Anuran assemblage changes along small-scale phytophysiognomies in natural Brazilian grasslands

ABSTRACT

We studied the species composition of frogs in two phytophysiognomies within Pampa biome (grassland and forest) of a Ramsar site in southern Brazil to assess the distribution of species and dissimilarities in community composition on a small spatial scale. We tested the hypothesis that the vegetation structure and the types of land cover present in each physiognomy influence species distribution and the compositional dissimilarity patterns between locations. We sampled individuals using pitfall traps and active search in the areas around the traps. We evaluated the existence of these differences by using permutational multivariate analysis of variance and multivariate dispersion. We found that the compositional dissimilarity was higher between the sampling sites from different phytophysiognomies than within the same phytophysiognomy. Also, the difference in anuran species composition between the sampling sites within the forest was considerably high. These differences were mainly due to the type of land cover present in each sampled site. Based on our results, we could assume that the phytophysiognomies evaluated here offer quite different colonization opportunities for anurans, especially those related to microhabitat characteristics, such as microclimate variables. Still, the presence of different types of land cover seems to be a decisive factor for the maintenance of some anuran species, since these can serve as an important source for obtaining food resources, in addition to facilitating the dispersion of individuals within and between locals, serve as sites for the regulation of physiological functions and also refuges against predators.

KEYWORDS
Beta diversity; Dissimilarity; forest; south Brazil; Taim

The composition of species in communities is a result of a complex interaction between organisms and environmental characteristics. In general, the presence of a singular species in anassemblage depends on intrinsic (such as the individual's physiological needs for homeostatic maintenance) and extrinsic processes(such as competition, predation and other biotic interactions). Combined, both of them determine the possibility of a given species to colonize a habitat, as well theirlong-term maintenance in there (Schluter & Ricklefs, 1993; Ximenez et al., 2014). Abiotic factors, such as clime, the complexity of the vegetation structure, the type of soil, the primary productivity and the physical-chemical composition of the water determine the local characteristics of each community. In terms of community structure patterns, several studies indicate that these factors are mainly responsible for the patterns of dissimilarity in species composition between communities (e.g. Iop et al., 2012; Dalmolin et al., 2019; Ceron et al., 2020). Still, it is assumed that heterogeneous habitats comprise a richerbiota (Semlitsch et al., 2015) since they offer many different microhabitats, which preciselyfavors the establishment of species without high niche overlap (Barrows & Allen, 2010; Oliveira et al., 2013).

The process of describing and evaluating the habitat structure is one of the challenges for carrying out ecological studies (Stein & Kreft, 2015). The concept of habitat heterogeneity is related to much confusion, being frequently used as a synonym of habitat complexity. In structural terms, environments can be described by (i) their complexity - vertical development of vegetation - and (ii) their heterogeneity - horizontal variation in descriptors such as density, dominance and frequency of plant species that compose them (August, 1983). In a review on the topic, Tews et al. (2004) defend the idea that the concept of habitat heterogeneity includes both vertical and horizontal structure of plant community - a concept adopted in the present study. The possibility of assessing the effect of habitat heterogeneity on the structuring of their communities comes up against the huge number of areas never sampled in the Neotropical region (Garcia & Vinciprova, 2003). This fact also has implications for the establishment of conservation strategies for potentially threatened species - as is the case with many Brazilian anurans (Silvano & Segalla, 2005).

The variety of available habitats and the high levels of species endemism throughout the Brazilian biomes makes it an adequate location to assess the role of environmental heterogeneity in the activity patterns and community structure of anurans (Silvano & Segalla, 2005). However, most studies on anuran ecology in Brazil are concentrated in forest environments, especially in the Cerrado and the Atlantic Forest biomes (Da Silva et al., 2011a; Iop et al., 2012; Prado & Rossa-Feres, 2014; Saccol et al., 2017). The Pampa grassland formations and the coastal portion ofsouthern Brazil are home to anuran species that are common in open habitats (Núñez et al., 2004, Vasconcelos et al., 2019; Dalmolin et al., 2019). Although the landscape of these environments consists predominantly of open areas (Machado et al., 2012), some forest patches can also be observed (such as swamp forests and dry or sandy forests).

The Brazilian coastal plains are formed by sediments from the Quaternary (Marques et al., 2015). It is estimated that these sediments originated from climatic and geological events that occurred in the Upper Pleistocene and the Holocene (more than 120 thousand years ago), although the plains are older and formed even during the Quaternary (in around 2.5 million years; Pessenda et al., 2012). The vegetation associated with this type of formation is typical of the Edaphic Systems of First Occupation and the Wet Dense Forest of Várzea, whose establishment occurs on quaternary sediments (Marques et al., 2015). The climatic, geological and oceanographic characteristics of the coastal region differ along its length, so that the topography and composition of soils vary widely on small spatial scales (Silveira, 1964). Consequently, landscapes with unique characteristics are formed in each portion of the Brazilian coast, and this has a direct impact on the type of structure and composition of vegetation, forest formations and associated fauna (Marques et al., 2015).

The landscapes of the southern the Brazilian coast differ from the landscapes found in the countryside or mountainous areas of the southern states of Brazil. In general, the coastal plains of the southern Brazil is made up of wide sedimentary and sandy plains associated with lakes of different sizes and small sets of mountains in the Serra Geral (Silveira, 1964; Gonçalves & dos Santos, 1985). It is estimated that historical and contemporary processes have a direct influence on the climate of this region, which is more humid and hotter; still, the association of these characteristics with the changes in the geomorphology of the region promoted successive historical changes in the vegetation of this region (Bartholomeu et al., 2014; Marques et al., 2015). Similarly, these same characteristics are expected to have influenced the selection process of anuran species occurring in the coastal region, since only one third of the more than 100 anuran species known to the state of Rio Grande do Sul occurs in the coastal plains (Colombo et al., 2008; Dalmolin et al., 2019; Knauth et al., 2019).

Studies that describe changes in amphibian assemblages over structural changes in vegetation cover can contribute to propose scenarios of landscape changes (Da Silva et al., 2011b). The process of forests expanding over grasslands under climate change or even converting forested areas into human use would reflect changes in the amphibian composition (Manenti et al., 2013; Oliveira & Rocha, 2015). In this sense, our work aimed to: (i) verify the distribution of anuran species on a small spatial scale; (ii) describe the association between the patterns ofdissimilarity in anuran composition between communities of different phytophysiognomies within Pampa biome (forest and grassland) and their environmental descriptors (types of land cover; vegetation cover; vegetation height).

The study design on a small spatial scale is interesting to assess adjacent habitats whose probability of colonization is theoretically the same for all species in the region (Ximenez et al., 2014). In addition, the habitat elements linked to the micro-habitat scale are important descriptors of the spatial partition of species (Oliveira & Rocha, 2015) in particular because they deal with transition zones in a refined way (Hofer et al., 2007). The detection of differences in anuran assemblages between habitats provides a strong indication of the segregation process of these species along a vegetation gradient (Da Silva et al., 2011a; Prado & Rossa-Feres, 2014). We expected to find greater differences in composition in places belonging to different phytophysiognomies than between areas belonging to the same phytophysiognomies, since the environmental and microenvironmental conditions must be similar in these areas, thus reducing their composition dissimilarity (Da Silva et al., 2011a). In addition, we expect vegetation descriptors to be the main drivers of patterns of dissimilarity in the composition of anurans among the sampled sites, since the structure and complexity of vegetation is important for the most diverse aspects of anuran biology (e.g. physiological controls, sources of food resources and refuge; Dalmolin et al., 2019, 2020).

MATERIAL AND METHODS

Study site. The study was carried out at the Estação Ecológica do Taim (ESEC Taim; S 32°32ʼ16.6ˮ, W 052°32ʼ16.7ˮ), located in the municipality of Rio Grande, Brazil’s extreme south (Fig. 1). ESEC Taim encompasses ecosystems of great relevance and, since 2017,was recognized as a Ramsar site (Ramsar Wetland of International Importance; Ramsar, 2018). The landscape is located on the Pampa biome and consists of wetlands and grasslands in association with remnants of “restinga” and flooded forests, which are dominated by cork trees and fig trees (Waechter & Jarenkow, 1998).

The climate of the region is humid sub-temperate, with an annual average temperature of 18.1ºC and average temperature of the coldest month of 12.7°C. The annual rainfall is 1162 mm, with periods of drought in the spring and a higher incidence of rain in the winter (Maluf, 2000). The maximum temperature varied between 9.1° C and 39.3° C during the study period (May 2011 to April 2012), with the hottest months being February and March 2012; the minimum temperature varied between 0.9 ° C and 28.1 ° C, with the coldest months being July and August 2011. Meteorological data were obtained from the database provided by the Meteorological Station of the Universidade Federal do Rio Grande (FURG), located in the municipality of Rio Grande.

Fig. 1.
Geographic location of the study site (C) on the Estação Ecológica do Taim (B) in the state of Rio Grande do Sul (A), Brazil.

Anuran sampling. Sampling was carried out in four sampling sites representing the two most remarkable phytophysiognomies of ESEC Taim. The four sampling sites were distributed equally, two in areas of grassland dominance and two in areas of forest dominance (Fig. 2; ^{Supplementary Material} Fig. 1). We were able to distinguish the boundary between these two environments precisely even by a visual inspection. This was possible because the forests are distributed in patches merged along the grassland with well-defined boundaries.

Fig. 2.
Phytophysiognomies of ESEC Taim, Rio Grande do Sul, Brazil: (A) sites of grassland dominance and (B) sites of forest dominance.

Sampling areas were chosen based on vegetation integrity, area with continuous habitat and presence of potential anuran breeding sites. In addition, we considered our accessibility to these areas since many regions of ESEC Taim are permanently flooded, making the displacement of researchers and the installation of pitfall traps not possible. Sampling areas are at least about 800 m from each other. Field campaigns were carried out every two weeks between May 2011 and April 2012. Detailed information about sampling effort was described below. Each campaign comprised four days when all sites were simultaneously sampled. Frogs were sampled by the combination of three methods: (i) pitfall traps with drift fence; (ii) auditory surveys; (iii) visual search.

Pitfall traps with drift fence (passive sampling). We installed sets of buckets arranged in pitfall lines in all the areas of grassland dominance and forest dominance. Grasslands represent the dominant habitat and represents the habitat matrix on entire landscape. Sampled forest are formed by two continuous forest patches with total area of 27 and 22 ha respectively. We installed two lines of buckets in each both, grassland and forests, resulting in eight lines and 32 buckets (see the Fig. 2 of supplementary material). Pitfalls were arranged following the description of similar studies (e.g. Cechin & Martins, 2000). Each pitfall was formed by four 100-l buckets buried about 13.3 m from each other and connected by a mosquito net (60 cm in height) that served as a guide fence. The guide fence was buried 10 cm into the ground to avoid individuals to transpose the fence (Cechin & Martins, 2000). All buckets had holes at the bottom to prevent them from being flooded by rain.

We maximized the minimum distance of trap lines to reduce the spatial dependence of sampling. None trap line was closer than 0.8 km from other habitat type. Spatial distribution of lines was limited by the low availability of areas free of flooding that could remove the buckets during extensive periods of rain. Considering this the exact point where each pitfall line was installed was defined by a result of a combination between several mandatory criteria that do not permit a random distribution of traps. We grouped the records obtained from each pair of lines of the same sampling point and consider it to be an assemblage of species. Pitfalls remained open during four consecutive days and the process was repeated in intervals of 15 days in a total of 192 hours/open buckets by month. In whole study sampled achieve a total of 2304 hours of open buckets During the samplings, we carried out daily inspections in the buckets. We identified every captured animal, which was later released about 20 meters from the trap. The collection was carried out with the authorization number 27755-1 from SISBIO.

Visual search and auditory surveys (active sampling). Visual search and auditory surveys (Heyer et al., 1994) were performed during the night, starting one hour after sunset. Active sampling was conducted in the vicinity of the pitfall traps. For standardization of sampling effort, we conducted a visual search in an area of about 1ha around each pitfall line. Based on our previous experience in auditory surveys, we were able to perform a safe species identification of calling males in a maximum area of 5 ha. This was the covered area for calling surveys at each pitfall line. Active sampling was made along four sampling nights and repeated monthly. Sampling started each night in a different environment (grassland or forest), alternately to avoid the same physiognomy to be always sampled at the same time. We applied two hours per person of active sampling. We must highlight that during auditory surveys, we took care to exclude any register that comes from a grassland/forest interface. By this we avoid to the risks of mismatches in habitat classification of calling males.

Phytophysiognomy evaluation. To better characterize the phytophysiognomies of the sampling areas, the vegetation structure was measured through plots (following Tozetti & Martins, 2008). We characterized the vegetation structure in the vicinity of the areas where pitfall traps were installed (see the Fig. 3 of supplementary material). We delimited two plots of 3 m x 10 m, which were distributed perpendicularly in each of the pitfall lines, one of which was placed at the beginning and the other at the end of the line. The following variables of the vegetation were measured in each plot: (i) Types (categories) of land cover - exposed (without any type of cover); cover by grassy vegetation; presence of shrubs; presence of trees; (ii) Vegetation cover (percentual); (iii) Vegetation height (centimeters).

Statistical analysis. We evaluated differences in the anuran composition in the areas within and between habitat types by using permutational multivariate analysis of variance (PERMANOVA) and multivariate dispersion (BETADISPER). PERMANOVA is an analysis of variance that accounts for the dissimilarity between different data matrices. This analysis is analogous to MANOVA (multivariate analysis of variance), but is differentiated by the incorporation of permutation algorithms (McArdle & Anderson, 2001; Anderson, 2001). In addition, PERMANOVA is considered a robust alternative for analysis of variance using distance data, since it allows the inclusion of matrices with multiple response variables and predictor variables (which is the case in this study). The composition dissimilarity matrix between communities was obtained through the presence and absence data of each specie in each transect. In PERMANOVA, we used the composition dissimilarity matrix as the response variable, and the environmental descriptors of each sample unit were used as predictor variables. We used the Jaccard index as a dissimilarity measure and performed 9999 permutations.

We use multivariate dispersion (BETADISPER) to visualize the patterns of beta diversity in our dataset. This analysis is commonly incorporated into studies that assess beta diversity patterns, as its tests the homogeneity of the data distribution (in our case the presence and absence of species in the sample units) in a multivariate space (Anderson, 2006). For this, non-Euclidean distances between objects and group centroids are handled by reducing the original distances to principal coordinates. Finally, we used a Similarity Percentages analysis (SIMPER) we used an analysis to discriminate which species were most associated with each type of land cover. PERMANOVA, BETADISPER and SIMPER were run using the “adonis”, “betadisper” and “simper” functions in the R package vegan (Oksanen et al., 2017).

RESULTS

We recorded 13 species of six families in both phytophysiognomies (Fig. 3; Tab. I). Leptodactylidae and Hylidae were the most representative, with six and four species respectively. The hylid Boana pulchella (Duméril & Bibron, 1841) occurred exclusively in grasslands and Scinax squalirostris (Lutz, 1925) occurred exclusively in forests. The number of recorded species varied during the sampling period, with the largest number of species recorded in May (12 species), September and November (11 species) in both phytophysiognomies.

Fig. 3.
Anuran species found in two phytophysiognomy types of Estação Ecológica do Taim, southern Brazil, between May 2011 and April 2012. The pink, green and mixed circles (green and pink) indicate the species that were recorded, respectively, in grassland areas, forest areas, and in both areas. (a) Elachistocleis bicolor (Guérin-Méneville, 1838); (b) Boana pulchella (Duméril & Bibron, 1841); (c) Leptodactylus gracilis (Duméril & Bibron, 1840); (d) Leptodactylus latinasus (Jiménez de la Espada, 1875); (e) Leptodactylus luctator (Hudson, 1892); (f) Odontophrynus maisuma (Reinhardt & Lütken, 1862); (g) Physalaemus biligonigerus (Cope, 1861); (h) Physalaemus gracilis (Boulenger, 1883); (i) Pseudis minuta (Günther, 1858); (j) Pseudopaludicola falcipes (Hensel, 1867); (k) Rhinella dorbignyi (Duméril & Bibron, 1841); (l) Scinax granulatus (Peters, 1871); (m) Scinax squalirostris (Lutz, 1925).

Tab. I.
Anuran species registered by the three sampling methods in two phytophysiognomy types at Estação Ecológica do Taim, southern Brazil, between May 2011 and April 2012. Black squares represent the occurrence of species in the relative month. Sampling methods: A (Auditory); P (Pitfall); V (Visual).

Family	Species	Sampling Method	Months with registers
			M	J	J	A	S	O	N	D	J	F	M	A
Bufonidae	Rhinella dorbignyi	A; P; V
Cycloramphidae	Odontophrynus maisuma	A; P; V
Hylidae	Boana pulchella	A; P; V
	Pseudis minuta	A; P; V
	Scinax granulatus	A; P; V
	Scinax squalirostris	A; P; V
Leptodactylidae	Physalaemus biligonigerus	P
	Physalaemus gracilis	A; P; V
	Pseudopaludicola falcipes	A; P
	Leptodactylus gracilis	A; P
	Leptodactylus luctator	A; P; V
	Leptodactylus latinasus	A; P; V
Microhylidae	Elachistocleis bicolor	P; V

We found that the patterns of compositional dissimilarity (β diversity) were higher between the sampling sites from different phytophysiognomies than within the same phytophysiognomy (F = 4.72; p=0.05; Tab. II; Fig. 4). Also, the difference in composition between the sampling sites in forest areas was considerably high. Through PERMANOVA, we identified that the types of land cover were the main environmental descriptor associated with the patterns of compositional dissimilarity between the sampled units evaluated (Tab. III). Of the total fraction of explained variation, 56% were explained by the differences in land cover between the sampling units (F = 2.23; p=0.04). Physalaemus biligonigerus, Rhinella fernandezae, Boana pulchella and Leptodactylus latinasus were the species that most contributed to the patterns of dissimilarity between the types of soil cover (Tab. IV). Physalaemus biligonigerus and Rhinella fernandezae were the species that most contributed to the patterns of compositional dissimilarity in completely exposed soils (without any type of vegetation), while Boana pulchella and Leptodactylus latinasus were the species that most contributed to the dissimilarity in areas with soils covered by some type of vegetation.

Tab. II.
Patterns of compositional dissimilarity between forest (green cells) and grassland (pink cells), Estação Ecológica do Taim, southern Brazil, between May 2011 and April 2012. Bold values represent the greatest values of dissimilarity between two points.

	CAA	CAB	CBA	CBB	MAA	MAB	MBA	MBB
CAA	-
CAB	0.125	-
CBA	0	0.125	-
CBB	0	0.125	0	-
MAA	0.375	0.500	0.375	0.375	-
MAB	0.250	0.375	0.250	0.250	0.429	-
MBA	0.125	0.250	0.125	0.125	0.286	0.143	-
MBB	0.125	0.250	0.125	0.125	0.286	0.143	0	-

Tab. III.
Effects environmental descriptors on patterns of compositional dissimilarity between phytophysiognomies of Estação Ecológica do Taim, southern Brazil accounted by PerMANOVA. Bold values represent statistically significant effect values.

Environmental descriptors	Sum of squares	R2	F	p
Types of land cover	0.27	0.56	2.23	0.04
Vegetation cover	0.06	0.13	1.61	0.24
Vegetation height	0.07	0.14	1.66	0.23

Fig. 4.
Dissimilarity in anuran composition in two different phytophysiognomies of Reserva Ecológica do Taim, southern Brazil. The black circles and red triangles represent, respectively, the points sampled in forest and grassland areas.

Tab. IV.
Cumulative contributions of most influential species in each type of land cover at Estação Ecológica do Taim, southern Brazil accounted by SIMPER.

Types of land cover	Species	Cumulative contributions
Exposed	Physalaemus biligonigerus	0.21
	Rhinella fernandezae	0.42
	Scinax granulatus	0.62
	Pseudopaludicola falcipes	0.81
Cover by grassy vegetation	Boana pulchella	0.20
	Leptodactylus latinasus	0.39
	Pseudis minuta	0.59
	Physalaemus biligonigerus	0.70
Presence of shrubs	Boana pulchella	0.25
	Physalaemus biligonigerus	0.38
	Elachistocleis bicolor	0.51
	Pseudopaludicola falcipes	0.63
	Rhinella fernandezae	0.75
Presence of trees	Leptodactylus latinasus	0.22
	Pseudis minuta	0.43
	Rhinella fernandezae	0.58
	Elachistocleis bicolor	0.73

DISCUSSION

Our results showed differences in species composition between grassland and forest habitats. They pointed out that differences in the types of land cover are capable to drive significant changes in anuran species composition, which confirms our initial hypothesis. We also highlight that we surveyed adjacent (spatially closely related) sites, which reveals that vegetation has a powerful effect on defining the composition of anuran assemblages. We may assume that the phytophysiognomies evaluated here offer quite different colonization opportunities to anurans, with drastic changes in anuran assemblages across a small geographical area. An obvious relationship between phytophysiognomies and colonization opportunities is related to microhabitat characteristics generated by shading, wind reduction, humidity maintenance, and reduction of dial range of air temperature. Microenvironmental conditions (especially those related to the microclimate) vary substantially between forests and grassland (D’Odorico et al., 2013) and, in our sampling site, mainly because the transition between them is relatively abrupt. Based on these differences, the heterogeneity affects a set of microhabitat proprieties, which could enable the maintenance of some species. The effects of substrate characteristics (e.g., litter deep, moisture retention), and dial variation in air temperature are well-known as determining factors for structuring anuran assemblages (Da Silva et al., 2011b; Van Dyke et al., 2017). Based on this, we point out that the relatively great difference in anuran species composition between sampling sites in forest is a result of the greater heterogeneity of this habitat in relation to grassland. It makes sense since forest sites would vary in both vertical and horizontal components of vegetation cover whereas grasslands vary mostly across the horizontal component (Dalmolin et al., 2019, 2020).

Differences in land cover were recognized as the main drivers of patterns of compositional dissimilarity in communities of different taxa, including birds (Lee & Martin, 2017), amphibians (Iop et al., 2020; Moreira et al., 2020) and reptiles (Wanger et al., 2010). In general, the presence of soils with some type of vegetation cover is an essential condition for the maintenance of species, especially in the southern region of Brazil, where the conversion of natural habitats to human activities is intense (Iop et al., 2020; Moreira et al., 2020). More precisely, habitats composed of different types of vegetation are structurally more complex and, a priori, add a greater set of ecological niches that can be occupied by species with different ecological requirements (Valério et al., 2016; Dalmolin et al., 2020). Thus, it is expected that habitats with greater complexity in the vertical and horizontal structure of the vegetation will be able to provide more diversified resources for the species, which leads to a considerable increase in the local diversity of species (Tews et al. 2004; Valério et al., 2016). For anuran communities, these findings are totally plausible, since heterogeneous habitats are essential to shelter the great diversity of reproductive modes and foraging strategies presented by this group (Haddad & Prado, 2005). For example, males of the species of the Hylidae family use vegetation as perches to vocalize; species of the Leptodactylidae family, on the other hand, vocalize on the soil and use vegetation as a substrate or as a way to camouflage themselves, thus, reduce the risk of predation (Santos et al., 2008; Dalmolin et al., 2020).

Land cover also influences other important aspects of amphibian biology. Soils provided with vegetation cover should facilitate the process of dispersion of frogs, since they offer both food and local resources that can be used as refuges. For communities present in environments subject to flooding (which is the case of Estação Ecológica do Taim), these resources provided by soils covered by vegetation seem to determine the local species richness, as was observed for the Brazilian Pantanal (Moreira et al., 2017). In addition, soils covered by vegetation have a greater capacity to retain humidity, which is essential for individuals to maximize their thermoregulation and internal water balance processes (Rittenhouse et al., 2009; Nowakowski et al., 2018; Morley et al., 2019; Rubalcaba et al., 2019). For species that have limited dispersion capacities and move preferentially through the terrestrial habitat (which is the case of Odontophrynus maisuma and Rhinella dorbignyi), the presence of soils covered by vegetation can be a decisive factor.

In general, it is expected that forest environments will have a greater variety of microhabitats (Afonso & Eterovick, 2007; Prado & Rossa-Feres, 2014; Van Dyke et al., 2017). As a result, the occurrence of a greater number of species with different functional traits and evolutionary histories is enhanced (Dalmolin et al., 2019). This possibility opens a precedent for new studies with the specific objective of evaluating the way species use these habitats, as well as their preferences in terms of microhabitats.

We must highlight the fact that Scinax squalirostris was recorded only in forests. This result is different from what was expected since this species is associated with low vegetation such as grass and shrubs (Ximenez et al., 2014; Maneyro et al., 2017; Dias et al., 2019). Another aspect to be taken into account is that hylids are less likely to be caught in pitfall traps because of their climbing habits (Oliveira et al., 2013; Dosso et al., 2019). This could have caused some sampling bias and, consequently, affected the observed patterns (which was not our case, since we used different, complementary sampling methods). In addition, some studies indicate that even specific sampling methods such as calling surveys are prone to the observer’s effect on the pattern of results (Dosso et al., 2019).

The fact that we have associated different sampling methods favored the record of species that make wide use of each habitat. Estimations based exclusively on auditory surveys, for example, prioritize the record of habitat use related to reproductive activity (Santos et al., 2016; Moser et al., 2019) limiting the projections regarding habitat requirements (Oliveira et al., 2016; Santos et al., 2016). This is not the case in our study, as previous work carried out in the same region reported the occurrence of nine to fifteen species (Ximenez et al., 2014; Ximenez & Tozetti, 2014; Dalmolin et al., 2019, 2020). Thus, our samplings can be considered effective for the representativeness of anuran species in Brazil’s extreme south.

The richness and diversity of anurans recorded in our study is relatively less than those recorded in other studies conducted on the coast of Rio Grande do Sul (e.g. Colombo et al., 2008; Knauth et al., 2019). In general terms, these differentiations may be the result of the strength of species selection based on different historical and contemporary processes. The southern end of the Brazilian coast is surrounded by mountain ranges, such as the Escudo Sul-Rio-Grandense, which is located in the southeastern region of Rio Grande do Sul (Gonçalves & dos Santos, 1985). It is assumed that the mountainous areas of the Atlantic Forest and Campos Sulinos biomes have contributed to the speciation processes and in the modulation of the patterns of richness and diversity of anurans in the region (Vasconcelos et al., 2019); this, therefore, these mountainous areas serve as barriers for the dispersion of species and individuals, contributing to the processes of allopatric, parapatric and sympatric speciation (Vences & Wake, 2007; Antonelli et al., 2009; Field et al., 2009), as well as for the evolution of the considerable diversity of reproductive modes in anurans (Haddad & Prado, 2005).

The mountainous areas of Rio Grande do Sul also directly influence the climate of the most diverse regions of the state, as well as the primary productivity of the associated ecosystems (Hawkins et al., 2003). These set of factors seems to be more pronounced for anurans in the extreme south of the Brazilian coast, since recent analyzes of the patterns of richness and distribution of subtropical species are in line with the expected pattern for the hypothesis of climatic variability (Stevens, 1989; Vasconcelos et al., 2019). According to this hypothesis, environments with a wide range of climatic variation select from the regional pool only those species that throughout their evolutionary history have developed physiological, morphological and behavioral characteristics that give them wide ranges of physiological tolerances to support such variations (Stevens, 1989; Vasconcelos et al., 2019). Consequently, these regions will be composed of species with broader ecological niches and, at the same time, with greater ranges of geographic distribution (Stevens, 1989; Vasconcelos et al., 2019). This is precisely the case for most representatives of the families Hylidae, Leptodactylidae and Microhylidae registered in our study areas. Finally, temperature is also considered the limiting factor for the distribution and occurrence of anurans in the southern portion of the Brazilian coast (Vasconcelos et al., 2019). It is worth mentioning that these regions suffer from constant anthropogenic changes, which can have direct effects on the microclimate conditions of the environments, so that the modification of the habitats and vegetation associated with them can cause the loss of species and ecological functions which are important for the maintenance of coastal ecosystems (Dalmolin et al., 2020).

The results herein obtained draw attention to the role of heterogeneity and environmental complexity on anuran composition. They also warn about the imminent risk of changing landscapes due to the invasion of forest species such as Eucalyptus and Pinus species. Although the sampled communities do not include any endangered species, it is worth mentioning that the study area is a Ramsar site, which is of particular concern when dealing with an area of enormous ecological relevance and importance for the maintenance of many ecosystem functions.

Acknowledgments.

This study had the collaboration of researchers linked to the Laboratory of Ecology of Terrestrial Vertebrates of FURG. Among them, we are especially grateful to Tatiane Penteado Gonçalves, Francis de Mattos Almeida. We also thank the ICMBio staff at the Estação Ecológica do Taim for their support in the field. This work was financially supported by the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) and Fundação de Amparo à Pesquisa do Estado do Rio Grande do Sul (Fapergs).

REFERENCES

• Afonso, L. G. & Eterovick, P. C. 2007. Spatial and temporal distribution of breeding anurans in streams in southeastern Brazil. Journal of Natural History 41:949-963.
• Anderson, M. J. 2001. A new method for non-parametric multivariate analysis of variance. Austral Ecology 26:32-46.
• Anderson, M. J. 2006. Distance-based tests for homogeneity of multivariate dispersions. Biometrics 62:245-253.
• Antonelli, A.; Nylander, J. A. A.; Persson, C. & Sanmartín, I. 2009. Tracing the impact of the Andean uplift on Neotropical plant evolution. Proceedings of the National Academy of Sciences 106(24):9749-9754.
• August, P. V. 1983. The role of habitat complexity and heterogeneity in structuring tropical mammal communities. Ecology 64:1495-1507.
• Barrows, C. W. & Allen, M. F. 2010. Patterns of occurrence of reptiles across a sand dune landscape. Journal of Arid Environments 74:186-192.
• Bartholomeu, R. L.; Barros, M. A.; Lopes, M. R. S.; Vilela, C. G. & Barth, O. M. 2014. Evolução Paleogeográfica da Planície Costeira da Praia Vermelha, Entrada da Baía de Guanabara, Rio de Janeiro, por meio de Registros Palinológicos. Anuário do Instituto de Geociências 37:92-103.
• Cechin, S. Z. & Martins, M. 2000. Eficiência de armadilhas de queda (pitfall traps) em amostragens de anfíbios e répteis no Brasil. Revista Brasileira de Zoologia 17(3):729-749.
• Ceron, K.; Santana, D. J.; Lucas, E. M.; Zocche, J. J. & Provete, D. B. 2020. Climatic variables influence the temporal dynamics of an anuran metacommunity in a nonstationary way. Ecology and Evolution 10:4630-4639.
• Colombo, P.; Kindel, A.; Vinciprova, G. & Krause, L. 2008. Composição e ameaças à conservação dos anfíbios anuros do Parque Estadual de Itapeva, município de Torres, Rio Grande do Sul, Brasil. Biota Neotropica 8:229-240
• Da Silva F. R.; Gibbs, J. P. & Rossa-Feres, D. C. 2011a. Breeding habitat and landscape correlates of frog diversity and abundance in a tropical agricultural landscape. Wetlands 31:1079-1087.
• Da Silva, F. R.; Prado, V. H. M. & Rossa-Feres, D. C. 2011b. Value of Small Forest Fragments to Amphibians. Science 332:1033-1033
• Dalmolin, D. A.; Tozetti, A. M. & Pereira, M. J. R. 2020. Turnover or intraspecific trait variation: explaining functional variability in a neotropical anuran metacommunity. Aquatic Sciences 82:62.
• Dalmolin, D. A.; Tozetti, A. M. & Ramos Pereira, M. J.. 2019. Taxonomic and functional anuran beta diversity of a subtropical metacommunity respond diferentially to environmental and spatial predictors. PLoS ONE 14(11):e0214902.
• Dias, K. S.; Dosso, E. S.; Hall, A. S.; Schuch, A. P. & Tozetti, A. M. 2019. Ecological light pollution affects anuran calling season, daily calling period, and sensitivity to light in natural Brazilian wetlands. Science of Nature 106.
• DʼOdorico, P.; He, Y.; Collins, S.; De Wekker, S. F. J.; Engel, V. & Fuentes, J. D. 2013. Vegetation-microclimate feedbacks. Global Ecology and Biogeography 22:364-379.
• Dosso, E. S.; Araujo, A.; Dias, K. S.; Martins, K. P.; Silveira, S. H. & Tozetti, A. M. 2019. The effect of observer during calling surveys on breeding sites: potential bias on estimating the abundance of calling males of of Pseudis minuta Günther, 1858 (Anura, Hylidae). Herpetology Notes 12:991-993.
• Field, R.; Hawkins, B. A.; Cornell, H. V.; Currie, D. J.; Diniz‐Filho, J. A. F.; Guégan, J‐F.; Kaufman, D. M.; Kerr, J. T.; Mittelbach, G. G.; Oberdorff, T.; O’Brien, E. M. & Turner, J. R. G. 2009. Spatial species‐richness gradients across scales: a meta‐analysis. Journal of Biogeography 36:132-147.
• Garcia, P. C. & Vinciprova, G. 2003. Anfíbios. In: Fontana, C. S.; Bencke, G. A. & Reis, R. eds. Livro Vermelho da Fauna Ameaçada de Extinção no Rio Grande do Sul. Porto Alegre, EDIPUCRS, p.147-164.
• Gonçalves, J. M. S. & dos Santos, N. M. 1985. Análise das classificações do relevo para o Rio Grande do Sul. Boletim Gaúcho de Geografia 13:3-20.
• Haddad, C. F. B. & Prado, C. P. A. 2005. Reproductive modes of the Atlantic forest frogs. BioScience 55(3):208-217.
• Hawkins, B. A.; Field, R.; Cornell, H. V.; Currie, D. J.; Guegan, J. F.; Kaufman, D. M.; Kerr, J. T.; Mittelbach, G. G.; Oberdorff, T.; OʼBrien, E. M.; Porter, E. E. & Turner, J. R. G. 2003. Energy, water, and broad‐scale geographic patterns of species richness. Ecology 84:3105-3117.
• Heyer, W. R.; Donnelly, M. A.; Mcdiarmid, R. W.; Hayek, L. C. & Foster, M. S. 1994. Measuring and monitoring biological diversity: standard methods for amphibians. Smithisonian Institution Press. 697p.
• Hofer, U.; Bersier, L. I. X. & Borcard, D. 2007. Ecotones and gradient as determinants of herpetofaunal community structure in the primary forest of Mount Kupe, Cameroon. Journal of Tropical Ecology (2000):517-533.
• Iop, S.; Caldart, V. M.; Santos, T. G. & Cechin, S. 2012. What is the Role of Heterogeneity and Spatial Autocorrelation of Ponds in the Organization of Frog Communities in Southern Brazil? Zoological Studies 51(7):1094-1104
• Iop, S.; dos Santos, T. G.; Cechin, Z. S.; Vélez‐Martin, E.; Pillar, V. D. & Prado, P. I. 2020. The interplay between local and landscape scales on the density of pond‐dwelling anurans in subtropical grasslands. Biotropica 52:913-926.
• Knauth, D. S.; Pires, M. M.; Stenert, C. & Maltchik, L. 2019. Disentangling the role of niche-based and spatial processes on anuran beta diversity in temporary ponds along a forest-grassland transition. Aquatic Sciences 81:1-13.
• Lee, M-B. & Martin, J. A. 2017. Avian Species and Functional Diversity in Agricultural Landscapes: Does Landscape Heterogeneity Matter? PLoS ONE 12(1):e0170540.
• Machado, I. F.; Moreira, L. F. B. & Maltchik, L. 2012. Efects of pine invasion on anurans assemblage in southern Brazil coastal ponds. Amphibia-Reptilia 33:227-237.
• Maluf, J. R. T. 2000. A new climatic classification for the State of Rio Grande do Sul, Brazil. Revista Brasileira de Agrometeorologia 8(1):141-150.
• Manenti, R.; Denoël, M. & Ficetola, G. F. 2013. Foraging plasticity favours adaptation to new habitats in fire salamanders. Animal Behaviour 86:375-382.
• Maneyro, R.; Loebmann, D.; Tozetti, A. M. & Fonte, L. F. M. 2017. Anfíbios das planícies costeiras do extremo sul do Brasil e Uruguai. Curitiba, Anolis Books. 176p.
• Marques, C. M.; Menezes, S. & Liebsch, D. 2015. Coastal plain forests in southern and southeastern Brazil: ecological drivers , floristic patterns and conservation status. Brazilian Journal of Botany 38:1-18.
• McArdle, B. H. & Anderson, M. J. 2001. Fitting multivariate models to community data: A comment on distance-based redundancy analysis. Ecology 82:290-297.
• Moreira L. F. B.; Dorado-Rodrigues, T. F.; Ferreira, V. L. & Strüssmann, C. 2017. An inter-dependence of flood and drought: disentangling amphibian beta diversity in seasonal floodplains. Marine and Freshwater Research 68:2115-2122.
• Moreira, L. F. B.; de Castilhos, H. Z. & Castroviejo‐Fisher, S. 2020. Something is not quite right: Effects of two land uses on anuran diversity in subtropical grasslands. Biotropica 52:1286-1297.
• Morley, S. A.; Peck, L. S.; Sunday, J. M.; Heiser, S. & Bates, A. E. 2019. Physiological acclimation and persistence of ectothermic species under extreme heat events. Global Ecology Biogeography 28:1018-1037.
• Moser, C. F.; Olmedo, G. M.; de Oliveira, M. & Tozetti, A. M. 2019. Movement ecology and habitat use in males of two species of Boana (Anura: Hylidae) during the breeding season. Herpetology Notes 12:885-893.
• Nowakowski, A. J.; Watling, J. I.; Thompson, M. E.; Brusch, G. A.; Catenazzi, A.; Whitfield, S. M.; Kurz, D. J.; Suárez‐Mayorga, A.; Aponte‐Gutiérrez, A.; Donnelly, M. A. & Todd, B. D. 2018. Thermal biology mediates responses of amphibians and reptiles to habitat modification. Ecology Letters 21:345-355.
• Núñez, D.; Maneyro, R.; Langone, J. & de Sá, R. O. 2004. Distribucion Geografica de la Faun de Anfibios del Uruguay. Smithsonian Herpetological Information Service 134:1-34 \rCopy in file-title page, 2 pages figures.
• Oksanen, J.; Kindt, R.; Legendre, P.; OʼHara, R. B. & Stevens, M. H. H. 2017. Vegan: Community ecology package. R package version 2.4-2. Electronic database Retrieved from http://CRAN.R-project.org/package=vegan.
  » http://CRAN.R-project.org/package=vegan.
• Oliveira, J. C. F. & Rocha, C. F. D. 2015. Journal of coastal conservation: a review on the anurofauna of Brazil’s sandy coastal plains. How much do we know about it? Journal of Coastal Conservation 19:35-49.
• Oliveira, M. C. L. M.; dos Santos, M. B.; Loebmann, D.; Hartman, A. & Tozetti, A. M. 2013. Diversity and associations between coastal habitats and anurans in southernmost Brazil. Anais da Academia Brasileira de Ciencias 85:575-583.
• Oliveira, M.; Aver, G. F.; Moreira, L. F. B.; Colombo, P. & Tozetti, A. M. 2016. Daily Movement and Microhabitat Use by the Blacksmith Treefrog Hypsiboas faber (Anura: Hylidae) during the Breeding Season in a Subtemperate Forest of Southern Brazil. South American Journal of Herpetology 11:89-97.
• Pessenda, L. C. R.; Vidotto, E.; Oliveira, P. E.; Buso Jr., A. A.; Cohen, M. C. L.; Rossetti, D. F. , Branco, F. R. & Bendassolli, J. A. 2012. Late quaternary vegetation and coastal environmental changes at Ilha do Cardoso mangrove, southeastern Brazil. Palaeogeography, Palaeoclimatology, Palaeoecology 363/364:57-68.
• Prado, V. H. M. & Rossa-Feres D. D. C. 2014. Multiple determinants of anuran richness and occurrence in an agricultural region in South-eastern Brazil. Environ Manage 53:823-837.
• RAMSAR: Wetland of International Importance. 2018. Avaiable at < Avaiable at https://www.ramsar.org />. Acessed 15 March 2018.
  » https://www.ramsar.org
• Rittenhouse, T. A. G.; Semlitsch, R. D. & Thompson, F. R. III. 2009. Survival costs associated with wood frog breeding migrations: effects of timber harvest and drought. Ecology 90:1620-1630.
• Rubalcaba J. G.; Gouveia, S. F. & Olalla‐Tárraga, M. A. 2019. A mechanistic model to scale up biophysical processes into geographical size gradients in ectotherms. Global Ecolology and Biogeography 28:793-803.
• Saccol, S. D. S. A.; Bolzan, A. M. R. & dos Santos, T. G. 2017. In the shadow of trees: does eucalyptus aforestation reduce herpetofaunal diversity in Southern Brazil? South American Journal of Herpetology 12:42-56.
• Santos, N. L. P. da S. dos; Colombo, P.; Avila, F. R. de; Oliveira, M. de & Tozetti, A. M. 2016. Calling Site Selection by the South American Tree-Frog Hypsiboas pulchellus (Anura, Hylidae) in Subtropical Wetlands. South American Journal of Herpetology 11:149-156.
• Santos, T. G.; Kopp, K.; Spies, M. R.; Trevisan, R. & Cechin, S. Z. 2008. Distribuição temporal e espacial de anuros em área de Pampa, Santa Maria, RS. Iheringia, Série Zoologia 98(2):244-253.
• Schluter, D. & Ricklefs, R. E. 1993. Species diversity: an introduction to the problem. In: Ricklefs, R. E. & Schluter, D. eds. Species diversity in ecological communities: historical and geographical perspectives. Chicago, University of Chicago Press. 414p.
• Semlitsch, R. D.; Peterman, W. E.; Anderson, T. L.; Drake, D. L. & Ousterhout, B. H. 2015. Intermediate Pond Sizes Contain the Highest Density, Richness, and Diversity of Pond-Breeding Amphibians. PLoS ONE 10(4):e0123055.
• Silvano, D. L. & Segalla, M. V. 2005. Conservação de Anfíbios no Brasil. Megadiversidade 1:79-86.
• Silveira, J. D. 1964. Morfologia do litoral. In: Azevedo, A. ed. Brasil: a terra e o homem. São Paulo, Editora Nacional. v. 1, p. 253-305.
• Stein, A. & Kreft, H. 2015. Terminology and quantification of environmental heterogeneity in species‐richness research. Biological Review 90:815-836.
• Stevens G. C. 1989. The latitudinal gradient in geographical range: how so many species coexist in the tropics. The American Naturalist 133(2):240-256
• Tews, J.; Brose, U.; Grimm, V.; Tielboörger, K.; Wichmann, M. C.; Schwager, M. & Jeltsch, F. 2004. Animal species diversity driven by habitat heterogeneity/diversity: the importance of keystone structure. Journal of Biogeography 31:79-92.
• Tozetti, A. M. & Martins, M. 2008. Habitat use by the South-American rattlesnake (Crotalus durissus) in south-eastern Brazil. Journal of Natural History 42(19-20):1435-1444.
• Valério, L. M.; Dorado-Rodrigues, T. F.; Chupel, T. F.; Penha, J. & Strüssmann, C. 2016. Vegetation Structure and Hydroperiod Affect Anuran Composition in a Large Neotropical Wetland. Herpetologica 72(3):181-188.
• Van Dyke, F.; Berthel, A.; Harju, S. M.; Lamb, R. L.; Thompson, D.; Ryan, J.; Pyne, E. & Dreyer, G. 2017. Amphibians in forest pools: Does habitat clustering affect community diversity and dynamics? Ecosphere 8(2):e01671.
• Vasconcelos, T. S.; Silva, F. R. da; dos Santos, T. G.; Prado, V. H. M. & Provete, D. B. 2019. Biogeographic Patterns of South American Anurans. Cham, Springer International Publishing. 149p.
• Vences, M. & Wake, D. B. 2007. Speciation, Species Boundaries and Phylogeography of Amphibians. In: Heatwole, H. H. & Tyler, M. eds. Amphibian Biology. Chipping Norton, Surrey Beatty & Sons, p. 2613-2669.
• Waechter, J. L. & Jarenkow, J. A. 1998. Composição e estrutura do componente arbóreo nas matas turfosas do Taim, Rio Grande do Sul. Biotemas 11(1):45-69.
• Wanger, T.; Iskandar, D.; Motzke, I.; Brook, B; Sodhi, N.; Clough, Y. & Tscharntke, T. 2010. Effects of Land‐Use Change on Community Composition of Tropical Amphibians and Reptiles in Sulawesi, Indonesia. Conservation Biology 24(3):795-802.
• Ximenez, S. S. & Tozetti, A. M. 2014. Seasonality in anuran activity and callingseason in a Brazilian subtemperate wetland. Zoological Studies 54:47.
• Ximenez, S. S.; Oliveira, M.; Santos, M. B. & Tozetti, A. M. 2014. The Influence of Habitat Configuration on Anuran Species Composition in Subtemperate Wetlands of Southernmost Brazil. South American Journal of Herpetology 9(3):235-243.

Supplementary material

The following online material is available for this article:

Fig. 1. Spatial distribution of traps in habitats. Each asterisk represents the position of a line of traps (set of four buckets). Red asterisks are traps in forest areas and yellow ones in grassland areas. The grasslandare the matrix of the habitat, while the forests are distributed in patches. The two samples sampled are delimited by the red line, being spot A (27 hectares) and spot B (22 hectares). On the right, images with details of the landscape and the line of traps indicated by the arrows. Fig. 2. Detail of the inside of a bucket, showing the Styrofoam plate used as protection for the captured animals. The guide fence was buried 10 cm into the ground to avoid individuals to transpose the fence. All buckets had holes at the bottom to prevent them from being flooded by rain.

Data availability

Data citations

Oksanen, J.; Kindt, R.; Legendre, P.; OʼHara, R. B. & Stevens, M. H. H. 2017. Vegan: Community ecology package. R package version 2.4-2. Electronic database Retrieved from http://CRAN.R-project.org/package=vegan.

Publication Dates

• Publication in this collection
  23 Aug 2021
• Date of issue
  2021

History

• Received
  19 Oct 2020
• Accepted
  05 May 2021