Butterfly (Insecta: Lepidoptera) assemblage structure along an upland-highland elevational gradient in a Brazilian tepui

Estrutura da assembleia de borboletas (Insecta: Lepidoptera) ao longo de um gradiente altitudinal de terras altas em um tepui brasileiro

ABSTRACT

Pantepui is a biogeographic province in the Guiana Highlands, considered an archipelago of tabletop mountains (tepuis) with a unique and poorly-known biota. Here, we investigate the butterfly assemblage structure along an elevational gradient on the Brazilian side of the Uei Tepui. We sampled butterflies in 14 plots, each at least 500 m apart, ranging from 1000 to 1820 m elevation. Each plot contained four baited traps in the canopy and four in the understory, set in pairs. We also performed standardized active sampling using an entomological net in each plot and additionally sampled butterflies opportunistically between 920 and 2100 m to complement the butterfly inventory. Over 21 days, we sampled 299 individuals from 93 species of butterflies. In the sampling plots, richness and abundance decrease considerably with altitude, and most species have been recorded (74%) in the Amazonia lowlands. However, there was a marked change in species richness, abundance, and composition at 1330 m compared to lower altitudinal plots. From this altitude, the species replacement (turnover) decreased, while the occurrence of endemic species strongly increased, especially in elevations above 1500 m. Our results highlight elevational effects such as spatial limitation and filtering on butterfly diversity, reinforcing the importance of conservation in this megadiverse region.

Keywords:
Amazonia; altitudinal gradient; beta-diversity; community; endemism

RESUMO

Pantepui é uma província biogeográfica nas terras altas do Escudo das Guianas, considerada um arquipélago de montanhas em forma de mesa (tepuis) com uma biota única e pouco conhecida. Aqui, investigamos a estrutura da assembleia de borboletas ao longo de um gradiente altitudinal no lado brasileiro do Uei Tepui. Amostramos borboletas em 14 parcelas, cada uma com pelo menos 500 m de distância, variando de 1000 a 1820 m de altitude. Cada parcela continha quatro armadilhas atrativas no dossel e quatro no sub-bosque, configuradas em pares. O esforço de coleta foi complementado por busca ativa padronizada usando rede entomológica e também de forma oportunista entre 920 e 2100 m para complementar o inventário de borboletas. Ao longo de 21 dias, amostramos 299 indivíduos de 93 espécies de borboletas. Nas parcelas, a riqueza e a abundância diminuíram consideravelmente com a altitude, e a maioria das espécies foi registrada (74%) nas terras baixas da Amazônia. No entanto, houve uma mudança marcante na riqueza de espécies, abundância e composição aos 1330 m em comparação com as parcelas altitudinais mais baixas. A partir desta altitude, a substituição de espécies (turnover) diminuiu, enquanto a ocorrência de espécies endêmicas aumentou fortemente, especialmente em elevações acima de 1500 m. Nossos resultados destacam os efeitos da elevação, como limitação espacial e o filtro ambiental, sobre a diversidade de borboletas, reforçando a importância da conservação nesta região megadiversa.

Palavras-chave:
Amazônia; beta-diversidade; comunidade; endemismo; gradiente altitudinal

INTRODUCTION

Biodiversity patterns along altitudinal gradients have been crucial to understanding species distribution in space and time (Rahbek et al. 2019). From an evolutionary perspective, there is a consensus that mountains greatly influence species diversification and endemism (Antonelli et al. 2018; Quintero and Jetz 2018; Perrigo et al. 2020). Mountains provide marked climatic variations in short spatial distances, and these differences result in large changes in the local geophysical and environmental conditions, which influence the occurrence and persistence of species (Körner 2007).

The Pantepui is a biogeographic province of the Guiana Highlands composed of a set of high mountains (1500-3000 m in elevation) surrounded by Amazonian lowland rainforests and savannas, considered an archipelago of sky islands (Mayr and Phelps 1967; Berry and Riina 2005). It is one of the oldest geological formations in South America (Rull et al. 2019a), characterized by typical flat-topped mountains, locally known as tepuis. The tepuis are remnants of a long-standing erosional process accumulated since the Precambrian times (Briceño and Schubert 1990). The upper portions (hereafter highlands - above 1500 m) display highly differentiated ecosystems regarding geological, geomorphological, geochemical, and biological aspects. These sharp differences along altitudinal zonation in tepuis are usually not found in other mountains where continuous formation has no significant interruption and present a more gradual elevational change (Viloria and Costa 2019). Due to these unique features related to geomorphological formations, biotic composition, and endemism, Pantepui is considered one of the areas of endemism in the Neotropics (Morrone 2017).

The elevational stages of Pantepui mountains are classified based on the phytogeographic criteria, with lowlands (0-500 m a.s.l. - above sea level), uplands (500-1500 m a.s.l.), and highlands (above 1500 m a.s.l.) representing the general distribution and diversity patterns of Pantepuian organisms (for more details, see Huber 1995 and Rull et al. 2019a). Although such elevational stages based on the phytogeographic criteria have been formalized, the actual categorization of the highlands can vary between 1200-1500 m depending on the tepui environment and biotic criteria (Huber 1987; Rull et al. 2019a). The highlands are the elevational stage where endemic species are usually found and despite the high vegetational endemism (~ 41%; Riina et al. 2019), species richness, abundance, and diversity of plants and animals tend to be lower compared to the uplands and lowlands (Berry and Riina 2005; Lew and Lim 2019; Ochoa and Rojas-Runjaic 2019; Nogueira et al. 2021).

Organisms with a strong association with plants usually respond similarly to changes in vegetation composition and structure (Dennis et al. 2004; Oliveira et al. 2021). For instance, most butterfly species are host-plant specialists (DeVries 1986), and their community similarity tends to decrease with more heterogeneous and diverse plant communities (Steffan-Dewenter and Tscharntke 2000; Kumar et al. 2009). Studies of butterflies along an elevational gradient in two mountain chains in Brazil (e.g., Serra do Mar and Espinhaço) showed a strong correlation between butterfly and plant composition and diversity (Carneiro et al. 2014; Beirão et al. 2020; Pires et al. 2020). Therefore, the phytogeographic definition can be used as a surrogate to investigate the diversity patterns of other organisms (Viloria and Costa 2019).

Understanding the variation of butterfly species composition along elevational gradients (spatial β-diversity) may provide valuable insights into how the community of this important group of bioindicators (Bonebrake et al. 2010) is structured in unique regions like the Pantepui. For instance, β-diversity partitioning into turnover and nestedness components can shed light on the underlying processes shaping community structure (Baselga 2010). Spatial β-diversity can be decomposed into turnover, which reflects species replacement between sites and is usually associated with environmental and biotic heterogeneity (Socolar et al. 2016), and nestedness, which represents an ordered species loss across sites, where sites with fewer species are subsets of richer ones (Baselga 2010; Legendre 2014).

The present study aimed to elucidate the structure of the butterfly community in Uei Tepui and evaluate the diversity and endemism patterns along the elevational gradient. We hypothesized that (1) butterflies’ richness, abundance, and diversity should decrease with elevation, and the main compositional change, including the appearance of endemic species, should occur between 1200 and 1500 m a.s.l., as reported for different taxa in Pantepui. (2) We also expect that turnover will be more dominant than nestedness, and the upland butterfly community will present higher species turnover than the highland community because of more space and resource availability on lower elevations.

MATERIAL AND METHODS

Study area

This study was carried out in the eastern portion of the Pantepui formation (Figure 1a, b). Sampling was undertaken at the Uei Tepui (5°5’N; 60°35’W) in the northern part of the Brazilian Amazon in the Roraima State (Figure 1c). Uei Tepui is located on the Brazilian and Venezuelan borders. It has a maximum elevation of 2150 m, with a summit area of 2.5 km² and a sloped area of 20 km² (Rull et al. 2019a). On the Brazilian side, the Uei Tepui is covered by two overlapping protected areas, the Raposa Serra do Sol Indigenous Land (Fundação Nacional dos Povos Indígenas - FUNAI) and the Mount Roraima National Park (Instituto Chico Mendes de Conservação da Biodiversidade - ICMBio; Figure 1d). The vegetation zonation along elevation can be seen in Figure 1, where Figures 1e-g depict the upland and Figure 1h-j the highland.

Figure 1
(A) Pantepui map in the northernmost part of South America with (B) the Eastern Pantepui region and (C) the study region showing the location of each sampling unit (plot) in Uei Tepui, Roraima State, Brazil. (D) Uei Tepui view and its flat-top summit with continuous upland-highland forests (photo: Ramiro Melinski); (E) Open area at 920 m a.s.l.; (F) Forest structure at 1000-1100 m a.s.l.; (G) Forest at 1300 m a.s.l.; (H) Epiphytes and mosses on trunks at 1500 m a.s.l.; (I) Butterfly trap at 1820 m a.s.l.; (J) Uei Tepui summit at 2100 m a.s.l.

The phytogeographical characterization of the elevational stages of the Uei Tepui aligns with the general characterization of tepuis (personal observation). The surrounding lowlands of the Uei Tepui are usually covered by open savannas with small herbaceous formations or by tropical wet forests (Rull et al. 2019a). The uplands include open savannas, lower montane and montane vegetation, with mixed forest formations also present (Rull et al. 2019a). In the highlands, evergreen cloud forests are found on the slopes. As the elevation increases, there is a decrease in the canopy height towards the summit, where shrubs, carnivorous plants, grasses, and open-rock communities often dominate (Rull et al. 2019a).

Sampling design and data collection

Butterflies were collected at the end of the wet season, from November 29^th to December 19^th of 2019, in 14 200-m long sampling plots (eight in the uplands, and six in the highlands) along a steep forest slope ranging from 1000 to 1820 m a.s.l. (~15 km). Most plots were at least 500 m apart from each other, except for the higher altitudinal plots (1820 m a.s.l.), which were separated by 200 m due to spatial limitations and logistical constraints.

In each plot, we placed four pairs of Van Someren-Rydon type traps (Rydon 1964) 50 m apart, each pair consisting of one trap in the understory (~ 1- 1.5 m high) and one in the canopy (~ 15- 20 m high). The higher altitudinal plots had only understory traps due to the low canopy height (~5 m; Figure 1i). Traps contained bait made of bananas and sugarcane juice fermented for 48 hours and were kept open for seven consecutive days, being checked every 48 hours (Freitas et al. 2014), resulting in a total sampling effort of 728 trap-days (56 trap-days per plot, except for the two plots at 1820 m a.s.l. which had half the trapping effort: 28 trap-days). We followed the butterfly sampling protocol of the Brazilian Biodiversity Monitoring Program (Programa Nacional de Monitoramento da Biodiversidade do Instituto Chico Mendes - Programa Monitora; Brasil 2022), using a bait proportion of 3 kg of bananas to 1 liter of sugarcane (Nobre et al. 2014). We choose to check the traps every 48 hours to be able to sample in a larger number of plots. Thus, every day, we checked different plots distant from each other. Standardized active searches using entomological nets were performed along all plots, enabling comparable analysis across sites. This sampling was conducted simultaneously with trap checks, with two collectors walking at 1 km h^-1 and collecting all butterflies sighted up to 2.5 m on each side of the transect. This method allowed us to sample both fruit-feeding and nectar-feeding butterflies.

Since this was the first biological expedition in this region, we took advantage of the opportunity to document as many butterfly species as possible. Therefore, in addition to standardized sampling, we conducted opportunistic surveys (e.g., non-standardized method) outside sampling units, including campsite areas and the Ingarikó people’s village, across an elevational range of 920 to 2100 m a.s.l. using entomological nets. We also surveyed the summit of Uei Tepui, but we did not detect any butterflies, likely due to the continuous rainy conditions. Butterflies sampled during these opportunistic surveys were not included in the statistical analysis, as mentioned below.

Butterflies were identified at the species and subspecies levels using an online guide (www.butterfliesofamerica.com), including a book chapter and taxonomic literature about Pantepui butterflies (Costa et al. 2017; Viloria and Costa 2019; Costa et al. 2020). Expert taxonomists in Neotropical and Pantepui butterfly fauna, Mauro Costa, Ángel Viloria, Andrew Nield, and Patrick Blandin, verified the butterfly identifications. Specimens were collected under permissions SISBIO 72093-1 and 57444. All collected butterflies were deposited in the Entomological Collection of the National Institute of Amazonian Research (INPA), and some individuals of the most abundant species were also deposited in the Community Ecology Lab of the Federal University of Amazonas (UFAM).

We compiled butterfly studies in the lowlands of the Brazilian Amazon with an available species list to better characterize the uniqueness of the butterfly fauna of Uei Tepui. The second author, Mauro Costa, classified the endemic species, and we also consulted the literature on Pantepui butterflies (Costa et al. 2017; Viloria and Costa 2019; Costa et al. 2020).

Statistical analysis

In this study, all analyses were performed using data from sampling units (plots from 1000 to 1820 m a.s.l.) where standardized methods were employed. For each plot, we combined butterflies sampled in the traps of both strata and those captured with the entomological net from the transect walk. To account for minor altitudinal variation among traps, we placed them at similar elevations whenever possible. To control any larger variation, we used the mean altitude of the four pairs of traps installed along each sampling plot to represent its elevation (Supplementary Material, Table S1).

To determine if our rapid inventory provides data to support our analysis, we used coverage-based rarefaction and extrapolation curves for each elevational stage (upland and highland) to compare and estimate species richness among elevational stages with different numbers of individuals collected (Colwell et al. 2012; Chao et al. 2014). The interpolated and extrapolated values for abundance data were based on Hill numbers (q = 0) and were generated in the iNEXT package (Hsieh et al. 2016).

To evaluate the butterfly richness, abundance, and composition along the Uei Tepui elevational gradient, we ran Generalized Linear Models (GLMs). Because the sampling effort was lower in the highest plots, we included an ‘offset’ term (log-transformed trap-days) in the models to standardize comparisons across plots. The offset acts as a known adjustment factor, allowing the model to account for differences in sampling effort by scaling the expected response without estimating an additional parameter (Crawley 2013). We used the number of species and the number of individuals as the dependent variables, with elevation as the predictor. We selected a quasi-Poisson distribution for both models due to overdispersion (Zuur et al. 2009). For community composition, we first reduced the dimensionality using Non-metric Multidimensional Scaling (NMDS) based on Bray-Curtis dissimilarities. The species-abundance data were standardized by the total abundance of butterflies per plot to make the ordination more sensitive to the composition rather than the total abundance. We performed a GLM using the one-dimensional NMDS solution as the dependent variable and altitude as an explanatory variable. For this analysis, we used the Gaussian error distribution (Crawley 2013). To better illustrate the species composition variation along the elevational gradient, we plot the relative species abundances along the elevation gradient (Dambros 2020; for the total species sampled, see Supplementary Material, Figure S2). We used “metaMDS” and “vegdist” functions from the package “vegan” (Oksanen et al. 2019) to calculate the NMDS axis.

We compared the Sørensen dissimilarities of butterflies’ composition between the elevational stages of upland and highlands. We defined the highlands where Pantepui endemic species started to occur regularly, with all plots above presenting endemic species, since the beginning of the highlands depend on the tepui environment and the focal organism (Huber 1987; Rull et al. 2019a). We used abundance data from standardized sampling plots to estimate which β-diversity component (turnover or nestedness) dominated in each elevational stage. To assess the spatial β-diversity, we obtained a multisite dissimilarity value for each component in the upland and highland butterfly communities using the “beta_div” function (see Legendre 2014). To assess whether multisite β-diversity differs among the elevational stages, we calculated bootstrap 95% confidence intervals. We identified significant differences in the lack of overlap between 95% confidence intervals (Manly 2007). All bootstrap analyses were based on 999 randomizations using half of the samples in each permutation.

We conducted all analyses using version 4.1.0 of the R program (R Core Team 2021).

RESULTS

We sampled 299 individuals from 93 species of butterflies in the Uei Tepui, 241 individuals of 57 species of fruit-feeding and 58 individuals of 36 species of nectar-feeding butterflies (Table S2). Among those, 89 individuals belonging to 43 species were collected with opportunistic surveys, with 22 species sampled exclusively with this method. These individuals collected with opportunistic surveys were not included in the analysis below.

In the standardized sampling plots, we collected 210 butterflies belonging to 71 species from five families: Nymphalidae (60 spp), Riodinidae (6 spp), Pieridae (2 spp), Lycaenidae (2 spp), and Hesperiidae (1 sp). Only four species had ten or more individuals collected: Taygetis laches (Fabricius, 1793) (n = 35), Nhambikuara furina (Hewitson, 1862) (n = 14), Antirrhea ulei (Strand, 1912) (n = 12), and Amiga arnaca (Nakahara, Willmott & Espeland, 2019) (n = 10). Singletons represented 49% and doubletons accounted for 21% of the sampled specimens. However, in the highland, there was a large dominance of A. ulei, which corresponded to 52% of the sampled individuals up to 1500 m a.s.l. The sampling completeness suggests that our sampling effort recorded ~84% of species in the uplands and ~73% of the estimated species richness for the highland (Figure S1).

Butterfly richness (t = -4.146, p = 0.001) and abundance (t = -4.893, p < 0.001) strongly decreased with altitude. There was a marked change in richness and abundance above 1330 m a.s.l. where the average number of species and individuals was nearly half of the lower altitude plots (Figure 2a,b). The one-axis NMDS solution converged and had moderate stress (0.13). Species composition also varied strongly with elevation (GLM; R² = 0.83, t = -7.38, p < 0.001; Figure 2c).

Figure 2
Linear regression of the effect of elevation on (A) number of species, (B) number of individuals, and (C) butterfly composition along the altitudinal gradient of Uei Tepui, Roraima State, Brazil. Points represent each sampling plot, and the shaded areas represent 95% of the confidence intervals.

Among standardized samples, 74% of the species (n = 53) were recorded at least once in previous studies in the lowlands of the Brazilian Amazon (reference studies in the Table S3). Most of these species (n = 50) were sampled from 1000-1100 m a.s.l. Endemic species were detected in all sampling plots above 1330 m a.s.l. (Figure 3). The first endemic species of Pantepui caught in the sampling plots was the new subspecies Caligo suzanna weidmanni (Blandin, Attal, Orellana, Nield, Lamas & Costa, 2025) (Costa et al. 2025) at the 1330 m elevation. Of the 13 species sampled in the plots above this elevation, six (46%) were endemic to Pantepui, with A. ulei being the most frequent species from 1390 m to 1820 m a.s.l. (Figure 3). The endemism rate increases to 55.5% (10 of 18 species; Figure 4) when considering individuals sampled in opportunistic sampling events above 1330 m a.s.l. (not included in the main analysis).

Figure 3
Direct ordination of butterfly species and their relative abundance (bar thickness) with the elevational gradient in Uei-tepui, Roraima State, Brazil. * butterflies endemic to Pantepui; † new species record to Brazil. The dashed line represents the lowest altitude where the first endemic species was collected in the sampling plots.

Figure 4
Endemic butterflies from Pantepui sampled in Uei Tepui, Roraima state, Brazil. (A) Antirrhea ulei; (B) Pereute lindemannae pemona*; (C) Archonias sisamnus ayanganna*; (D) Memphis montesino; (E) Hyposcada zarepha bonplandii; (F) Oleria boyeri; (G) new species of Oleria sp.*; (H) Greta clavijoi; (I) Dismorphia zathoe proserpina*; (J) Melete leucardia reyi**; (K) Junonia evarete oscura**; (L) Caligo suzanna weidmanni. * = sampled opportunistically; ** sampled opportunistically in an open field in the upland.

Species turnover accounted for 89% of the observed compositional variation among our sampling plots across the elevational gradient. Although the total β-diversity (species dissimilarity) was higher in the highland plots than in the upland plots, this difference was not statistically significant (Figure 5a). Interestingly, despite the relatively small difference in altitude between the upland sampling plots, the spatial turnover was significantly greater in the upland butterfly communities (93%) compared to the highland communities (69%) (Figure 5b).

Figure 5
The relative contribution of (A) total β-diversity and (B) spatial turnover among the elevational stages. The bars represent the observed multisite component values, and the black lines represent 95% confidence intervals based on 999 bootstraps. Different letters indicate significant differences between upland and highland butterfly communities based on the lack of confidence interval overlap.

DISCUSSION

For the first time, we were able to study and describe the butterfly assemblage structure across a continuous elevation gradient in this isolated tepui. We found that despite the unique environmental aspects of the Brazilian side of the Uei Tepui, the butterfly community followed the general patterns of community structure observed along elevational gradients in the Pantepui mountains. We found a marked change in species richness, abundance, and composition around 1330 m. Above this point, we also found lower species replacement and higher occurrence of endemic species.

Butterfly richness and abundance have been shown to decrease with altitude in most of the studies along altitudinal gradients of Brazilian mountain chains (Espinhaço [Beirão et al. 2020; Pires et al. 2020], Serra da Mantiqueira [Henriques et al. 2022], and Serra do Mar [Carneiro et al. 2014]). This is a typical pattern found for butterflies and other organisms around the world (Acharya and Vijayan 2015; Perillo et al. 2017; Castro et al. 2019; Berrones-Morales et al. 2020; Afzal et al. 2021; Dzekashu et al. 2022). Specifically in the Guiana Highlands, several taxa present a similar pattern, including plants (Berry and Riina 2005), other invertebrates (Ochoa and Rojas-Runjaic 2019; Nogueira et al. 2021) and mammals (Lew and Lim 2019).

There are some general explanations for the decline in species richness and abundance with increasing elevation. Area reduction toward the summit is a notable limiting factor for species diversity (Rahbek 1997; Safont et al. 2014; Ochoa and Rojas-Runjaic 2019; Riina et al. 2019). This area effect is likely to be particularly relevant in the Uei Tepui, which has one of the smallest summits and slope areas among tepuis (Rull et al. 2019a). Spatial limitation tends to select more specialized biota, decreasing diversity and increasing endemism (Viloria and Costa 2019). This pattern was described for plant communities in different tepuis in the Guiana Highlands (Berry and Riina 2005), and this compositional change may also influence the butterfly community.

The presence of host plants and life strategies strongly influences the distribution of butterflies, with larval specificity and habitat specialization being critical determinants of species occurrence (Dennis et al. 2004; Koh et al. 2004; Hodkinson 2005). Consequently, the strong compositional dissimilarity of butterflies along the elevational gradient reported here may also result from the phytogeographical changes along the tepuis (Viloria and Costa 2019).

Besides the spatial limitation, higher elevations act as a strong environmental filter on butterflies (Nogué et al. 2012; Chazot et al. 2014; Henriques et al. 2022). Climatic conditions, especially temperature, greatly influence ectothermic organisms, acting not only in the time of their activity during the day but also in growth, development, and seasonality (Hodkinson 2005). During our study, the rainfall was prevalent, especially in the highlands, where the temperature was much cooler than in the uplands. This likely explains the low number of butterfly individuals sampled during the expedition compared to other mountainous regions in the Neotropics (Marín et al. 2015; Henriques et al. 2019; Álvarez et al. 2021). However, because there are no studies in the Pantepui that include the number of species and abundance data, and the majority of butterfly studies in tepuis focus on species descriptions or the occurrence of new species in certain tepui (Nakahara et al. 2014; Costa et al. 2014a; Fratello et al. 2015; Costa et al. 2019), we cannot state that our observed abundance was lower than expected by our sampling effort. The absence of species commonly found in Neotropical montane environments, such as the Pronophilina subtribe, might be due to these climatic and seasonal variables. While these butterflies are present mainly in the Andes (Pyrcz et al. 2009; Pyrcz and Rodríguez 2007; Pyrcz et al. 2016), they are also found in some tepuis, including the adjacent Roraima massif (Pyrcz and Fratello 2005).

Although we are not able to verify whether the 200 m distance between plots interfered with sampling independence in the highest plots (i.e., 1820 m a.s.l.), the lack of butterflies (we sampled only one individual in the traps and none were observed during active sampling) suggests that this effect was minimal. This minimum distance between sampling plots has been considered adequate to maintain independence between sampling units in the Atlantic Forest (see Ribeiro et al. 2012; Freitas et al. 2014), although such information is lacking for mountainous and Amazonian systems. However, most butterfly individuals move short distances and tend to remain within the same habitat type (Lourenço et al. 2021). In addition, oxygen availability decreases at higher altitudes, which can affect the ability to sustain flight performance in insects (Hodkinson 2005). Because habitat types and abiotic conditions can shift abruptly over short spatial scales along altitudinal gradients (Körner 2007), butterflies at higher elevations are expected to have narrower home ranges. Therefore, this spatial proximity between sampling units is unlikely to have affected our results.

Due to the lower abundance of butterflies observed this study, it is suggested that future studies use both rotten fish and fermented fruit as bait to increase the sampling effectiveness in high altitudinal sites (Álvarez et al. 2021). We did not include those items because our sampling was part of the Brazilian Biodiversity Monitoring Program (Brasil 2022), therefore it was important to maintain the standardized protocol for bait and the number of traps used across other Protected Areas (Nobre et al. 2014).

In this study, butterfly composition changed gradually across the sampling plots along the elevational gradient of the Uei Tepui, but around 1330 m a.s.l., endemic species from Pantepui started to appear regularly. From this altitude, the butterfly richness and abundance decreased dramatically compared to the sampling plots in the 1000-1100 m a.s.l. Despite the general definition of highlands as being above 1500 m a.s.l. (Rull et al. 2019a), our butterfly occurrence data show that the highlands in the Uei Tepui begin around 1330 m. This result is in line with Pantepui characterization, which suggest a significant faunal change and the appearance of endemic species at around 1200-1500 m (Huber 1987). The butterfly endemism rate in the highlands (46%) aligns with the rate of plant species exclusively distributed in the highlands of Pantepui (Berry and Riina 2005; Riina et al. 2019), providing compelling evidence for the cohesive interaction between butterflies and plant composition.

The β-diversity values were similar between the upland and the highland communities. The species replacement (turnover) was the main driver of the differences in composition among our sampling plots along the elevation gradient. This pattern is commonly found in studies of elevational gradients (Beirão et al. 2020; Pires et al. 2020; Cómbita et al. 2022; Montañez-Reyna et al. 2023). As we expected, the species turnover presents a higher contribution in the upland than in the highland communities. Some studies show that species diversity and turnover rates tend to decrease at higher elevations due to stronger environmental filtering and reduced environmental heterogeneity (Beirão et al. 2020; Castro et al. 2020). Our findings underscore the significant decline in species richness at the highest elevation sampled (1820 m a.s.l.), which is crucial in reducing turnover and increasing nestedness.

Although Pantepui is one of the most singular provinces for Neotropical biodiversity and endemism (Rull and Vegas-Villarúbia 2017), its remoteness and difficult access have limited our understanding of its diversity (Costa et al. 2014b; Carvalho et al. 2023). Sporadic expeditions, which began in the mid-19th century and continue to this day, have provided valuable insights into the origin and diversity of organisms in this unique environment (Rull 2019; Rabelo et al. 2021). However, long-term monitoring of both fauna and flora is essential to understand the environmental influences on communities’ structure and contribute to conservation strategies (Freitas et al. 2024). This is especially important in the context of climate change which is considered one of the greatest threats to mountain biota worldwide (Forister et al. 2010; Rull and Vegas-Villarúbia 2017; Rull et al. 2019b; Rödder et al. 2021; Chowdhury 2023). Therefore, initiatives like the Brazilian Biodiversity Monitoring Program (Programa Monitora,Brasil 2022) and Protected Areas of Amazonia (ARPA) programs (arpa.mma.gov.br) are crucial for advancing this understanding.

CONCLUSION

Our study was the first to evaluate the butterfly assemblage structure along a tepui altitudinal gradient using standardized and combined sampling methods. Our results indicate that the Uei Tepui harbors a structured butterfly community shaped by elevation, with reduced richness and abundance at higher altitudes and a distinct shift in composition around 1330 m a.s.l., including a higher occurrence of endemic species. These findings support broader biogeographical patterns observed in other tepuis and taxa, reinforcing the role of elevation, spatial limitation, and environmental filtering in shaping Neotropical montane diversity. The distinctiveness of the Pantepui region, combined with its high rates of endemism and vulnerability to climate change, highlights the urgency of expanding research and conservation efforts in these isolated mountain systems.

ACKNOWLEDGMENTS

The Instituto Chico Mendes de Conservação da Biodiversidade (ICMBio) funded this research. The authors are grateful for logistical support from ICMBio and Conselho do Povo Indígena Ingarikó (COPING). This study was carried out with the consent and collaboration of the indigenous Ingarikó people and with the cooperation of Brazil’s National Indigenous Foundation, FUNAI. Special thanks to Rodrigo and Oswaldo Ingarikó for their essential field assistance and to the Karumanpaktëi village for their logistical support and cordial reception. IFO is grateful to the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES for her doctoral (#88882.347460/2019-01) and doctoral sandwich fellowships (PDSE; #88881.623028/2021-01). RMR thanks Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq, #142352/2017-9) for a doctoral fellowship. FPW and FBB thank CNPq for their productivity fellowships (311504/2020-5 and 313986/2020-7, respectively). Butterfly specimens were collected under permissions SISBIO 72093-1 and 57444. We thank Thiago Laranjeiras, Marina Beirão, Helena Romanowski, Eduardo Carneiro, Onildo Marini-Filho, Pedro Pequeno and anonymous reviewers for constructive comments on an earlier version of this manuscript. We also thank Ángel Viloria, Andrew Neild, and Patrick Blandin for the butterfly identifications.

REFERENCES

• Acharya, B.K.; Vijayan, L. 2015. Butterfly diversity along the elevation gradient of Eastern Himalaya, India.Ecological Research30: 909-919.
• Afzal, S.; Nesar, H.; Imran, Z.; Ahmad, W. 2021. Altitudinal gradient affect abundance, diversity and metabolic footprint of soil nematodes in Banihal-Pass of Pir-Panjal mountain range.Scientific Reports11: 16214.
• Álvarez, C.F.; Clavijo-Giraldo, A.; Inés Uribe, S.; Pyrcz, T.W.; Iserhard, C.A.; Lucci Freitas, A.V.; Marín, M.A. 2021. Sampling performance of bait traps in high Andean fruit-feeding butterflies. Neotropical Biodiversity 7: 507-513.
• Antonelli, A.; Kissling, W.D.; Flantua, S.G.; Bermúdez, M.A.; Mulch, A.; Muellner-Riehl, A.N.; et al. 2018. Geological and climatic influences on mountain biodiversity.Nature Geoscience11: 718-725.
• Baselga, A. 2010. Partitioning the turnover and nestedness components of beta diversity. Global Ecology and Biogeography 19: 134-143.
• Beirão, M.V.; Neves, F.S.; Fernandes, G.W. 2020. Climate and plant structure determine the spatiotemporal butterfly distribution on a tropical mountain.Biotropica53: 191-200.
• Berrones-Morales, M.; Vanoye-Eligio, V.; Coronado-Blanco, J.M.; Gaona-García, G.; Sánchez-Ramos, G. 2020. Species diversity of fruit flies (Diptera: Tephritidae) through different ecosystems in a Neotropical transition zone in Mexico.Journal of Insect Conservation24: 219-231.
• Berry, P.E.; Riina, R. 2005. Insights into the diversity of the Pantepui flora and the biogeographic complexity of the Guayana Shield.Biologiske Skrifter55: 145-167.
• Bonebrake, T.C.; Ponisio, L.C.; Boggs, C.L.; Ehrlich, P.R. 2010. More than just indicators: a review of tropical butterfly ecology and conservation.Biological Conservation143: 1831-1841.
• Brasil. 2022. Ministério do Meio Ambiente- MMA. Instrução normativa nº 2 de 28 de Janeiro de 2022. Reformula conceitos, princípios, finalidades, instrumentos e procedimentos para a implementação do Programa Nacional de Monitoramento da Biodiversidade do Instituto Chico Mendes - Programa Monitora (Processo 02070.023604/2021-69). ( (https://www.gov.br/icmbio/pt-br/acesso-a-informacao/institucional/legislacao/instrucoes-normativas/Instruesnormativas.pdf ). Accessed on 8 Aug 2024.
  » https://www.gov.br/icmbio/pt-br/acesso-a-informacao/institucional/legislacao/instrucoes-normativas/Instruesnormativas.pdf
• Briceño, H.O.; Schubert, C. 1990. Geomorphology of the Gran Sabana, Guayana Shield, southeastern Venezuela. Geomorphology 3: 125-141.
• Carneiro, E.; Mielke, O.H.H.; Casagrande, M.M.; Fiedler, K. 2014. Community structure of skipper butterflies (Lepidoptera, Hesperiidae) along elevational gradients in Brazilian Atlantic forest reflects vegetation type rather than altitude.Plos One9: e108207.
• Carvalho, R.L.; Resende, A.F.; Barlow, J.; França, F.M.; Moura, M.R.; Maciel, R.; et al. 2023. Pervasive gaps in Amazonian ecological research.Current Biology33: 3495-3504.
• Castro, D.M.; Callisto, M.; Solar, R.R.; Macedo, D.R.; Fernandes, G.W. 2019. Beta diversity of aquatic invertebrates increases along an altitudinal gradient in a Neotropical mountain.Biotropica51: 399-411.
• Castro, F.S.D.; Da Silva, P.G.; Solar, R.; Fernandes, G.W.; Neves, F.D.S. 2020. Environmental drivers of taxonomic and functional diversity of ant communities in a tropical mountain. Insect Conservation and Diversity 13: 393-403.
• Chao, A.; Gotelli, N.J.; Hsieh, T.C.; Sander, E.L.; Ma, K.H.; Colwell, R.K.; et al. 2014. Rarefaction and extrapolation with Hill numbers: A framework for sampling and estimation in species diversity studies. Ecological Monographs 84: 45-67.
• Chazot, N.; Willmott, K.R.; Santacruz Endara, P.G.; Toporov, A.; Hill, R.I.; Jiggins, C.D.; et al. 2014. Mutualistic mimicry and filtering by altitude shape the structure of Andean butterfly communities.The American Naturalist183: 26-39.
• Chowdhury, S. 2023. Threatened species could be more vulnerable to climate change in tropical countries. Science of the Total Environment 858: 159989.
• Colwell, R.K.; Chao, A.; Gotelli, N.J.; Lin, S-Y.; Mao, C.X.; Chazdon, R.L.; et al. 2012. Models and estimators linking individual- based and sample-based rarefaction, extrapolation and comparison of assemblages. Journal of Plant Ecology 5: 3-21.
• Cómbita, J.L.; Giraldo, C.E.; Escobar, F. 2022. Environmental variation associated with topography explains butterfly diversity along a tropical elevation gradient. Biotropica 54: 146-156.
• Costa, M.; Viloria, Á.L.; Attal, S.; Benmesbah, M.; Orellana, A.; Neild, A.F.E.; Lamas, G.; Blandin, P. 2025. Lepidoptera from the Pantepui. Part XVII. A new subspecies of Caligo suzanna (Deyrolle, 1872). Antenor 11: 150-184.
• Costa, M.; Viloria, Á.L.; Attal, S.; Neild, A.F.E.; Fratello, S.; Callaghan, C.; et al. 2017. Lepidoptera del Pantepui. Parte IV. Nuevos Riodinidae Riodininae y Pieridae Pierinae. Bulletin de la Societé Entomologique de France 122: 269-286.
• Costa, M.; Viloria, Á.L.; Attal, S.; Orellana, A.; Benmesbah, M. 2019. Lepidoptera del Pantepui. Parte VI. Nuevos Hesperiidae Hesperiinae y Nymphalidae Satyrinae, Nymphalinae. Bulletin de la Societé Entomologique de France 124: 77-102.
• Costa, M.; Viloria, Á.L.; Attal, S.; Orellana, A.; Neild, A.F.E.; Benmesbah, M.; et al. 2020. Lepidoptera del Pantepui. Parte X. Nuevos Pieridae (Dismorphiinae) y Hesperiidae (Pyrrhopyginae). Antenor 7: 82-105.
• Costa, M.; Viloria, Á.L.; Attal, S.; Orellana, A. 2014a. Lepidoptera del Pantepui. Parte II. Descripción n de nuevos Nymphalidae (Papilionoidea). Bulletin de la Societé entomologique de France 119: 39-52.
• Costa, M.; Viloria, Á.L.; Huber, O.; Attal, S.; Orellana, A. 2014b. Lepidoptera del Pantepui. Parte I. Endemismo y caracterización biogeográfica. Entomotrópica 28: 193-217.
• Crawley, M.J. 2013. The R book. Chichester, West Sussex, United Kingdom: Wiley. 1051p.
• Dambros, C.S. 2020. Csdambros/R-fuctions: First release. (https://github.com/csdambros/R-functions/tree/v1.0).
  » https://github.com/csdambros/R-functions/tree/v1.0
• Dennis, R.L.; Hodgson, J.G.; Grenyer, R.; Shreeve, T.G.; Roy, D.B. 2004. Host plants and butterfly biology. Do host‐plant strategies drive butterfly status?Ecological Entomology29: 12-26.
• DeVries, P.J. 1986. Hostplant records and natural history notes on Costa Rican butterflies (Papilionidae, Pieridae & Nymphalidae).Journal of Research on the Lepidoptera24: 290-333.
• Dzekashu, F.F.; Yusuf, A.A.; Pirk, C.W.; Steffan‐Dewenter, I.; Lattorff, H.M.G.; Peters, M.K. 2022. Floral turnover and climate drive seasonal bee diversity along a tropical elevation gradient.Ecosphere13: e3964.
• Forister, M.L.; McCall, A.C.; Sanders, N.J.; Fordyce, J.A.; Thorne, J.H.; O’Brien, J.; et al. 2010. Compounded effects of climate change and habitat alteration shift patterns of butterfly diversity. Proceedings of the National Academy of Sciences 107: 2088-2092.
• Fratello, S.A.; Nakahara, S.; Brévignon, C.R.; Harvey, D.J. 2015. Two new species of Euptychia Hübner, 1818 (Lepidoptera: Nymphalidae: Satyrinae) from the Guiana Shield, with notes on E. marceli Brévignon, 2005 and E. rufocincta Weymer, 1911. Journal of the Lepidopterists’ Society 69: 293-306.
• Freitas, A.V.L.; Gueratto, P.E.; Carreira, J.Y.O.; Lourenço, G.M.; Shirai, L.T.; Santos, J.P; et al. 2024. Fruit-Feeding Butterfly Assemblages: Trends, Changes, and the Importance of Monitoring Schemes in Neotropical Environments. In: León-Cortés, J.L.; Córdoba-Aguilar, A. (Eds.). Insect Decline and Conservation in the Neotropics Springer International Publishing, Switzerland. p. 205-233.
• Freitas, A.V.L.; Iserhard, C.A.; Santos, J.P.; Carreira, J.Y.O.; Ribeiro, D.B.; Melo, D.H.A.; et al. 2014. Studies with butterfly bait traps: an overview. Revista Colombiana de Entomología 40: 203-212.
• Henriques, N. R.; Beirão, M.D.V.; Brasil, E.; Cornelissen, T. 2019. Butterflies (Lepidoptera: Papilionoidea) from the campos rupestres of Serra de São José, Minas Gerais, Brazil. Biota Neotropica 19: e20180655.
• Henriques, N.R.; Lourenço, G.M.; Diniz, E.S.; Cornelissen, T. 2022. Is elevation a strong environmental filter? Combining taxonomy, functional traits and phylogeny of butterflies in a tropical mountain.Ecological Entomology47: 613-625.
• Hodkinson, I.D. 2005. Terrestrial insects along elevation gradients: species and community responses to altitude. Biological Reviews 80: 489-513.
• Hsieh, T.C.; Ma, K.H.; Chao, A. 2016. iNEXT: An R package for rarefaction and extrapolation of species diversity (Hill numbers). Methods in Ecology and Evolution 7: 1451-1456.
• Huber, O. 1987. Consideraciones sobre el concepto de Pantepui. Pantepui 2: 2-10.
• Huber, O. 1995. Geographical and physical features. In: Berry, P.; Holst, B.K.; Yaskievych, K. (Eds.). Flora of the Venezuelan Guayana, Introduction Missouri Botanical Garden Press, St. Louis. p.1-51.
• Koh, L.P.; Sodhi, N.S.; Brook, B.W. 2004. Ecological correlates of extinction proneness in tropical butterflies. Conservation Biology 18: 1571-1578.
• Körner, C. 2007. The use of ‘altitude’ in ecological research.Trends in Ecology & Evolution22: 569-574.
• Kumar, S.; Simonson, S.E.; Stohlgren, T.J. 2009. Effects of spatial heterogeneity on butterfly species richness in Rocky Mountain National Park, CO, USA.Biodiversity and Conservation18: 739-763.
• Legendre, P. 2014. Interpreting the replacement and richness difference components of beta diversity. Global Ecology and Biogeography 23: 1324-1334.
• Lew, D.; Lim, B.K. 2019. Mammals. In: Rull, V.; Vegas-Vilarrúbia, T.; Huber, O.; Señaris, C. (Eds.). The Pristine “Lost World” of the Neotropical Guiana Highlands Elsevier, London, UK. p.333-372.
• Lourenço, G.M.; Dattilo, W.; Ribeiro, S.P.; Freitas, A.V. 2021. Biological aspects and movements of Neotropical fruit-feeding butterflies. Neotropical Entomology 51: 43-53.
• Manly, B.F.J. 2007. Randomization, bootstrap and Monte Carlo methods in biology (3rd ed.). Chapman and Hall/CRC. 480p.
• Marín, M.A.; Giraldo, C.E.; Marín, A.L.; Álvarez, C.F.; Pyrcz, T.W. 2015. Differences in butterfly (Nymphalidae) diversity between hillsides and hilltop forest patches in the northern Andes. Studies on Neotropical Fauna and Environment 50: 194-203.
• Mayr, E.; Phelps Jr, W.H. 1967. The origin of the bird fauna of the South Venezuelan highlands. Bulletin of the AMNH 136: 269328.
• Montañez-Reyna, M.; Leon-Cortes, J.L.; Falcon-Brindis, A. 2023. The diversity of moths (Erebidae: Arctiinae: Arctiini) from threatened mountain cloud forests in the Mesoamerican biodiversity hotspot. Journal of Insect Conservation 27: 141-154.
• Morrone, J.J. 2017. Neotropical biogeography: Regionalization and evolution. CRC Press, Boca Raton. 312p.
• Nakahara, S.; Fratello, S.A.; Harvey, D.J. 2014. A new species of Euptychia Hübner, 1818 (Lepidoptera: Nymphalidae: Satyrinae: Satyrini) from Mount Roraima, Guyana. Zootaxa 3881: 291-300.
• Nobre, R.A., Kinouche, M.R., Constantino, P.A.L., Pereira, R.C., Uehara-Prado, M. 2014. Monitoramento da Biodiversidade: Roteiro Metodológico de aplicação. Brasília, ICMBio. 43p.
• Nogué, S.; Rull, V.; Vegas‐Vilarrúbia, T. 2012. Elevational gradients in the neotropical table mountains: patterns of endemism and implications for conservation.Diversity and Distributions19: 676-687.
• Nogueira, A.A.; Brescovit, A.D.; Perbiche-Neves, G.; Venticinque, E.M. 2021. Beta diversity along an elevational gradient at the Pico da Neblina (Brazil): is spider (Arachnida-Araneae) community composition congruent with the Guayana region elevational zonation?Diversity13: 620.
• Ochoa, J.A.; Rojas-Runjaic, F.J.M. 2019. Scorpions. In: Rull, V.; Vegas-Vilarrúbia, T.; Huber, O.; Señaris, C. (Eds.). The Pristine “Lost World” of the Neotropical Guiana Highlands Elsevier, London, UK . p.223-241.
• Oksanen, J.; Blanchet, F.G.; Friendly, M.; Kindt, R.; Legendre, P.; McGlinn, D.; et al. 2019. Vegan: Community Ecology Package. R package version 2.5-6. Available online: https://CRAN.R-project.org/package=vegan
  » https://CRAN.R-project.org/package=vegan
• Oliveira, I.F.; Baccaro, F.B.; Werneck, F.P.; Zacca, T.; Haugaasen, T. 2021. Marked differences in butterfly assemblage composition between forest types in Central Amazonia, Brazil. Forests 12: 942.
• Perillo, L.N.; Neves, F.D.S.; Antonini, Y.; Martins, R.P. 2017. Compositional changes in bee and wasp communities along Neotropical mountain altitudinal gradient. PLoS One 12: e0182054.
• Perrigo, A.; Hoorn, C.; Antonelli, A. 2020. Why mountains matter for biodiversity. Journal of Biogeography47: 315-325.
• Pires, A.C.V.; Barbosa, M.; Beiroz, W.; Beirão, M.V.; Marini-Filho, O.J.; Duarte, M.; et al. 2020. Altitudinal variation in butterfly community associated with climate and vegetation.Anais da Academia Brasileira de Ciências92: e20190058.
• Pyrcz, T.W.; Clavijo, A.; Uribe, S.; Marín, M.A.; Álvarez, C.F.; Zubek, A. 2016. Páramo de Belmira as an important centre of endemism in the northern Colombian Andes: New evidence from Pronophilina butterflies (Lepidoptera: Nymphalidae, Satyrinae, Satyrini). Zootaxa 4179: 77-102.
• Pyrcz, T.W.; Fratello, S. 2005. Cloud forest butterfly fauna of the Pantepui-Poor or poorly known? Description of new species and records of new genera of Pronophilina: Eretris agata and Oxeoschistus romeo (Nymphalidae: Satyrinae).Journal of the Lepidopterists’ Society59: 200.
• Pyrcz, T.W.; Rodríguez, G. 2007. Mariposas de la tribu Pronophilini en la Cordillera Occidental de los Andes de Colombia (Lepidoptera: Nymphalidae: Satyrinae). SHILAP Revista de Lepidopterología 35: 455-489.
• Pyrcz, T.W.; Wojtusiak, J.; Garlacz, R. 2009. Diversity and distribution patterns of Pronophilina Butterflies (Lepidoptera: Nymphalidae: Satyrinae) along an altitudinal transect in North-Western Ecuador. Neotropical Entomology 38: 716-726.
• Quintero, I.; Jetz, W. 2018. Global elevational diversity and diversification of birds.Nature555: 246-250.
• R Core Team. 2021. R: A language and environment for statistical computing R Foundation for Statistical Computing.
• Rabelo, R.M.; Oliveira, I.F.; Magnusson, W.E. 2021. Finding a lost species in the ‘Lost World’: predicted habitat occupancy by an endemic butterfly in a Neotropical sky‐island archipelago.Insect Conservation and Diversity15: 128-135.
• Rahbek, C. 1997. The relationship among area, elevation, and regional species richness in neotropical birds.The American Naturalist149: 875-902.
• Rahbek, C.; Borregaard, M.K.; Antonelli, A.; Colwell, R.K.; Holt, B.G.; Nogues-Bravo, D.; et al. 2019. Building mountain biodiversity: Geological and evolutionary processes.Science365: 1114-1119.
• Ribeiro, D.B.; Batista, R.; Prado, P.I.; Brown, K.S.; Freitas, A.V. 2012. The importance of small scales to the fruit-feeding butterfly assemblages in a fragmented landscape. Biodiversity and Conservation 21: 811-827.
• Riina, R.; Berry, P.E.; Huber, O.; Michelangeli, F.A. 2019. Vascular plants and bryophytes. In: Rull, V.; Vegas-Vilarrúbia, T.; Huber, O.; Señaris, C. (Eds.). The Pristine “Lost World” of the Neotropical Guiana Highlands Elsevier, London, UK . p.121-147.
• Rödder, D.; Schmitt, T.; Gros, P.; Ulrich, W.; Habel, J.C. 2021. Climate change drives mountain butterflies towards the summits. Scientific Reports 11: 14382.
• Rull, V. 2019. Origin and evolution of the Pantepui biota. In: Rull, V.; Vegas-Vilarrúbia, T.; Huber, O.; Señaris, C. (Eds.). The Pristine “Lost World” of the Neotropical Guiana Highlands Elsevier, London, UK . p. 69-91.
• Rull, V.; Huber, O.; Vegas-Vilarrúbia, T.; Señaris, C. 2019a. Definition and characterization of the Pantepui biogeographical province. In: Rull, V.; Vegas-Vilarrúbia, T.; Huber, O.; Señaris, C. (Eds.). The Pristine “Lost World” of the Neotropical Guiana Highlands Elsevier, London, UK . p. 3-32.
• Rull, V.; Nogué, S.; Safont, E.; Vegas-Vilarrúbia, T. 2019b. Pantepui and global warming. In: Rull, V.; Vegas-Vilarrúbia, T.; Huber, O.; Señaris, C. (Eds.). The Pristine “Lost World” of the Neotropical Guiana Highlands Elsevier, London, UK . p. 403-417.
• Rull, V.; Vegas-Vilarrúbia, T. 2017. Potential responses of vascular plants from the pristine “Lost World” of the Neotropical Guayana highlands to global warming: Review and new perspectives.Frontiers in Plant Science8: 81.
• Rydon, A. 1964. Notes on the use of butterfly traps in East Africa. Journal of the Lepidopterist’s Society 18: 51-58.
• Safont, E.; Rull, V.; Vegas-Vilarrúbia, T.; Holst, B.K.; Huber, O.; Nozawa, S.; et al. 2014. Establishing a baseline of plant diversity and endemism on a neotropical mountain summit for future comparative studies assessing upward migration: an approach from biogeography and nature conservation. Systematics and Biodiversity 12: 292-314.
• Socolar, J.B.; Gilroy, J.J.; Kunin, W.E.; Edwards, D.P. 2016. How should beta-diversity inform biodiversity conservation?.Trends in Ecology & Evolution31: 67-80.
• Steffan‐Dewenter, I.; Tscharntke, T. 2000. Butterfly community structure in fragmented habitats.Ecology Letters3: 449-456.
• Viloria, Á.L.; Costa, M. 2019. Butterflies. In: Rull, V.; Vegas-Vilarrúbia, T.; Huber, O.; Señaris, C. (Eds.). The Pristine “Lost World” of the Neotropical Guiana Highlands Elsevier, London, UK . p. 193-222.
• Zuur, A.F.; Ieno, E.N.; Walker, N.J.; Saveliev, A.A.; Smith, G.M. 2009. Mixed effects models and extensions in ecology with R. New York: Springer. 563p.

• CITE AS:
  Oliveira, I.F.; Costa, M.; Rabelo, R.M.; Werneck, F.P.; Baccaro, F.B. 2025. Butterfly (Insecta: Lepidoptera) assemblage structure along an upland-highland elevational gradient in a Brazilian tepui. Acta Amazonica 55: e55bc24182.

Data availability

All data supporting the findings of this study are available in the Supplementary Material.

SUPPLEMENTARY MATERIAL

Oliveira et al. Butterfly (Insecta: Lepidoptera) assemblage structure along an upland-highland elevational gradient in a Brazilian tepui

Table S1
The mean altitude of each sampling plot.

Transect	Plot	Altitude (m a.s.l.)
SS1	P1	1044
SS1	P2	1042
SS1	P3	1072
SS1	P4	1108
SS2	P1	1014
SS2	P2	1022
SS2	P3	1016
SS2	P4	1024
SS3	P1	1330
SS3	P2	1390
SS3	P3	1565
SS3	P4	1582
SS4	P1	1820
SS4	P2	1820

Table S2
The complete list of butterfly species and respective abundances sampled in the first biological expedition to the Brazilian side of Uei Tepui, and the distinct sampling methods used: standardized methods [understory (U) and canopy (C) traps (T) in the sampling plots, and transect walk (TW) using entomological net], and opportunistically surveys (OS, t = traps out of the plots, and N = entomological net).

Family/Subfamily/Species	Altitude (m a.s.l.)	Methods	N˚ of individuals
Hesperiidae			2
Hesperiinae			1
Parphorus storax (Mabille, 1891)	1019	TW	1
Pyrginae			1
Hyalothyrus infernalis (Möschler, 1877)	956	OS (N)	1
Lycaenidae			3
Theclinae			3
Arawacus aetolus (Sulzer, 1776)	1021	TW	1
Calycopis sp Scudder, 1876	1018	TW	2
Nymphalidae			279
Biblidinae			24
Callicore cynosura amazona (H. Bates, 1864)	1041	T (C)	1
Catonephele acontius (Linnaeus, 1771)	from 1015 to 1112	T (U,C), OS (N)	13
Catonephele numilia (Cramer, 1776)	from 1013 to 1109	T (C)	2
Eunica viola Bates, 1864	1043	T (C)	1
Nessaea batesii magniplaga Röber, 1928	from 1015 to 1018	T (U), OS (t)	5
Temenis laothoe laothoe (Cramer, 1777)	from 1014 to 1018	T (C), OS (t)	2
Charaxinae			27
Archaeoprepona d. demophon (Linnaeus, 1758)	from 1018 to 1073	T(C), OS (t)	2
Archaeoprepona licomedes (Cramer, 1777)	1018	OS (t)	3
Memphis a. acidalia (Hübner, [1819])	from 1048 to 1063	T (U,C)	3
Memphis laertes (Cramer, 1775)	from 1043 to 1048	T (U,C)	3
Memphis montesino* Pyrcz, 1995	1388	T (C)	1
Memphis phantes (Hopffer, 1874)	1580	T (C)	1
Mesoprepona pheridamas (Cramer, 1777)	1018	OS (t)	1
Prepona claudina (Godart, [1824])	from 1014 to 1048	T (C)	5
Prepona eugenes eugenes H. Bates, 1865	from 1013 to 1071	T (C), OS (t)	4
Prepona laertes demodice (Godart, [1824])	from 1018 to 1109	T (C), OS (N)	3
Zaretis itys (Cramer, 1777)	1018	OS (N)	1
Cyrestinae			1
Marpesia furcula oechalia (Westwood, 1850)	1019	TW	1
Danainae			15
Greta clavijoi* Neild, 2008	from 1486 to 1571	TW, OS (N)	3
Hyposcada zarepha bonplandii* Neild, 2008	from 1571 to 1581	TW, OS (N)	2
Hypothyris ninonia connexa Hall, 1939	from 1021 to 1071	TW	3
Lycorea halia halia (Hübner, 1816)	1013	TW	1
Melinaea mneme mauensis Weymer, 1891	1018	OS (N)	1
Melinaea satevis crameri Godman & Salvin, 1898	from 920 to 1018	OS (N)	2
Oleria boyeri* Neild, 2008	1581	TW	2
Oleria n. sp Hübner, 1816	1571	OS (N)	1
Heliconiinae			14
Dryas iulia*** (Fabricius, 1775)	1047	T (C)	1
Heliconius erato magnifica Riffarth, 1900	from 920 to 1019	TW, OS (N)	6
Heliconius melpomene pyrforus Kaye, 1907	1486	TW	2
Heliconius numata superioris Butler, 1875	from 1016 to 1026	TW	4
Heliconius sara thamar (Fabricius, 1793)	1021	TW	1
Limenitidinae			2
Adelpha delinita delinita Fruhstorfer, 1913	1018	OS (N)	1
Adelpha thesprotia (C. Felder & R. Felder, 1867)	1014	T (C)	1
Nymphalinae			9
Anartia jatrophae*** (Linnaeus, 1763)	1021	TW	2
Baeotus japetus (Staudinger, [1885])	1077	T (C)	1
Colobura dirce (Linnaeus, 1758)	1018	OS (N)	2
Junonia evarete oscura* Neild, 2008	920	OS (N)	1
Junonia genoveva vivida W.T.M. Forbes, [1929]	920	OS (N)	2
Siproeta stelenes meridionalis (Fruhstorfer, 1909)	1018	OS (N)	1
Satyrinae			187
Amiga arnaca (Fabricius, 1776)	from 956 to 1037	T (U), TW OS (N)	11
Antirrhea ulei* Strand, 1912	from 1388 to 1820	T (U,C), TW, OS (N)	14
Bia actorion actorion (Linnaeus, 1763)	from 956 to 1056	TW, OS (t,N)	4
Caeruleuptychia urania (A. Butler, 1867)	from 1013 to 1106	T (U)	2
Caligo idomeneus (Linnaeus, 1758)	from 1016 to 1063	T (U)	5
Caligo illioneus illioneus (Cramer, 1775)	920	OS (N)	1
Caligo suzanna weidmanni* Blandin, Attal, Orellana, Lamas & Costa, 2025	from 1018 to 1571	T (C), OS (t,N),	3
Catoblepia berecynthia halli Bristow, 1981	from 1022 to 1047	T (U,C)	3
Catoblepia xanthus (Linnaeus, 1758)	1109	OS (t)	1
Cithaerias andromeda andromeda (Fabricius, 1775)	from 1021 to 1392	TW	5
Emeryus argulus magnum Zacca, Casagrande & Mielke, 2020	920	OS (N)	1
Eryphanis automedon automedon (Cramer, 1775)	1047	OS (t)	1
Haetera piera piera (Linnaeus, 1758)	from 1017 to 1392	TW	3
Hermeuptychia canthe [Hübner, 1811]	from 920 to 1026	TW, OS (N)	7
Manataria maculata maculata (Hopffer, 1874)	1571	OS (N)	1
Modesta h. harpyia (C. Felder & R. Felder, 1867)	1037	T (C)	1
Modica myncea (Cramer, 1780)	1019	T (U), TW	2
Morpho achilles (Linnaeus, 1758)	from 1013 to 1015	T (U), OS (t,N)	3
Morpho deidamia (Hübner, [1819])	from 1016 to 1106	T (U), OS (t,N)	9
Morpho helenor extremus Le Moult, 1933	927	OS (N)	1
Nhambikuara furina (Hewitson, 1862)	from 1019 to 1109	T (U,C), TW, OS (N)	16
Opsiphanes cassina numatius Fruhstorfer, 1912	1014	T (C)	1
Opsiphanes invirae sieberti Bristow, 1991	from 1013 to 1048	T (C)	9
Oressinoma typhla** E. Doubleday, [1849]	from 1538 to 1571	TW, OS (N)	2
Pareuptychia ocirrhoe (Fabricius, 1776)	1021	TW	1
Pareuptychia hervei Brévignon, 2005	1019	T (U)	1
Pareuptychia lydia (Cramer, 1777)	from 1022 to 1048	T (C), TW	6
Pareuptychia metaleuca tekolokem Brévignon, 2005	from 1019 to 1041	T (U,C)	2
Pierella astyoche (Erichson, [1849])	from 1017 to 1045	TW	2
Pierella hyalinus hyalinus (Gmelin, [1790])	from 1014 to 1090	T (U), TW	5
Pierella lamia (Sulzer, 1776)	from 1039 to 1042	TW	3
Pierella lena lena (Linnaeus, 1767)	1019	TW	1
Posttaygetis penelea (Cramer, 1777)	from 1014 to 1017	T (U), TW	3
Taygetina gulnare (A. Butler, 1870)	1019	TW	2
Taygetis laches (Fabricius, 1793)	from 1013 to 1538	T (U), TW, OS (N)	48
Taygetis leuctra A. Butler, 1870	1017	T (U)	1
Taygetis mermeria (Cramer, 1776)	1022	T (U)	1
Taygetis virgilia (Cramer, 1776)	from 1014 to 1037	T (U), OS (t)	5
Pieridae			7
Coliadinae			1
Eurema albula albula*** (Cramer, 1775)	1021	TW	1
Dismorphinae			1
Dismorphia zathoe proserpina * Grose-Smith & W.F.Kirby, 1897	1571	OS (N)	1
Pierinae			5
Archonias brassolis (Fabricius, 1776)	1019	TW	1
Archonias sisamnus ayanganna* Bollino & Costa, 2007	1571	OS (N)	1
Melete leucardia reyi* A.F.E. Neild & Costa, 2017	927	OS (N)	1
Pereute lindemannae pemona* de Marmels, Clavijo & M. Chacín, 2003	1571	OS (N)	2
Riodinidae			8
Riodininae			8
Detritivora cleonus (Stoll, 1781)	from 1017 to 1071	TW	2
Eurybia nicaeus nicaeus (Fabricius, 1775)	1016	TW	1
Mesosemia maera Hewitson, 1873	1486	TW	1
Mesosemia sp. Hübner, [1819]	1017	TW	1
Nymphidium menalcus (Stoll, 1782)	from 1018 to 1071	TW, OS (N)	2
Pirascca sagaris (Cramer, 1775)	1021	TW	1
Total			299

• * Endemic to the Pantepui (Viloria and Costa 2019; Costa et al. 2025). With the opportunistic sampling, we were also able to record endemic species outside the sampling plots. Some of these were found below 1330 m a.s.l.; however, this is consistent with studies on butterflies from the Pantepui region, where the butterfly endemism appears gradually from ~1000 m a.s.l. and increases up to 1500 m (Viloria and Costa 2019).
• ** First record in Brazil, but it is distributed from Costa Rica to Ecuador, Colombia, Venezuela, and Bolivia (Losada et al. 2018; pers. comm.).
• *** Common butterfly species in open areas and not usually sampled in the forest.

Table S3
Butterfly species recorded in the sampling plots of this study, and the butterfly records in different studies across the Brazilian Amazon lowlands. x = recorded species; x sp = same species recorded but not the same subspecies.

This study	Emmel and Austin (1990)	Ramos (2000)	Barlow et al. (2007)	Mielke et al. (2010)	Casagrande et al. (2012)	Ribeiro and Freitas (2012)	Graça et al. (2017a)	Graça et al. (2017b)	Teston and Silva (2017)	Spaniol et al. (2019)	Oliveira et al. (2021)	Rabelo et al. (2021)	Mota et al. (2022)
A. thesprotia	x												x
A. arnaca	x	x							x	x	x		x
A. jatrophae***				x
A. ulei*
A. aetolus					x
A. d. demophon	x	x	x	x	x	x	x		x	x	x	x	x
A. brassolis
B. japetus			x			x				x			x
B. a. actorion	x	x	x	x	x	x	x	x	x	x	x	x
C. urania
C. idomeneus	x		x	x				x	x	x			x
C. suzanna weidmanni*			x sp
C. c. amazona	x sp		x sp										x sp
Calycopis sp
C. b. halli	x sp	x sp	x sp	x sp	x sp	x sp			x sp	x sp	x sp	x sp	x sp
C. acontius	x	x			x	x	x	x	x	x	x	x	x
C. numilia		x				x					x		x
C. a. andromeda			x		x			x		x
D. cleonus	x			x	x
D. iulia***	x			x sp	x						x
E. viola
E. a. albula***	x			x							x		x
E. n. nicaeus	x			x	x						x		x
G. clavijoi*
H. p. piera	x			x sp	x			x	x	x	x	x	x
H. e. magnifica	x sp			x sp	x sp						x sp	x sp	x sp
H. m. pyrforus	x sp			x sp	x sp
H. n. superiores	x sp			x sp	x						x sp		x
H. s. thamar	x sp		x sp	x							x sp	x sp	x sp
H. canthe
H. z. bonplandii*
H. n. connexa	x sp												x sp
L. h. halia				x sp									x sp
M. myncea	x		x			x		x			x		x
M. h. harpyia			x		x sp			x			x sp		x sp
M. f. oechalia
M. a. acidalia			x	x sp		x	x		x	x			x
M. laertes			x			x	x	x
M. montesino*
M. phantes			x	x		x		x		x	x sp
M. maera
Mesosemia sp
M. achilles	x	x sp	x	x sp	x				x		x	x	x sp
M. deidamia	x		x		x				x			x	x
N. b. magniplaga
N. furina			x		x
N. menalcus
O. boyeri*
O. c. numatius			x sp
O. i. sieberti	x sp	x sp	x sp	x sp		x sp	x sp	x sp		x sp	x sp	x sp	x sp
O. typhla**
P. ocirrhoe		x							x			x	x
P. hervei
P. lydia			x			x				x			x
P. m. tekolokem		x sp
P. storax	x			x
P. astyoche	x			x sp	x			x		x	x		x sp
P. h. hyalinus	x		x		x			x					x
P. lamia	x				x			x		x
P. l. lena	x			x	x		x			x	x sp	x	x
P. sagaris
P. penelea	x		x	x			x				x	x	x
P. claudina	x		x	x sp		x	x	x	x				x
P. e. eugenes													x
P. l. demodice		x sp	x	x	x sp					x			x sp
T. gulnare													x
T. laches		x	x		x	x	x	x	x	x	x		x
T. leuctra		x	x
T. mermeria	x	x	x						x		x	x	x
T. virgilia	x	x	x		x				x		x	x	x
T. l. laothoe	x		x	x		x	x		x	x	x	x	x

• * Endemic from Pantepui (Viloria and Costa 2019; Costa et al. 2025)
• ** First record in Brazil, but it is distributed from Costa Rica to Ecuador, Colombia, Venezuela, and Bolivia (Losada et al. 2018; pers. comm.).
• *** Common butterfly species in open areas and not usually sampled in the forest.
• Butterfly studies in the Brazilian Amazonia lowlands: Emmel and Austin (1990) in Rondônia State; Barlow et al. (2007) in Amapá and Pará States; Graça et al. (2017a) in Roraima State; Casagrande et al. (2012), Ribeiro and Freitas (2012), Graça et al. (2017b), Spaniol et al. (2019), Oliveira et al. (2021), and Rabelo et al. (2021) in Amazonas State, and Mota et al. (2022) in Mato Grosso State.

Figure S1
(A) Sample completeness and (B) individual-based rarefaction curves for upland and highland elevations (triangles and circles, respectively) of butterfly assemblages in the Uei Tepui, Roraima State. Solid lines represent the interpolated and dashed lines the extrapolates values. Shaded areas represent 95% confidence intervals.

Figure S2
Direct ordination of all butterfly species sampled in this study (standardized and opportunistically sampled) and their relative abundance (bar thickness) along the elevational gradient of the Uei-tepui, Roraima State, Brazil. * butterflies endemic to Pantepui; † new species record to Brazil.

REFERENCES

Barlow, J.O.S.; Overal, W.L.; Araujo. I.S.; Gardner, T.A.; Peres, C.A. 2007. The value of primary, secondary and plantation forests for fruit‐feeding butterflies in the Brazilian Amazon.Journal of Applied Ecology44: 1001-1012.

Casagrande, M.M.; Mielke, O.H.; Carneiro, E.; Rafael, J.A.; Hutchings, R.W.; 2012. Hesperioidea e Papilionoidea (Lepidoptera) coligidos em expedição aos Rios Nhamundá e Abacaxis, Amazonas, Brasil: novos subsídios para o conhecimento da biodiversidade da Amazônia Brasileira.Revista Brasileira de Entomologia56: 23-28.

Costa, M.; Viloria, Á.L.; Attal, S.; Benmesbah, M.; Orellana, A.; Neild, A.F.E.; Lamas, G.; Blandin, P. 2025. Lepidoptera from the Pantepui. Part XVII. A new subspecies of Caligo suzanna (Deyrolle, 1872). Antenor 11: 150-184.

Emmel, T.C.; Austin, G.T. 1990. The Tropical Rain Forest Butterfly Fauna of Rondonia, Brazil: Species Diversity and Conservation. Tropical Lepidoptera 1: 1-12.

Graça, M.B.; Pequeno, P.A.; Franklin, E.; Souza, J.L.; Morais, J.W. 2017a Taxonomic, functional, and phylogenetic perspectives on butterfly spatial assembly in northern Amazonia.Ecological Entomology42: 816-826.

Graça, M.B.; Souza, J.L.; Franklin, E.; Morais, J.W.; Pequeno, P.A. 2017b. Sampling effort and common species: optimizing surveys of understorey fruit-feeding butterflies in the Central Amazon.Ecological Indicators73: 181-188.

Losada, M.E.; Neild, A.F.E.; Viloria, Á.L. 2018. The life cycle of Oressinoma typhla Doubleday, [1849] (Lepidoptera: Nymphalidae: Satyrinae). Tropical Lepidoptera Research 28: 54-60.

Mielke, O.H.H.; Carneiro, E.; Casagrande, M.M. 2010. Lepidopterofauna (Papilionoidea e Hesperioidea) do Parque Estadual do Chandless e arredores, Acre, Brasil. Biota Neotropica 10: 285-299.

Mota, L.L.; Boddington, S.J.; Brown Jr, K.S.; Callaghan, C.J.; Carter, G.; Carter, W.; et al. 2022. The butterflies of Cristalino Lodge, in the Brazilian southern Amazonia: An updated species list with a significant contribution from citizen scienceBiota Neotropica22.

Rabelo, R.M.; Pereira, G.C.N.; Valsecchi, J.; Magnusson, W.E. 2021. The role of river flooding as an environmental filter for Amazonian butterfly assemblages. Frontiers in Ecology and Evolution 9: 693178.

Ramos, F.A. 2000. Nymphalid butterfly communities in an Amazonian forest fragment. Journal of Research on the Lepidoptera 35: 29-41.

Ribeiro, D.B.; Freitas, A.V.L. 2012. The effect of reduced-impact logging on fruit-feeding butterflies in Central Amazon, Brazil.Journal of Insect Conservation16: 733-744.

Spaniol, R.L.; Duarte, L.S.; Mendonça, M.S.; Iserhard, C.A. 2019. Combining functional traits and phylogeny to disentangling Amazonian butterfly assemblages on anthropogenic gradients. Ecosphere 10: e02837.

Teston, J.A.; Silva, P.L. 2017. Diversity and seasonality of frugivorous butterflies (Lepidoptera, Nymphalidae) in the Tapajós National Forest, Pará, Brazil. Biota Amazônia 7: 79-83.

Viloria, Á.L.; Costa, M. 2019. Butterflies. In: Rull, V.; Vegas-Vilarrúbia, T.; Huber, O.; Señaris, C. (Eds.). The Pristine “Lost World” of the Neotropical Guiana Highlands. Elsevier: London, UK, p. 193-222.

Edited by

• ASSOCIATE EDITOR:
  Ana Lúcia Tourinhohttps://orcid.org/0000-0001-8335-328X

Publication Dates

• Publication in this collection
  04 Aug 2025
• Date of issue
  2025

History

• Received
  24 May 2024
• Accepted
  30 Apr 2025