets of those present in functionally water bodies (Almeida-Gomes et al., 2019; Becker
richer communities (Carvalho et al., 2023; Dalmolin et al., 2010). Also, the dependence or independence on
et al., 2019; de Oliveira et al., 2023). Similar functional free water for reproduction may affect the distribution
roles can be fulfilled by different species in different patterns of frog species in different ways along an eleva-
assemblages and, thus, changes in species composition tional gradient within the forest (Siqueira et al., 2021).
between sites do not necessarily involve changes in the Moreover, a higher species richness at intermediate ele-
functional composition of the assemblages (Dehling & vations (i.e., a hump-shaped pattern) was observed for
Dehling, 2021). pond-breeding frogs along an altitudinal gradient in a
The decrease in species richness and taxonomic diver- subtropical Atlantic Forest area (Carvalho-Rocha
sity (TD) in sites at higher elevations is a consistent pat- et al., 2021). A similar pattern emerged in the TD and FD
tern found in studies on altitudinal gradients, especially of treefrogs (Hylidae) when considering occurrence data
due to their comparatively low productivity, limited across the entire Atlantic Forest domain (Paz
resource availability, small area, and harsh environmental et al., 2022). The Atlantic Forest is the biome that harbors
conditions such as low temperatures (McCain, 2009; the highest diversity of frog reproductive strategies world-
Rahbek, 1995, 2005; Terborgh, 1977). In some studies, ele- wide (Haddad & Prado, 2005; Nunes-de-Almeida
vation had a negative influence on FD in birds (Dehling et al., 2021) and contains one of the greatest concentra-
et al., 2014), ants (Reymond et al., 2013) and plants tions of threatened frog species (Luedtke et al., 2023).
(Dehling et al., 2014; Thakur & Chawla, 2019), but no Amphibians are highly vulnerable to extinction driven by
effect was observed for dung beetles (Nunes et al., 2016), climate change and land use, which are predicted to
termites (Nunes et al., 2017), or butterflies (Kaltsas cause distribution range reductions or the disappearance
et al., 2018). Thus, the nuanced intricacies underlying the of many species within the next decades (Ceron
variation of FD along elevational gradients are not yet well et al., 2023; Ramalho et al., 2021, 2023).
understood. Advancing our understanding of FD variation The present study aimed to understand how biodiver-
along elevational gradients requires integrating multiple sity patterns, both taxonomic and functional, are shaped
biodiversity dimensions and exploring how environmental along altitudinal gradients in tropical forests, providing
filtering and biotic interactions shape functional traits. insights into the ecological processes driving these
Studies on amphibians in Brazilian tropical forests, partic- changes. We studied different aspects of frog community
ularly in the Atlantic Forest, may provide critical insights diversity along a tropical altitudinal gradient in south-
into these processes due to their high degree of endemism eastern Brazil. We recorded frog species composition,
and vulnerability, contributing to both ecological theory abundances, and functional traits to examine changes in
and conservation efforts. both TD and FD along an elevational gradient. The study
Amphibians affect ecosystem structure through eco- was performed at a local scale (α-diversity) and between
logical functions such as energy flows, and, in those spe- altitudes (β-diversity). Moreover, for β-diversity, we stud-
cies having aquatic larvae in their life cycle, they can ied the spatial partitioning of TD and FD into turnover
regulate primary production in aquatic ecosystems by and nestedness components. Following the pattern
direct consumption and nutrient cycling (Costa & detected for the species richness in a previous study
Vonesh, 2013; Hocking & Babbitt, 2014; Seale, 1980). (Siqueira et al., 2021), we expected that TD would
Thus, it is important to consider life-history traits when decrease with increasing altitude. Since FD is more
assessing the distribution models of frog species across related to ecological processes (Almeida-Gomes
landscapes (Ceron et al., 2023; Duellman, 1988). Frog TD et al., 2019; Leão-Pires et al., 2018), we would expect that
and FD have been reported to vary across habitats it would also decrease with increasing altitude, but due
(Almeida-Gomes et al., 2019; Dehling & Dehling, 2021; to functional redundancy among species (Almeida-
Ernst et al., 2006; Lourenço-de-Moraes et al., 2019; Sun Gomes et al., 2019), this relationship would be less con-
et al., 2021), including along altitudinal gradients sistent than for TD. Moreover, since at higher altitudes
(Dalmolin et al., 2019; Paz et al., 2022). there are many habitat-specialist and/or microendemic
In the Brazilian Atlantic Forest, the distribution Atlantic Forest frogs (Garey & Provete, 2016; Siqueira,
patterns of frog species and their functional traits along Vrcibradic, Almeida-Gomes, et al., 2011; Siqueira,
SIQUEIRA ET AL. 3

Vrcibradic, Dorigo, & Rocha, 2011), we would expect that conservation units: the Reserva Ecolo
gica de Guapiaçu
the turnover component could act more on β-diversity (REGUA) and the Parque Estadual dos Três Picos
than the nestedness component (both for TD and FD). (PETP). The relief in the region is rugged, and its highest
Communities at high altitudes would not constitute a tax- peak reaches approximately 2360 m a.s.l. The climate of
onomic and functional subset of assemblages found at the general area is warm and humid, with annual rainfall
low elevations along the same gradient. In addition, we of 2000–2500 mm (Attias et al., 2009). Worldclim data
also analyzed whether altitude could explain the dissimi- indicate that mean annual temperatures range from 22 C
larities (in both TD and FD) among communities. at 100 m a.s.l. to 14.4 C at 1900 m a.s.l. along the eleva-
Finally, we evaluated how some functional traits vary tional gradient considered in the present study (climatic
with altitude. Thus, this study addresses some critical data are summarized in Table S2).
gaps in our knowledge of how elevational gradients influ-
ence both the composition and functionality of frog com-
munities in one of the world's most biodiverse 2.2 | Frog sampling and methods
ecosystems.
Our sampling of frogs was carried out between November
2007 and March 2010 at 14 elevational sites between
2 | MATERIALS AND METHODS 100 and 1900 m of altitude, with a total effort of 1000
person-hours of search time in 78 days of fieldwork
2.1 | Study area (a detailed description can be seen in Table S1).
Frogs were detected using time-constrained visual
We collected data in an area within the eastern portion of encounter surveys (Crump & Scott, 1994) at night (18:00–
the Serra dos Órgãos mountain range (Table S1), which 00:00 h), using headlamps. During searches, each
encloses one of the largest forest remnants of the Atlantic observer moved at a slow walking pace, carefully search-
Forest of the state of Rio de Janeiro, southeastern Brazil ing all types of potential microhabitats for anurans, such
(Figure 1). Fieldwork was carried out in two adjacent as leaf litter, water bodies (e.g., streams, marshes),

F I G U R E 1 Map showing the region of the Serra dos Órgãos mountains and the location of the sampling sites (black dots), in the state of
Rio de Janeiro (bottom left), southeastern Brazil (upper left).
4 SIQUEIRA ET AL.

vegetation (e.g., bromeliads, trees), rocks, and fallen logs mapbiomas.org/), using QGIS version 3.28.9 (QGIS
and tree trunks on forest ground. This method is Development Team, 2023). Because HA_200m (HA at
acknowledged as the best for assessing frogs in Atlantic local scale) and HA_1000m (HA at landscape scale) were
Forest areas because it enables a thorough search of positively and significantly correlated (r = 0.62,
almost all habitats available in each study site and, thus, p = 0.02), we used only HA_200m in statistical analyses.
the recording of species usually not detected by other We also obtained climatic data for the studied areas
sampling methods such as plots (quadrat sampling) and (Table S1). However, as all variables were correlated with
pitfall traps (Almeida-Gomes et al., 2010; Siqueira, Vrci- altitude (Pearson's correlation, p < 0.001 in all cases), we
bradic, Almeida-Gomes, et al., 2011; Siqueira, Vrcibradic, did not include those data in our analysis.
Dorigo, & Rocha, 2011). We collected voucher individuals
per species using a research permit previously granted by
the Instituto Chico Mendes de Conservação da Biodiver- 2.3 | Data analysis
sidade (ICMBio, permit #11701-1). Voucher specimens
were deposited at the Museu Nacional (MNRJ), in Rio de To increase the robustness of our efforts to capture the
Janeiro, Brazil (Text S1). local set of species, we pooled the data from all sampling
For each altitude, we determined the FD of frog spe- events per altitude to quantify differences in composition
cies based on three functional traits that are commonly and assess diversity. For each altitude range, we deter-
considered in studies with Neotropical anurans mined the TD and FD of the local frog assemblage. To
(Almeida-Gomes et al., 2019; Lourenço-de-Moraes measure TD, we used Hill number (calculated using the
et al., 2019): body size (snout-vent length, in mm), habit R package hillR; Chao et al., 2014), with Shannon
(arboreal, terrestrial, and rheophilic), and reproductive entropy (q = 1) (Jost, 2006). We calculated Hill–Shannon
mode (RM) (see Haddad & Prado, 2005 for categories of diversity considering that it responds strongly to both
RMs). Data were obtained and categorized mainly follow- very high and very low rarity values, emphasizes neither
ing (Haddad et al., 2013) but complemented with our rare nor common species, and has been considered a con-
own field data. RM (Table S3) is a useful variable in eco- sistent measure to estimate variation in biodiversity
logical studies using frogs as a model, because it describes (Roswell et al., 2021). The concept of biological diversity
a combination of many reproductive traits, such as type has evolved from a simple count of species to more
of reproduction (oviparous or viviparous), oviposition refined measures that are sensitive to relative abundances
habitat, spawning type, oviposition substrate, egg- (Jost, 2019).
surrounding medium, nest construction, oviposition For FD analysis, we used three traits (body size, habit,
microhabitat, embryonic development (direct or indi- and RM) for each frog species for constructing the dis-
rect), embryonic nutrition, larval and hatchling nutrition, tance matrix, as these are among the most commonly
and place of larval development (Nunes-de-Almeida used traits in analyses of anuran ecological functions
et al., 2021). We adopted the classification of Haddad and (Almeida-Gomes et al., 2019; Dehling & Dehling, 2021;
Prado (2005) for the RM of frog species, assuming that Paz et al., 2022). To create the distance matrices, we used
undescribed species potentially would have the same the method Gower distance, since we included continu-
RMs as their congeners. ous (body size) and categorical (habit and RM) variables
The proportion of vegetation cover was used as a (Mammola et al., 2021; Pavoine et al., 2009; Villéger
proxy for HA. Vegetation cover was considered as the et al., 2008). Because Gower's distance may lead to dis-
amount of native vegetation (i.e., excluding anthropized similarity with an unbalanced contribution of certain
habitats), mainly arboreal, except for the highest site traits, particularly categorical and correlated traits, we
(where shrubs and native grasses were also part of the used the “gawdis” function in the R package gawdis
cover). We included this predictor because, even though (de Bello et al., 2021). The Gawdis function uses weight-
PEPT and REGUA are protected areas, some sites within ing to minimize the differences in the correlation
them have experienced varied degrees of anthropic between the dissimilarity of each trait and provides a
disturbance (Almeida-Gomes & Rocha, 2014). HA was realistic measure of multi-trait dissimilarity between spe-
calculated from the central point of each altitudinal sam- cies in which each trait has a balanced contribution
pling site as the proportion of vegetation cover included (de Bello et al., 2021). Then, FD was calculated as the
in a buffer defined by two radii: 200 m (HA_200m, local functional dispersion index (FDis), that is, the mean dis-
scale) and 1000 m (HA_1000m, landscape scale) tance in multidimensional trait space of individual spe-
(Table S1). HA values were obtained from previously cies to the centroid of all species (Laliberté &
classified, georeferenced 30-m resolution land cover data Legendre, 2010). FDis is unaffected by species richness
(year 2007) available from MapBiomas (https://brasil. (Laliberté et al., 2014) and is not strongly influenced by
SIQUEIRA ET AL. 5

outliers (Laliberté & Legendre, 2010). Because FDis cal- The normality criterion was evaluated with the
culation requires that the trait distance matrix be Euclid- Shapiro–Wilk test in R. Data processing was performed
ean, we corrected it using the “cailliez” method in Microsoft Excel® and analysis in R version 4.1.2
(Cailliez, 1983). The “cailliez” method adjusts the dis- (R Core Team, 2021), using the RStudio interface version
tance matrix by adding a constant to all elements, ensur- 2023.6.1.524 (RStudio Team, 2023).
ing that the matrix becomes Euclidean (Cailliez, 1983).
Given that common organisms are more relevant for eco-
system services in assemblages than those rare ones, it is 3 | RESULTS
recommended to use species abundances in ecological
studies (Winfree et al., 2015). FDis was computed from We recorded 2720 individuals belonging to 63 frog species
the species-species trait distance matrix, representing the in the Serra dos Órgãos mountains (Table S4). Five frog spe-
dispersion of traits in the community (independently of cies comprised 69% of all individuals, with the most abun-
richness), accounting for species abundances, using the R dant species being Ischnocnema parva (almost 30% of
package FD (Laliberté et al., 2014). sampled frogs; N = 793 individuals), followed by Ischnoc-
The relationships of TD and FD versus altitude and nema aff. guentheri (ca. 13%; N = 344) (Figure S1). Body
HA_200m were evaluated using generalized linear size (snout-vent length) of the species varied from 9 mm
models (GLMs) with Gaussian distribution, as the resid- (Brachycephalus didactylus) to 137 mm (Rhinella icterica).
uals were normally distributed ( p > 0.05). We also evalu- The mean of body sizes was 41.7 ± 23.1 mm (range: 9.0–
ated the effects of elevation on relative species richness 137.0 mm) and varied among altitudes (Figure S2). Regard-
(expressed as proportions of the total number of species) ing habits, arboreal frogs represented 51% (32 species) of all
for some functional–trait attributes (body size, habit, and sampled species, terrestrial frogs 38% (24 species), and rheo-
RM; excluding the categories which occurred in less than philic frogs 11% (7 species). The rheophilic habit was not
half of the elevational gradient) using GLMs with Gauss- recorded at elevations of 1600 and 1900 m (Figure S3). We
ian distribution and a quadratic term being added when found a total of 16 RMs among the 63 species recorded,
the relationship was not linear. As the richness data were with “RM23 (direct development of terrestrial eggs)” being
expressed as proportions, all the values were arcsine- found in 24% of the species (N = 15 species), “RM1 (eggs
transformed prior to performing the analyses. We used and exotrophic tadpoles in lentic water)” in 17% (11 species),
“glm” function in the R package stats (R Core and “RM2 (eggs and exotrophic tadpoles in lotic water)” in
Team, 2021). 14% (9 species). Also, “RM23” was the only RM present at
We examined the relative importance of turnover and all altitudes (Figure S4).
nestedness to total taxonomic and functional Values of TD (Table S5) were highest at 500 m (10.6)
dissimilarities (β-diversities) between altitudes. We used and 600 m (10.2) of elevation, whereas FD was highest at
presence-absence data to obtain the dissimilarities 100 m (6.3). Both TD (1.4) and FD (0.7) were lowest
(Jaccard family). We used “as.phylo” function in the ape at 1600 m. Elevation significantly and negatively affected
R package (Paradis et al., 2004) to convert UPGMA den- TD (p < 0.001, adjR2 = 0.76) and FD ( p < 0.05,
drograms into tree objects. We assessed taxonomic adjR2 = 0.22), with a monotonic linear decrease of both
(“beta.multi” function) and functional (“phylo.beta. variables with increasing altitude (Figure 2). We observed
multi” function) dissimilarities (βJAC), including the con- minimal variation in HA (VC = 0.88–1.00 within a 200 m
tributions of turnover (βJTU) and nestedness (βJNE), using radius) along the elevational gradient, with the lowest
the betapart package (Baselga et al., 2023). To analyze values of vegetation cover at 100 m (0.91) and 1900 m
the extent of the effect of elevation on the taxonomic and (0.88) of elevation (Table S1). We did not detect a signifi-
functional dissimilarities (βJAC), we used distance-based cant influence of HA on either TD ( p = 0.648) or FD
redundancy analysis (db-RDA) on each dissimilarity (p = 0.205).
matrix (TD and FD), using the function “dbrda” of the R We evaluated the variation in 10 frog functional trait
package vegan. db-RDA is an ordination method similar attributes (i.e., those which occurred in at least half of
to redundancy analysis, but it allows the determination the elevational gradient). The proportion of small
of non-Euclidean dissimilarity indices (Legendre & (<30 mm of body size) and medium-sized (30–100 mm)
Anderson, 1999). The significance of the models was frogs was not significantly influenced by elevation
tested through an ANOVA-like permutation test, with (Table 1). In terms of habitat, the proportion of rheophi-
999 permutations, using the “anova.cca” function of the lic species significantly decreased with altitude, while the
vegan package in the R. If statistically significant, we ran opposite pattern was observed for the terrestrial frogs
the adjusted-R2 (adjR2) in “RsquareAdj” function of the (Table 1 and Figure 3). The proportion of frog species
vegan package (Oksanen et al., 2022). with RMs such as RM1 and RM19 decreased with
6 SIQUEIRA ET AL.

F I G U R E 2 Relationship between
taxonomic (a) and functional (b) frog
diversity and altitude (in meters) in the
Serra dos Órgãos mountains,
southeastern Brazil. The fitted values
(black lines) with 95% confidence
intervals (gray area) were generated
from the generalized linear models
(GLMs) with Gaussian-distributed
residuals.

T A B L E 1 Results of generalized linear models (GLMs) of the intermediate elevations, especially at 500 and 600 m of
effects of elevation on relative species richness (proportion in altitude (Table 1 and Figure 4).
relation to the total number of species) for 10 functional traits.
We found a significant influence of elevation on taxo-
Functional nomic (F1,13 = 3.09; p = 0.001; adjR2 = 0.14) and func-
traits Results tional (F1,13 = 3.24; p = 0.001; adjR2 = 0.15) dissimilarity
Body size values (Table S6). In general, the differences in species
<30 mm E = 0.0001 (0.00006), p = 0.14 composition were greater than those in functional traits
between the same pairs of altitudes (Figure 5). The total
30–100 mm E= 0.0001 (0.00005), p = 0.08
taxonomic and functional dissimilarities were predomi-
Habitat
nantly explained by the turnover rather than the nested-
Arboreal E= 0.000004 (0.00006), p = 0.95 ness component (Figure 6).
Rheophilic E = 0.0002 (0.00005), p = 0.0003*,
adjR2 = 0.64
Terrestrial E = 0.0001 (0.00003), p = 0.02*, 4 | DISCUSSION
adjR2 = 0.31
Reproductive We observed a consistent pattern of variation in TD along
modes the altitudinal gradient, characterized by a monotonic
RM1 E = 0.0002 (0.00006), p = 0.003*, decrease, as expected. A previous study based on the
adjR2 = 0.49 same frog species' dataset demonstrated a linear decline
RM2 E = 0.000009 (0.00008), p = 0.92 in species richness (based on interpolated occurrence
RM3 E ≤ 0.00001 (<0.00001), p = 0.009*, data for all species) with increasing altitude (Siqueira
adjR2 = 0.46 et al., 2021). By including relative abundance data into
RM19 E = 0.0002 (0.00004), p = 0.00004*, analyses, we observed a substantial increase in explana-
adjR2 = 0.74 tory power (reaching 76%), surpassing the 56% observed
RM23 E = 0.0002 (0.00003), p = 0.00002*, when the community composition was based solely on
adjR2 = 0.77 occurrence data (Siqueira et al., 2021). Other studies have
also found lower frog species richness/TD at higher alti-
Note: RM1: eggs and exotrophic tadpoles in lentic water. RM2: eggs and
exotrophic tadpoles in lotic water. RM3: eggs and early larval stages in
tudes (Pitogo et al., 2021; Poynton, 2003; Qian
constructed subaquatic chambers; exotrophic tadpoles in streams. RM19: et al., 2007; Zancolli et al., 2014), but different results
eggs on humid rocks, in rock crevices, or on tree roots above water; have also been found depending on taxon, guild, and/or
exotrophic semiterrestrial tadpoles living on rocks and rock crevices in a spatial scale (Fu et al., 2006; Goyannes-Araujo
water film or in the water–land interface. RM23: direct development of
terrestrial eggs. Values show coefficient estimates (E) with standard errors
et al., 2015; Hutter et al., 2017; Naniwadekar &
(in parentheses), p-values, and adjusted-R2 (adjR2) for significant Vasudevan, 2007). At high elevations, the complexity of
relationships. Asterisk indicates significant p-values. Negative parameters forest structure typically diminishes in the Atlantic For-
indicate negative effects. est, with marked changes in vegetation structure
(Nettesheim et al., 2018; Rezende et al., 2015). Also, com-
altitude, whereas RM23 had the highest proportion of munity species compositions are filtered under severe
species at the highest altitudes (Table 1 and Figure 4). environmental conditions, which are common at high
The proportion of species with RM3 peaked at altitudes (Reymond et al., 2013).
SIQUEIRA ET AL. 7

F I G U R E 3 Relationship between
the relative species richness of
rheophilic (a) and terrestrial (b) frogs
(expressed in terms of proportions;
values arcsine-transformed) and altitude
(in meters) in the Serra dos Órgãos
mountains, southeastern Brazil. The
fitted values (black lines) with 95%
confidence intervals (gray area) were
generated from the generalized linear
models (GLMs) with Gaussian-
distributed residuals.

F I G U R E 4 Significant relationships
between proportion of frog species
(values arcsine-transformed)
representing some reproductive modes
and altitude (in meters) in the Serra dos
Órgãos mountains, southeastern Brazil.
The fitted values (black lines) with 95%
confidence intervals (gray area) were
generated from the generalized linear
models (GLMs) with Gaussian-
distributed residuals (with a quadratic
term being added in ‘b”). (a) RM1: eggs
and exotrophic tadpoles in lentic water.
(b) RM3: eggs and early larval stages in
constructed subaquatic chambers;
exotrophic tadpoles in streams.
(c) RM19: eggs on humid rocks, in rock
crevices, or on tree roots above water;
exotrophic semiterrestrial tadpoles
living on rocks and rock crevices in a
water film or in the water–land
interface. (d) RM23: direct development
of terrestrial eggs.

We also detected a negative relationship between FD FD values for dung beetles (Nunes et al., 2016) and ter-
and altitude. A similar trend was observed in bird assem- mites (Nunes et al., 2017), however, showed no signifi-
blages from the Ecuadorian Andes and from Mount Kili- cant relationship with elevation. Similarly, no
manjaro (Tanzania), where FD consistently decreased relationship was observed between FD and altitude for
with increasing elevation in both regions, primarily butterflies in the mountains of Greece (Kaltsas
driven by the availability of food resources (Hanz et al., 2018). Thus, there does not seem to be a wide-
et al., 2019). A trend in the decrease of FD with the spread pattern of either increase or decrease in FD along
increase in altitude was also found for ants along an alti- altitude. This has been interpreted as indicative that envi-
tudinal gradient (1800–2550 m) in the Swiss Alps ronmental conditions may filter species in terms of the
(Reymond et al., 2013). Regarding amphibians, another Grinnellian niche (the range of each biotic and abiotic
study found a reduction in FD (and also in TD) with alti- factor determining the distribution and abundance of a
tude in hylid treefrogs of the Brazilian Atlantic Forest single species) but not in terms of the Eltonian niche (the
(Paz et al., 2022). In a tropical montane habitat in Brazil, species' position in ecological networks) (Devictor
8 SIQUEIRA ET AL.

high heterogeneity of habitat in the lowlands because

there are various types of habitats, including streams
inside the forest, which are surrounded by a matrix con-
taining lentic water bodies such as ponds or wetlands
(Almeida-Gomes et al., 2016). At the other extreme, the
lowest FD value was observed at 1600 m, where the rela-
tive abundance of the most common frog species
(N = 239 individuals) is much higher than that of the
others (≤4 individuals each; Table S1), and most of
the species present were functionally similar. At the high-
est altitudes sampled (1600 and 1900 m), frogs experience
cooler temperatures (e.g., from 100 to 1900 m, the mean
temperature decreases by almost 8 C) and limiting water
availability, particularly streams, puddles, and marshes.
F I G U R E 5 Relationship between taxonomic and functional These factors act as environmental filters for many frog
dissimilarity values (β-diversity) for frog assemblages calculated for species (Paz et al., 2022). Interestingly, at the highest
each pair of altitudes and elevational distance (m) along an sampled altitude (1900 m) FD was unexpectedly high, in
altitudinal gradient in the Serra dos Órgãos mountains, spite of the low species richness (only four species); this
southeastern Brazil.
result probably reflects the fact that these four species
occurred with low abundances and were representative
of three different RMs (RM23, RM36, and RM37). For
dung beetle communities in Brazil, almost 100% of func-
tional diversity was due to the α-component (Nunes
et al., 2016). In this case, low values of FD and the
absence of a significant relationship can be attributed to
ecological redundancy, stemming from the widespread
presence and dominance of taxonomically close and, con-
sequently, functionally similar species across the entire
altitudinal gradient.
While the proportion of rheophilic species tended to
decrease with altitude, the opposite pattern was observed
for the terrestrial frogs. Furthermore, the proportion of
F I G U R E 6 Contributions of turnover and nestedness for
taxonomic and functional β-diversity of frogs along an altitudinal
frog species with some RMs dependent on water for
gradient in the Serra dos Órgãos mountains, southeastern Brazil. reproduction decreased with altitude (RM1, RM19) or
The respective values, in proportion, are shown next to the bars. peaked at intermediate elevations (RM3), whereas a RM
featuring direct development of terrestrial eggs (RM23)
had the highest proportion of species at the highest alti-
et al., 2010). Our study also emphasizes that the distribu- tudes. The proportional increase in the richness of terres-
tion of frog species is significantly influenced by environ- trial species with elevation appears to reflect mainly the
mental factors correlated with altitude (see Table S2). proportional decrease in rheophilic species (rather than
Nonetheless, the observed variation in FD along altitude, an actual increment in the number of terrestrial ones), as
albeit modest (explaining 22% of the variance), highlights evidenced by the stronger negative effect of elevation in
the importance of considering the functional roles of frog the latter (R2 = 0.54) compared to its positive effect in
species in ecological networks along an altitudinal gradi- the former (R2 = 0.31). The occurrence and availability of
ent in tropical forests. streams tend to decrease at higher altitudes, which
At lower altitudes (100–300 m), we found the highest restricts habitats to species depending on streams to live
values of FD. In these areas, in general, Adenomera mar- and reproduce. Our casual observations during fieldwork
morata, Hylodes pipilans, Ischnocnema aff. guentheri, Olo- suggest that lentic bodies of water in our study area
lygon humilis, Physalaemus signifer, and Rhinella ornata become rare above 400 m a.s.l., whereas both lentic and
were the most abundant frogs (Figure S1). These six spe- lotic habitats were absent from the two highest sites, at
cies differ in their functional traits, including RMs, 1600 and 1900 m a.s.l. (see Siqueira et al., 2021). In the
habits, and body sizes, which contribute to the observed Philippines, altitude also had a strong effect on frog func-
high values of FD. This result seems to be an effect of the tional traits, particularly those related to habitat use,
SIQUEIRA ET AL. 9

larval ecology, and body size (Pitogo et al., 2021). Species sensitivity to forest loss, resulting in an ordered loss of
undergoing direct development were positively associ- sets of species and traits (Almeida-Gomes et al., 2019).
ated, while those with exotrophic tadpole development in Both TD and FD are expected to be correlated in
lotic habitats were negatively associated with altitude amphibians, as closely related species tend to share simi-
(Pitogo et al., 2021). Likewise, in a montane region lar ecological traits due to phylogenetic conservatism
within Amazonia, some functional groups of frogs, such (Campos et al., 2019). Thus, it is not surprising that in
as large-bodied species, terrestrial, arboreal, and pond our study both TD and FD were similarly affected by alti-
and stream breeders, were absent or almost absent from tude. However, patterns of TD and FD across the gradi-
higher elevations (Villacampa et al., 2019). Environmen- ent were not entirely congruent, with TD having
tal factors thus seem to play a crucial role in explaining maximal values at intermediate altitudes and FD at the
the variation of functional traits along altitudinal gradi- lowest altitude. Phenomena like ecological convergence
ents. Incorporating functional traits in analyses of and adaptive radiation may cause some dissociation
richness-elevation relationships can be essential for between phylogenetic and functional aspects of anuran
understanding species distribution patterns and associ- diversity, a result that stresses the importance of consid-
ated mechanisms. ering both TD and FD in analyses of the influence of
We found a significant influence of altitude on both environmental gradients in anuran communities.
taxonomic and functional dissimilarities of frog assem- We detected little variation in the amount of habitat
blages. Likewise, elevation explained taxonomic and func- along the elevational gradient in the Serra dos Órgãos moun-
tional turnover in phyllostomid bats in Amazonia tains, with relatively low impact on forest cover detected at
(Carvalho et al., 2023). Also, in a South American subtrop- the lowest (100 m) and highest (1900 m) altitudes. Therefore,
ical landscape, whereas the taxonomic beta diversity of we did not detect a significant influence of vegetation cover
pond frogs was influenced by environmental filters, the on TD and FD. In the lowlands adjacent to the Serra dos
dissimilarity in traits between sites was explained by spa- Órgãos mountains, the forest is bordered by anthropically
tial distances (Dalmolin et al., 2019). We noted greater tax- modified areas, mostly by cattle ranching and agriculture.
onomic dissimilarity compared to functional dissimilarity The impact of fragmentation in the lowlands affects TD and
when examining similar altitudinal distances (see FD of frogs across the landscape, as previously mentioned
Figure 5). This points to a potential redundancy in the (Almeida-Gomes et al., 2019; Almeida-Gomes &
functional roles within frog assemblages along the eleva- Rocha, 2014). In our study, the sampling sites were located
tional gradient, suggesting that some species may not nec- within conservation units, so the degree of anthropic distur-
essarily be replaced by others with distinct characteristics. bance was probably not extensive enough to have a signifi-
The overall taxonomic and functional dissimilarities cant effect on the dynamics of local frog assemblages.
were mainly explained by the turnover component. Some of the species sampled during the study are con-
Therefore, poor communities in terms of taxonomic and sidered to be under some degree of threat according to
functional diversity were not subsets of richer communi- the classification system of the International Union for
ties. Instead, they presented distinct species compositions the Conservation of nature (IUCN): Dendrophryniscus
and functional attributes. This result was similar to that davori (as critically endangered—CR), Boana secedens,
found for dung beetles in another mountainous area in Brachycephalus garbeanus, and Cycloramphus brasiliensis
Brazil (Nunes et al., 2016). Conversely, for butterflies (all as endangered—EN). Of these, one (B. secedens) is a
(Kaltsas et al., 2018) and termites (Nunes et al., 2017), TD lowland species, another (C. brasiliensis) is found mainly
was predominantly explained by turnover, while FD was at mid-elevations, and the other two are associated with
associated with nestedness. Other taxonomic and func- higher altitudes (>1500 m). This reinforces the relevance
tional dissimilarities among frog assemblages across envi- of the study area for the conservation of Atlantic Forest
ronmental gradients have been studied in the anurans and the importance of preserving the environ-
Neotropical region. In a mosaic of grasslands and Arau- ment along the entire altitudinal gradient.
caria Forest in Brazil, tadpole dissimilarity was explained
by species replacement associated with pond area
(Knauth et al., 2018). In the Atlantic Forest, the nested- 5 | CONCLUSIONS
ness component appears to drive metacommunities sub-
jected to an environmental gradient of disturbance Our results highlight the importance of functional traits
intensity. In a fragmented forest landscape, functional for understanding frog distribution patterns along envi-
dissimilarity in frogs was primarily attributed to nested- ronmental gradients. Since not all ecological functions
ness, wherein forest-dependent species, particularly those can be fulfilled in all local species assemblages, changes
closely associated with streams, exhibited greater in species composition can be accompanied by alterations
10 SIQUEIRA ET AL.

in species' functional roles. This results in a higher diver- RE FER EN CES

sity of functions on the landscape scale. The high turn- Almeida-Gomes, M., & Rocha, C. F. D. (2014). Landscape connec-
over of species composition and functional roles tivity may explain anuran species distribution in an Atlantic
emphasizes the importance of conserving the entire forest fragmented area. Landscape Ecology, 29(1), 29–40.
mountain habitat, including the upper elevations, thus https://doi.org/10.1007/s10980-013-9898-5
Almeida-Gomes, M., & Rocha, C. F. D. (2015). Habitat loss reduces
ensuring that the preserved area encompasses the
the diversity of frog reproductive modes in an Atlantic Forest
restricted habitats of frogs with more specialized ecologi- fragmented landscape. Biotropica, 47(1), 113–118. https://doi.
cal attributes. A total of 84% of the frog species recorded org/10.1111/btp.12168
in our surveys are forest specialists (Table S4), which typ- Almeida-Gomes, M., Rocha, C. F. D., & Vieira, M. V. (2016).
ically have more restricted habitat requirements com- Local and landscape factors driving the structure of tropical
pared to taxonomically related generalist species anuran communities: Do ephemeral ponds have a nested
(Almeida-Gomes et al., 2019). Community-level studies pattern? Biotropica, 48(3), 365–372. https://doi.org/10.1111/
on Brazilian frogs remain limited, especially for montane btp.12285
Almeida-Gomes, M., Vieira, M. V., Rocha, C. F. D., & Melo, A. S.
species vulnerable to the impacts of environmental and
(2019). Habitat amount drives the functional diversity and nest-
climatic changes (Ramalho et al., 2021, 2023), posing edness of anuran communities in an Atlantic Forest fragmen-
challenges for conservation and management efforts. ted landscape. Biotropica, 51(6), 874–884. https://doi.org/10.
1111/btp.12687
A C K N O WL E D G M E N T S Almeida-Gomes, M. A., Almeida-Santos, M., Goyannes-Araújo, P.,
We thank the Instituto Estadual do Ambiente (INEA; Borges-Júnior, V. N. T., Vrcibradic, D., Siqueira, C. C.,
permit #005/2008) and the Instituto Chico Mendes de Ariani, C. V., Dias, A. S., Souza, V. V., Pinto, R. R., Van
Sluys, M., & Rocha, C. F. D. (2010). Anurofauna of an Atlantic
Conservação da Biodiversidade (ICMBio; permit
Rainforest fragment and its surroundings in Northern Rio de
#11701-1) for permission to work at the Parque Estadual
Janeiro State, Brazil. Brazilian Journal of Biology, 70(3),
dos Três Picos (PETP), and Nicholas Locke for providing 871–877.
infrastructure, logistic support, and permission to work Attias, N., Raíces, D. S. L., Pessoa, F. S., Albuquerque, H., Jordão-
in the Reserva Ecol
ogica de Guapiaçu (REGUA). We also Nogueira, T., Modesto, T. C., & Bergallo, H. G. (2009). Potential
thank Adriano Luz, Alexander Jan Davis, and Moisés distribution and new records of Trinomys species (Rodentia:
Vasconcelos for providing infrastructure and local sup- Echimyidae) in the state of Rio de Janeiro. Zoologia, 26(2),
port during fieldwork at sites within the PETP. We would 305–315.
Baselga, A., Orme, D., Villeger, S., Bortoli, J., Leprieur, F.,
also like to thank the many colleagues who helped us
Logez, M., Martinez-Santalla, S., Martin-Devasa, R., Gomez-
during fieldwork, and J. P. Pombal Jr., C. A. G. Cruz, Rodriguez, C., Crujeiras, R. M., & Henriques-Silva, R. (2023).
A. C. C. Lourenço, I. Nunes, M. Targino (Museu Nacio- Package “betapart”: Partitioning beta diversity into turnover and
nal, Rio de Janeiro), and C. Canedo (UERJ) for helping nestedness components. v. 1.6. http://www.pwrc.usgs.gov/BBS/
us with the identification of frog species. Becker, C. G., Fonseca, C. R., Haddad, C. F. B., & Prado, P. I.
(2010). Habitat split as a cause of local population declines of
FUNDING INFORMATION amphibians with aquatic larvae: Contributed paper. Conserva-
This study was supported by the Conselho Nacional de tion Biology, 24(1), 287–294. https://doi.org/10.1111/j.1523-
1739.2009.01324.x
Desenvolvimento Científico e Tecnolo
gico (CNPq)
Cailliez, F. (1983). The analytical solution of the additive constant
(Grant/Award Numbers: 302974/2015-6, 424473/2016-0, problem. Psychometrika, 48(2), 305–308.
304375/2020-9); Cientistas do Nosso Estado Program, Campos, F. S., Lourenço-De-Moraes, R., Rudoy, A., Rödder, D.,
Fundação Carlos Chagas Filho de Amparo à Pesquisa do Llorente, G. A., & Solé, M. (2019). Ecological trait evolution in
Estado do Rio de Janeiro (FAPERJ) (Grant/Award Num- amphibian phylogenetic relationships. Ethology Ecology & Evo-
bers: E-26/202.803/2018, E-26/201.083/2022); Prociência lution, 31(6), 526–543. https://doi.org/10.1080/03949370.2019.
and Programa de Apoio à Docência (PAPD)/Universi- 1630012
Carvalho, W. D., Fluck, I. E., de Castro, I. J., Hil
ario, R. R.,
dade do Estado do Rio de Janeiro (UERJ); “Edital Espé-
Martins, A. C. M., de Toledo, J. J., da Silva Xavier, B.,
cies Ameaçadas,” Fundação Biodiversitas/CEPAN and
Dambros, C., & Bobrowiec, P. E. D. (2023). Elevation drives tax-
RAN/ICMBio(Grant/Award Numbers: 0158A/012006). onomic, functional and phylogenetic β-diversity of phyllosto-
mid bats in the Amazon biome. Journal of Biogeography, 50(1),
CONFLICT OF INTEREST STATEMENT 70–85. https://doi.org/10.1111/jbi.14533
The authors declare no conflicts of interest. Carvalho-Rocha, V., Peres, C. A., & Neckel-Oliveira, S. (2021). Hab-
itat amount and ambient temperature dictate patterns of
ORCID anuran diversity along a subtropical elevational gradient. Diver-
sity and Distributions, 27(2), 344–359. https://doi.org/10.1111/
Carla C. Siqueira https://orcid.org/0000-0002-5518-
ddi.13187
7090
SIQUEIRA ET AL. 11

Ceron, K., Sales, L. P., Santana, D. J., & Pires, M. M. (2023). Applied Ecology, 47(1), 15–25. https://doi.org/10.1111/j.1365-
Decoupled responses of biodiversity facets driven from anuran 2664.2009.01744.x
vulnerability to climate and land-use changes. Ecology Letters, Duellman, W. E. (1988). Patterns of species diversity in anuran
26(6), 869–882. https://doi.org/10.1111/ele.14207 amphibians in the American Tropics. Annals of the Missouri
Chao, A., Chiu, C.-H., & Jost, L. (2014). Unifying species diversity, Botanical Garden, 75(1), 79–104. https://doi.org/10.2307/
phylogenetic diversity, functional diversity, and related similar- 2399467
ity and differentiation measures through Hill numbers. Annual Ernst, R., Linsenmair, K. E., & Rödel, M. O. (2006). Diversity ero-
Review of Ecology, Evolution, and Systematics, 45(1), 297–324. sion beyond the species level: Dramatic loss of functional diver-
https://doi.org/10.1146/annurev-ecolsys-120213-091540 sity after selective logging in two tropical amphibian
Christmann, T., Rosado, B. H. P., Delhaye, G., Matos, I. S., communities. Biological Conservation, 133(2), 143–155. https://
Drummond, J. S., Roland, H. L., Moraes, Y. C., & Oliveras doi.org/10.1016/j.biocon.2006.05.028
Menor, I. (2021). Functional assembly of tropical montane tree Ferreira, R. B., Beard, K. H., & Crump, M. L. (2016). Breeding guild
islands in the Atlantic Forest is shaped by stress tolerance, determines frog distributions in response to edge effects and
bamboo presence, and facilitation. Ecology and Evolution, habitat conversion in the Brazil's Atlantic Forest. PLoS One,
11(15), 10164–10177. https://doi.org/10.1002/ece3.7824 11(6), e0156781. https://doi.org/10.1371/journal.pone.0156781
Costa, Z. J., & Vonesh, J. R. (2013). Interspecific differences in the Fu, C., Hua, X., Li, J., Chang, Z., Pu, Z., & Chen, J. (2006). Eleva-
direct and indirect effects of two neotropical hylid tadpoles on tional patterns of frog species richness and endemic richness in
primary producers and zooplankton. Biotropica, 45(4), 503–510. the Hengduan Mountains, China: Geometric constraints, area
https://doi.org/10.1111/btp.12032 and climate effects. Ecography, 29(6), 919–927. https://doi.org/
Crump, M. L., & Scott, N. J., Jr. (1994). Visual encounter surveys. In 10.1111/j.2006.0906-7590.04802.x
W. R. Heyer, M. A. Donnely, W. M. Roy, L. C. Hayek, & M. S. Garey, M. V., & Provete, D. B. (2016). Species composition, conservation
Foster (Eds.), Measuring and monitoring biological diversity: status, and sources of threat of anurans in mosaics of highland
Standard methods for amphibians (1st ed., pp. 84–92). Smithso- grasslands of southern and southeastern Brazil. Oecologia Australis,
nian Institution Press. 20(2), 94–108. https://doi.org/10.4257/oeco.2016.2002.07
Dalmolin, D. A., Tozetti, A. M., & Pereira, M. J. R. (2019). Taxo- Goyannes-Araujo, P., Siqueira, C. C., Laia, R. L., Almeida-
nomic and functional anuran beta diversity of a subtropical Santos, M., Guedes, D. M., & Rocha, C. F. D. (2015). Anuran
metacommunity respond differentially to environmental and species distribution along an elevational gradient and seasonal
spatial predictors. PLoS One, 14(11), e0214902. https://doi.org/ comparisons of leaf litter frogs in an Atlantic Rainforest area of
10.1371/journal.pone.0214902 southeastern Brazil. The Herpetological Journal, 25(2), 75–81.
de Bello, F., Botta-Duk
at, Z., Lepš, J., & Fibich, P. (2021). Towards Haddad, C. F. B., & Prado, C. P. A. (2005). Reproductive modes of
a more balanced combination of multiple traits when comput- the Atlantic forest frogs. Bioscience, 55(3), 207–217.
ing functional differences between species. Methods in Ecology Haddad, C. F. B., Toledo, L. F., Prado, C. P. A., Loebmann, D.,
and Evolution, 12(3), 443–448. https://doi.org/10.1111/2041- Gasparini, J. L., & Sazima, I. (2013). Guia dos Anfíbios da Mata
210X.13537 Atlântica: Diversidade e Biologia. Anolis Books.
de Oliveira, A. B. d. S., Borges, S. H., Paes, A. T., Pereira, R. C. d. S., Hanz, D. M., Böhning-Gaese, K., Ferger, S. W., Fritz, S. A.,
Melinski, R. D., Lima, A. P., Magnusson, W. E., & Neuschulz, E. L., Quiti
an, M., Santill
an, V., Töpfer, T., &
Baccaro, F. B. (2023). Beta diversity and microhabitat use of ant Schleuning, M. (2019). Functional and phylogenetic diversity of
assemblages in a white-sand vegetation gradient in central bird assemblages are filtered by different biotic factors on tropi-
Amazonia. Journal of Insect Conservation, 27(6), 855–864. cal mountains. Journal of Biogeography, 46(2), 291–303. https://
https://doi.org/10.1007/s10841-023-00517-4 doi.org/10.1111/jbi.13489
Dehling, D. M., Peralta, G., Bender, I. M. A., Blendinger, P. G., Hocking, D. J., & Babbitt, K. J. (2014). Amphibian contributions to
Böhning-Gaese, K., Muñoz, M. C., Neuschulz, E. L., ecosystem services. Herpetological Conservation and Biology,
Quiti
an, M., Saavedra, F., Santill
an, V., Schleuning, M., & 9(1), 1–17.
Stouffer, D. B. (2020). Similar composition of functional roles Hutter, C. R., Lambert, S. M., & Wiens, J. J. (2017). Rapid diversifi-
in Andean seed-dispersal networks, despite high species and cation and time explain amphibian richness at different scales
interaction turnover. Ecology, 101(7), e03028. https://doi.org/ in the tropical Andes, Earth's most biodiverse hotspot. The
10.1002/ecy.3028 American Naturalist, 190(6), 828–843. https://doi.org/10.5061/
Dehling, D. M., Töpfer, T., Schaefer, H. M., Jordano, P., Böhning- dryad.1555n
Gaese, K., & Schleuning, M. (2014). Functional relationships Jost, L. (2006). Entropy and diversity. Oikos, 113(2), 363–375.
beyond species richness patterns: Trait matching in plant-bird Jost, L. (2019). What do we mean by diversity?: The path towards
mutualisms across scales. Global Ecology and Biogeography, quantification. Metode, 2019(9), 55–61. https://doi.org/10.7203/
23(10), 1085–1093. https://doi.org/10.1111/geb.12193 metode.9.11472
Dehling, J. M., & Dehling, D. M. (2021). Conserving ecological Kaltsas, D., Dede, K., Giannaka, J., Nasopoulou, T.,
functions of frog communities in Borneo requires diverse forest Kechagioglou, S., Grigoriadou, E., Raptis, D., Damos, P.,
landscapes. Global Ecology and Conservation, 26, e01481. Vasiliadis, I., Christopoulos, V., Loukaki, E., Franses, R.,
https://doi.org/10.1016/j.gecco.2021.e01481 Vlachaki, D., & Avtzis, D. N. (2018). Taxonomic and functional
Devictor, V., Clavel, J., Julliard, R., Lavergne, S., Mouillot, D., diversity of butterflies along an altitudinal gradient in two
Thuiller, W., Venail, P., Villéger, S., & Mouquet, N. (2010). NATURA 2000 sites in Greece. Insect Conservation and Diver-
Defining and measuring ecological specialization. Journal of sity, 11(5), 464–478. https://doi.org/10.1111/icad.12292
12 SIQUEIRA ET AL.

Knauth, D. S., Moreira, L. F. B., & Maltchik, L. (2018). Partitioning and functional diversity of termites along a tropical elevational
tadpole beta diversity in highland ponds with different hydro- gradient. Biotropica, 49(2), 186–194. https://doi.org/10.1111/
periods. Freshwater Science, 37(2), 380–388. https://doi.org/10. btp.12365
1086/697926 Nunes-de-Almeida, C. H. L., Haddad, C. F. B., & Toledo, L. F.
Laliberté, E., & Legendre, P. (2010). A distance-based framework (2021). Amphibian reproductive modes: A revised classification
for measuring functional diversity from multiple traits. Ecology, of the amphibian reproductive modes. Salamandra, 57(3),
91(1), 299–305. 413–427.
Laliberté, E., Legendre, P., & Shipley, B. (2014). FD: Measuring Oksanen, J., Simpson, G.´., Blanchet, F., Kindt, R., Legendre, P.,
functional diversity from multiple traits, and other tools for Func- Minchin, P., O'Hara, R., Solymos, P., Stevens, M., Szoecs, E.,
tional Ecology. R package (1.0-12). Wagner, H., Barbour, M., Bedward, M., Bolker, B., Borcard, D.,
Leão-Pires, T. A., Luiz, A. M., & Sawaya, R. J. (2018). The complex Borman, T., Carvalho, G., Chirico, M., De Caceres, M., …
roles of space and environment in structuring functional, taxo- Weedon, J. (2022). vegan: Community Ecology Package. R pack-
nomic and phylogenetic beta diversity of frogs in the Atlantic age version 2.7–0. https://github.com/vegandevs/vegan,
Forest. PLoS One, 13(4), e0196066. https://doi.org/10.1371/ https://vegandevs.github.io/vegan/
journal.pone.0196066 Paradis, E., Claude, J., & Strimmer, K. (2004). APE: Analyses of
Legendre, P., & Anderson, M. J. (1999). Distance-based redundancy phylogenetics and evolution in R language. Bioinformatics, 20,
analysis: Testing multispecies responses in multifactorial eco- 289–290.
logical experiments. Ecological Monographs, 69(1), 1–24. Pavoine, S., Vallet, J., Dufour, A. B., Gachet, S., & Daniel, H. (2009).
Lourenço-de-Moraes, R., Campos, F. S., Ferreira, R. B., Solé, M., On the challenge of treating various types of variables: Applica-
Beard, K. H., & Bastos, R. P. (2019). Back to the future: Con- tion for improving the measurement of functional diversity.
serving functional and phylogenetic diversity in amphibian- Oikos, 118(3), 391–402. https://doi.org/10.1111/j.1600-0706.
climate refuges. Biodiversity and Conservation, 28(5), 1049– 2008.16668.x
1073. https://doi.org/10.1007/s10531-019-01706-x Paz, A., Hernandez, L., Melo, L. S. O., Lyra, M., Haddad, C. F. B., &
Luedtke, J. A., Chanson, J., Neam, K., Hobin, L., Maciel, A. O., Carnaval, A. C. (2022). Extreme environments filter function-
Catenazzi, A., Borzée, A., Hamidy, A., Aowphol, A., Jean, A., ally rich communities of Atlantic Forest treefrogs along altitu-
Sosa-Bartuano, Á., Fong G, A., de Silva, A., Fouquet, A., dinal and latitudinal gradients. Ecography, 2022(9), e06138.
Angulo, A., Kidov, A. A., Muñoz Saravia, A., Diesmos, A. C., https://doi.org/10.1111/ecog.06138
Tominaga, A., & Stuart, S. N. (2023). Ongoing declines for the Petchey, O. L., & Gaston, K. J. (2002). Functional diversity (FD),
world's amphibians in the face of emerging threats. Nature, species richness and community composition. Ecology Letters,
622(7982), 308–314. https://doi.org/10.1038/s41586-023-06578-4 5, 402–411.
Magnago, L. F. S., Edwards, D. P., Edwards, F. A., Magrach, A., Pitogo, K. M. E., Saavedra, A. J. L., Aurellado, M. E. B., de
Martins, S. V., & Laurance, W. F. (2014). Functional attributes Guia, A. P. O., & Afuang, L. E. (2021). Functional traits and
change but functional richness is unchanged after fragmenta- environment drive montane amphibian distribution in the
tion of Brazilian Atlantic forests. Journal of Ecology, 102(2), southern Philippines. Biodiversity and Conservation, 30(14),
475–485. https://doi.org/10.1111/1365-2745.12206 4177–4197. https://doi.org/10.1007/s10531-021-02299-0
Mammola, S., Carmona, C. P., Guillerme, T., & Cardoso, P. (2021). Poynton, J. C. (2003). Altitudinal species turnover in southern
Concepts and applications in functional diversity. Functional Tanzania shown by anurans: Some zoogeographical consider-
Ecology, 35(9), 1869–1885. https://doi.org/10.1111/1365-2435. ations. Systematics and Biodiversity, 1(1), 117–126. https://doi.
13882 org/10.1017/S1477200002001019
McCain, C. M. (2009). Global analysis of bird elevational diversity. QGIS Development Team. (2023). QGIS geographic Information sys-
Global Ecology and Biogeography, 18(3), 346–360. https://doi. tem. QGIS Association. https://www.qgis.org
org/10.1111/j.1466-8238.2008.00443.x Qian, H., Wang, X., Wang, S., & Li, Y. (2007). Environmental deter-
Naniwadekar, R., & Vasudevan, K. (2007). Patterns in diversity of minants of amphibian and reptile species richness in China.
anurans along an elevational gradient in the Western Ghats, Ecography, 30(4), 471–482. https://doi.org/10.1111/j.2007.0906-
South India. Journal of Biogeography, 34(5), 842–853. https:// 7590.05025.x
doi.org/10.1111/j.1365-2699.2006.01648.x R Core Team. (2021). R: A language and environment for statistical
Nettesheim, F. C., Garbin, M. L., Pereira, M. G., de computing (4.1.2). R Foundation for Statistical Computing.
Araujo, D. S. D., & de Viveiros Grelle, C. E. (2018). Local-scale Rahbek, C. (1995). The elevational gradient of species richness: A
elevation patterns of Atlantic Forest tree community variation uniform pattern? Ecography, 18(2), 200–205. https://doi.org/10.
and assembly drivers in a conservation hotspot in southeastern 1111/j.1600-0587.1995.tb00341.x
Brazil. Flora: Morphology, Distribution, Functional Ecology of Rahbek, C. (2005). The role of spatial scale and the perception of
Plants, 248, 61–69. https://doi.org/10.1016/j.flora.2018.08.016 large-scale species-richness patterns. Ecology Letters, 8(2), 224–
Nunes, C. A., Braga, R. F., Figueira, J. E. C., De Siqueira 239. https://doi.org/10.1111/j.1461-0248.2004.00701.x
Neves, F., & Fernandes, G. W. (2016). Dung beetles along a Ramalho, Q., Tourinho, L., Almeida-Gomes, M., Vale, M. M., &
tropical altitudinal gradient: Environmental filtering on taxo- Prevedello, J. A. (2021). Reforestation can compensate negative
nomic and functional diversity. PLoS One, 11(6), e0157442. effects of climate change on amphibians. Biological Conserva-
https://doi.org/10.1371/journal.pone.0157442 tion, 260, 109187. https://doi.org/10.1016/j.biocon.2021.109187
Nunes, C. A., Quintino, A. V., Constantino, R., Negreiros, D., Reis Ramalho, Q., Vale, M. M., Manes, S., Diniz, P., Malecha, A., &
Júnior, R., & Fernandes, G. W. (2017). Patterns of taxonomic Prevedello, J. A. (2023). Evidence of stronger range shift
SIQUEIRA ET AL. 13

response to ongoing climate change by ectotherms and high- Terborgh, J. (1977). Bird species diversity on an Andean elevational
latitude species. Biological Conservation, 279, 109911. https:// gradient. Ecology, 58(5), 1007–1019.
doi.org/10.1016/j.biocon.2023.109911 Thakur, D., & Chawla, A. (2019). Functional diversity along eleva-
Reymond, A., Purcell, J., Cherix, D., Guisan, A., & Pellissier, L. tional gradients in the high altitude vegetation of the western
(2013). Functional diversity decreases with temperature in high Himalaya. Biodiversity and Conservation, 28(8–9), 1977–1996.
elevation ant fauna. Ecological Entomology, 38(4), 364–373. https://doi.org/10.1007/s10531-019-01728-5
https://doi.org/10.1111/een.12027 Tilman, D. (2001). Functional diversity. In S. A. Levin (Ed.), Ency-
Rezende, V. L., de Miranda, P. L. S., Meyer, L., Moreira, C. V., clopedia of biodiversity (Vol. 3, pp. 109–120). Academic Press.
Linhares, M. F. M., de Oliveira-Filho, A. T., & Eisenlohr, P. V. Villacampa, J., Whitworth, A., Allen, L., & Malo, J. E. (2019). Alti-
(2015). Tree species composition and richness along altitudinal tudinal differences in alpha, beta and functional diversity of an
gradients as a tool for conservation decisions: The case of amphibian community in a biodiversity hotspot. Neotropical
Atlantic semideciduous forest. Biodiversity and Conservation, Biodiversity, 5(1), 60–68. https://doi.org/10.1080/23766808.2019.
24(9), 2149–2163. https://doi.org/10.1007/s10531-015-0939-z 1659022
Roswell, M., Dushoff, J., & Winfree, R. (2021). A conceptual guide Villéger, S., Mason, N. W. H., & Mouillot, D. (2008). New multidi-
to measuring species diversity. Oikos, 130(3), 321–338. https:// mensional functional diversity indices for a multifaceted frame-
doi.org/10.1111/oik.07202 work in functional ecology. Ecology, 89(8), 2290–2301. https://
RStudio Team. (2023). RStudio: Integrated development for R. RStu- doi.org/10.1890/07-1206.1
dio, PBC. http://www.rstudio.com/ Winfree, R., Fox, J. W., Williams, N. M., Reilly, J. R., &
Seale, D. B. (1980). Influence of amphibian larvae on primary pro- Cariveau, D. P. (2015). Abundance of common species, not spe-
duction, nutrient flux, and competition in a pond ecosystem. cies richness, drives delivery of a real-world ecosystem service.
Ecology, 61(6), 1531–1550. Ecology Letters, 18(7), 626–635. https://doi.org/10.1111/ele.
Siqueira, C. C., Vrcibradic, D., Almeida-Gomes, M., Menezes, V. A., 12424
Borges-Junior, V. N. T., Hatano, F. H., Pontes, J. A. L., Zancolli, G., Steffan-Dewenter, I., & Rödel, M. O. (2014). Amphib-
Goyannes-Araújo, P., Guedes, D. M., Van Sluys, M., & ian diversity on the roof of Africa: Unveiling the effects of habi-
Rocha, C. F. D. (2011). Species composition and density esti- tat degradation, altitude and biogeography. Diversity and
mates of the anurofauna of a site within the northernmost large Distributions, 20(3), 297–308. https://doi.org/10.1111/ddi.12161
Atlantic Forest remnant (Parque Estadual do Desengano) in the
state of Rio de Janeiro, Brazil. Biota Neotropica, 11(4), 131–137.
Siqueira, C. C., Vrcibradic, D., Almeida-Gomes, M., & SU PP O R TI N G I N F O RMA TI O N
Rocha, C. F. D. (2021). Assessing the importance of reproduc- Additional supporting information can be found online
tive modes for the evaluation of altitudinal distribution patterns in the Supporting Information section at the end of this
in tropical frogs. Biotropica, 53(3), 786–797. https://doi.org/10. article.
1111/btp.12933
Siqueira, C. C., Vrcibradic, D., Dorigo, T. A., & Rocha, C. F. D.
(2011). Anurans from two high-elevation areas of Atlantic for- How to cite this article: Siqueira, C. C.,
est in the state of Rio De Janeiro, Brazil. Zoologia, 28(4), 457– Vrcibradic, D., & Rocha, C. F. D. (2025). Effects of
464. https://doi.org/10.1590/S1984-46702011000400007 elevation on taxonomic and functional diversity of
Sun, Z. J., Zhu, W., Zhu, W. B., Zhao, C. L., Liao, C. L., Zou, B.,
frog communities in a tropical forest. Ecological
Xu, D., Fan, W. B., Su, S. Q., Jiang, J. P., & Zhao, T. (2021). Spa-
tiotemporal patterns of anuran functional diversity in temper-
Research, 1–13.
ate montane forests. Zoological Research, 42(4), 412–416. https://doi.org/10.1111/1440-1703.12551
https://doi.org/10.24272/J.ISSN.2095-8137.2020.341