Anurans of a remnant forest in Urussanga , Santa Catarina , southern of Brazil

We provide a list of amphibians from a Submontane Forest remnant in Urussanga municipality, Santa Catarina, southern of Brazil. We sampling through visual and acoustic surveys in four distinct sites. Twenty-six species of anurans belonging to nine families were recorded, where nine species were rare, nine abundant, and eight common. Eleven reproductive modes were observed, with exotrophic tadpoles in still water (mode 1) being most frequent mode (38%). The period of highest vocalization activity occurred between October and January. Temperature, rainfall and air humidity were not correlated with the richness and abundance of calling males. Only photoperiod was positively correlated with the abundance of calling males and species richness.


INTRODUCTION
Amphibian biodiversity has become increasingly threatened worldwide (Convention on Biological Diversity 2010).In Brazil, major threats to conservation are related to habitat destruction as a result of deforestation, expansion of agriculture, mining, burning, and development of urban infrastructure (Silvano & Segalla 2005).
Unlike its flora, the fauna of Santa Catarina is not well known (Cherem et al. 2004), despite the state's high biodiversity.In contrast, anuran composition in the states of Paraná and Rio Grande do Sul in southern Brazil is wellcharacterized, because a large number of studies have been conducted for several decades (Armstrong & Conte 2010; Bernarde & Anjos 1999;Both et al. 2008;Colombo et al. 2008;Conte & Machado 2005;Conte & Rossa-Feres 2007;Machado & Maltchik 2007;Santos et al. 2008;Zank et al. 2013).In Santa Catarina, little is known about species richness and geographical distribution of frogs due to the paucity of studies related to the subject (Garcia et al. 2008;Lucas 2008).Most previous research has been concentrated in the western part of the state (Bastiani et al. 2012;Bastiani & Lucas 2013;Lingnau 2009;Lucas & Fortes 2008;Lucas & Garcia 2011;Lucas & Marocco 2011) and near the central coast (Garcia 1996;Garcia & Vinciprova 1998;Garcia et al. 2003;Kwet 2006).A new species was recently described, as well, from the state (Bruschi et al. 2014).
Information regarding the temporal and spatial distribution of organisms is essential for understanding their ecological relationships with the environment (Ferreira et al. 2012) and species conservation (Eterovick et al. 2005).Information related to the abundance and species richness of amphibians in Santa Catarina is rare, but is of high important to rebuild the life history of these organisms and understand population fluctuations and the general trend of declines in amphibian populations (Lucas 2008).The objectives of the present study were: 1) to characterize species richness, constancy, and reproductive modes of anuran species in an area of the Atlantic Forest in southern Santa Catarina; 2) to verify the vocalization seasons of anuran males in four water bodies located in the study area; 3) to verify correlations among climatic predictors (temperature, precipitation and photoperiod) and species richness and male vocalization activity.

Study site
The study was conducted on the premises of the Vale dos Figos Hotel, in the municipality of Urussanga, in southern Santa Catarina (28°28ʹ53.19ʺS, 049°18ʹ38.07ʺW, 195-222 m above sea level) (Figure 1).The vegetation comprises (permit SISBIO 29935-1).The individuals were manually captured, anesthetized, and sacrificed with a 5% lidocaine paste, then fixed in 10% formalin and preserved in 70% alcohol.The nomenclature adopted for this study followed Frost (2015).

Data analysis
Sampling efficiency was evaluated by a species accumulation curve, constructed from the matrix of monthly data collected on species presence/absence.Calculations were performed based on 1,000 randomizations using the nonparametric estimators, Bootstrap and Jacknife 1, based on presence-absence (incidence) of species, considering or not rare species, with the aid of the EstimateS 9.0 program (Cowell 2013).Each recorded species was classified according to the constancy of occurrence index (Dajoz 1983) and considered to be abundant (present in >50% of samples), common (present in 25%-50% of samples), or rare (present in <25% of samples).
The abundance of each species was determined based on the highest number of vocalizing males recorded during one of three passes through each site on a given sampling date, as suggested by Gottsberger & Gruber (2004).This procedure was adopted to avoid underestimating population abundance resulting from the use of average abundance between successive samplings (Conte & Rossa-Feres 2007).The diversity and evenness for each sampling site were evaluated by Shannon-Wiener's diversity index and Pielou's evenness index, respectively (Krebs 1999).Significant differences observed in specific diversity and evenness between the four sampling sites were evaluated using a Chi-square test with p ≤ 0.05 (Zar 1999).
The influence of environmental variables (temperature, rainfall, relative humidity, and photoperiod) on species remnants of secondary submontane tropical rainforest (IBGE 2012) at different degrees of conservation and interspersed with exotic species and orchards.The climate, according to the Köeppen classification system (1948), is Cfa with average temperature ranging from 17.0-19.3°Cand minimum and maximum averages ranging from 12.0-15.1°Cand from 23.4-25.9°C,respectively.Annual precipitation ranges from 1,200-1,700 mm, with rain events distributed throughout the year (EPAGRI 2001).The area is drained by a tributary of the Urussanga River and is situated in the geomorphologic region of the coal outcrop.

Data collection
Field sampling was conducted monthly (n = 1 d/mo) from November 2010 to December 2011 by active search (Crump 1971) and visual and acoustic encounters (Zimmerman 1994) at four distinct sites: stream with sandy bottom (SSB); stream with pebbles (SP); marsh in open area (MOA); and marsh in forest remnant (MFR).Sites were chosen based on ease of access and potential to attract frogs.Reproductive modes of recorded frogs were classified according to the literature (Haddad & Prado 2005;Haddad et al. 2008Haddad et al. , 2013)).
Sampling was conducted between 15:00 and 00:00 h.Each site was visited three times on each sampling date, with equal amounts of time spent between different habitats within a site (45 minutes).All species seen or heard were recorded.Habitats were also visited ad libitum during the day, on separate dates from samplings, to record daytime activities.When required, specimens were collected and deposited in the Museu de Zoologia Profª.Morgana Cirimbeli Gaidzinski (MUESC) at Universidade do Extremo Sul Catarinense, Criciúma, Santa Catarina, Brazil richness and abundance of vocalizing males was verified by applying Spearman's correlation coefficient (r s ) (Zar 1999).The month of September was excluded from this analysis because no species vocalization activity was recorded for that month.
Abiotic data (temperature, relative humidity, and rainfall) associated with sampling dates were obtained from the Experimental Station of EPAGRI at Urussanga (6 km from the study site).Data relating to photoperiod were obtained from the Brazilian National Observatory (http:// euler.on.br/ephemeris/index.php).Temperature, rainfall, and humidity data were reported as averages of five days per month (four days before and on the day of sampling).

RESULTS
Twenty-six species distributed amongst nine families and 17 genera were recorded.Hylidae was the richest (n = 13 species), followed by Leptodactylidae (n = 5) and Bufonidae (n = 2).Six families contributed with only one species each (Table 1).The environments with the highest species richness were MOA (n = 15 species), followed by SP (n = 13 species), SSB (n = 11), and MFR (n = 8).The recorded variation in richness between environments was not statistically significant, as demonstrated by the Chi-square test (Table 1).
The general diversity index for the area was high (H = 3.037), although values for diversity and evenness were distinct among the sampling sites (Hʹ for SSB = 1.907 ns ; SP = 2.078 ns ; MOA = 2.062 ns ; MFR = 1.094 ns ), no significant differences were detected (p > 0.05) (Table 1).The species accumulation curve showed that the sampling effort was sufficient to register 86.66% of species potentially present in the study area.The difference between the estimated maximum number of species and the number registered was approximately four species (Figure 2).
Nine species were considered rare, nine were abundant, and eight were common species (Table 2).Vocalization activity observed during the study indicated greater activity between October and January (Figure 1; Table 2).The period with the lowest number of vocalizing male species was in autumn (eight species), whereas the period with the highest number occurred in summer (18 species).Among inventoried species, Vitreorana uranoscopa vocalized only in February after a strong storm, Rhinella abei, Aplastodiscus ehrhardti, Bokermannohyla hylax, Hypsiboas bischoffi, and Ololygon catharinae were the only species recorded between April and July, and no species were recorded in September.Eleven reproductive modes were recorded for the study area (Table 2).Mode 1 (eggs and tadpoles in still water) Table 2. Vocalization activity period, constancy of occurrence index and reproductive modes of amphibian species recorded in Urussanga, SC, between November 2010 and December 2011.Where: OF = occurrence frequency; CI = constancy index (Dajoz, 1983), C (Common species), A (abundant species) and R (rare species); RM = reproductive modes (sensu Haddad & Prado, 2005).was the most frequent, being represented in 10 of the 26 species (38%).

DISCUSSION
The 26 recorded species in the present study corresponded to 18% of 144 species listed for Santa Catarina (Lucas 2008) and about 4.9% of 529 recorded species in the Atlantic Forest (Haddad et al. 2013).The number of recorded species in this study was similar to those recorded in other studies conducted in Santa Catarina (e.g., Bastiani & Lucas 2013;Hartmann et al. 2008 Temporary habitats (including the MOA site) tend to have less pressure from predators than permanent environments (Heyer et al. 1975), thereby allowing diversity and species richness (Skelly 1997).In contrast with permanent environments tend to be colonized by species with long reproductive periods, such as Dendropsophus minutus and Hypsiboas faber and species of the family Leptodactylidae (Maragno et al. 2013;Wells 1977).The abundance of rare species observed in this study may be due to occasional records while anurans were moving to breeding sites or to shelter areas, which were not the sampling sites themselves (e.g., Phyllomedusa distincta, Scinax granulatus, and Scinax perereca).Some species are explosive breeders (e.g., Proceratophrys boiei and Vitreorana uranoscopa) and appeared/vocalized during or after heavy rains (Duellman & Trueb 1994).
In subtropical regions, temperature and photoperiod are the most important factors regulating anuran vocalization activity (Canavero et al. 2009).In this study, vocalization activity was not correlated with temperature, relative humidity, or precipitation, but it may be associated with stability of variables during the seasons and year (EPAGRI 2001).Photoperiod was the only environmental variable that influenced variations in species richness and abundance of vocalizing males.Temperature is known to fluctuate directly with the photoperiod, and temperatures are highly variable throughout the year in southern Brazil (0-39°C; EPAGRI 2001).Anurans may use photoperiod as an indicator of suitable conditions for reproduction (Both et al. 2008;Duellman & Trueb 1994).In this study, most anuran species were found to reproduce throughout the year, with a decline from late summer (February) until winter (July), changing from higher to lower photoperiods.
The diversity of reproductive modes found in a coalfield in Urussanga municipality, was likely due to heterogeneity of the Atlantic Forest (e.g., presence of bromeliads, mountain streams, and moist leaf litter).Heterogeneity allows the occurrence of generalized reproductive strategies (direct development of terrestrial eggs [mode 23], foam nest floating on pond and exotrophic tadpoles in ponds [mode 11]) to specialist reproductive modes, as for V. uranoscopa and H. meridionalis that require more pristine environments (eggs from which exotrophic tadpoles hatch and drop into running water [mode 25] and eggs and tadpoles in streams [mode 3], respectively).Faunal studies conducted in the state of Santa Catarina, especially in the far southern region, are important due the paucity and absence of temporal and spatial studies in an area that hosts the national critical area for pollution control and three areas of full protection (one state and two federal conservation areas).The present study demonstrates that even small and disturbed areas harbor important, threatened species (in both national and state lists), and may help to fill the knowledge gaps about the diversity in state.Ugioni, Guilherme de Lucca, and Fábio Hammen Llanos for their assistance in the research project, and the anonymous reviewers for their valuable comments.

Figure 1 .
Figure 1.Map showing the study site in the state of Santa Catarina, Southern Brazil.

Figure 4 .
Figure 4. A. Graphic representing Spearman's correlation between photoperiod variation and monthly richness of species (rs = 0.8 and p ≤ 0.001) and, B. Graphic representing Spearman's correlation (rs = 0.67 and p ≤ 0.001) between photoperiod variation and abundance of vocalizing males, between November 2010 and December 2011.

Figure 3 .
Figure 3. Curve of species accumulation, based on the monthly occurrence of anurans, recorded on the four sampling sites in Urussanga, SC, from November 2010 to December 2011.Legend: Solid black line with asterisk: Jacknife1, Solid black line with square: Bootstrap and Solid black line: Observed richness.

Table 1 .
Anurans amphibians recorded in four sampling sites in Urussanga, SC, between November 2010 and December 2011, where: SSB = stream with sandy bottom; SP = stream with pebbles; MOA = marsh in open area and MFR = marsh in forest remnant.*Species without vocalization activity.