Amphibian Richness Patterns in Atlantic Forest Areas Invaded by American Bullfrogs

bs_bs_banner

Austral Ecology (2014) 39, 864–874

Amphibian richness patterns in Atlantic Forest areas invaded by American bullfrogs

CAMILA BOTH,1,2* BRUNO MADALOZZO,3 RODRIGO LINGNAU4 AND TARAN GRANT1,5 1Programa de Pós-Graudação em Zoologia, Pontifícia Universidade Católica do Rio Grande do Sul, Porto Alegre, 3Programa de Pós-Graduação em Biodiversidade Animal, Universidade Federal de Santa Maria, Santa Maria, 4Universidade Tecnológica Federal do Paraná, Francisco Beltrão, and 5Departamento de Zoologia, Instituto de Biociências, Universidade de São Paulo, São Paulo, Brazil; and 2School of Biological Sciences, University of Sydney, Sydney, New South Wales, Australia (Email: [email protected])

Abstract The relationship between invasion success and native biodiversity is central to biological invasion research. New theoretical and analytical approaches have revealed that spatial scale, land-use factors and community assemblages are important predictors of the relationship between community diversity and invasibility and the negative effects of invasive species on community diversity. In this study we assess if the abundance of Lithobates catesbeianus, the American bullfrog, negatively affects the richness of native amphibian species in Atlantic Forest waterbodies in Brazil. Although this species has been invading Atlantic Forest areas since the 1930s, studies that estimate the invasion effects upon native species diversity are lacking. We developed a model to understand the impact of environmental, spatial and species composition gradients on the relationships between bullfrogs and native species richness. We found a weak positive relationship between bullfrog abundance and species richness in invaded areas. The path model revealed that this is an indirect relationship mediated by community composition gradients. Our results indicate that bullfrogs are more abundant in certain amphibian communities, which can be species-rich. Local factors describing habitat heterogeneity were the main predictors of amphibian species richness and composition and bullfrog abundance. Our results reinforce the important role of habitats in determining both native species diversity and potential invasibility.

Key words: Anura, Brazil, diversity, invasion, Lithobates catesbeianus, Ranidae.

INTRODUCTION Charles Elton predicted a scenario of ecological and economical losses linked with species invasions The immigration of novel species into communities or (Simberloff 2011), and since his seminal synthesis in uncolonized patches is an important process for com- ‘The Ecology of Invasions by Animals and Plants’ munity development, structure and composition (Elton 1958) there is growing interest to identify (MacArthur & Wilson 1967; Connell & Slatyer 1977; species traits linked to invasion propensity and com- Loreau & Mouquet 1999).We are currently witnessing munity properties facilitating or preventing invasions. an age of biological invasions, with novel species In his book, Elton observed that ‘tropical systems are less being incorporated into local communities not only invaded than species poor temperate and boreal systems’. by dispersal, but also by intentional or accidental Elton associated ‘tropical resistance’ with highly introductions. Such invasions are not restricted to diverse communities, harbouring large numbers of taxa, regions, biomes, or continents (Soulé 1990), and enemies and parasites, which, in turn, can impede the they have attracted much attention because they can establishment of novel species. This idea is currently be economically expensive. For instance, exotic species recognized as the diversity–invasibility hypothesis (see can negatively affect crops, ﬁsheries, public health Justus 2008). Elton himself recognized that this should systems etc., resulting in the expense of billions of only partially explain the low invasion rates in the dollars (Pimentel et al. 2001). Further, invasions can tropics. Studies subsequent to Elton (1958) explained also be expensive in an ecological and evolutionary the diversity–invasibility hypothesis in the light of context, since they have been linked to recent species resource exploitation. Following this rationale, diverse extinctions (Clavero & García-Berthou 2005). tropical communities are more resistant because richer communities tend to be saturated with species that *Corresponding author. exploit resources in complementary ways (MacArthur Accepted for publication April 2014. & Levins 1967; MacArthur 1970).

The diversity–invasibility hypothesis is still a central species and community diversity patterns, asking issue in ecology (Fridley et al. 2007).There is evidence whether invasions are related to less diverse commu- from observational studies that alien species diversity nities or differences in community composition, while is low in richer systems, although some studies have controlling for environmental and spatial variation. found no relationship at all (e.g. Case & Bolger 1991). Classical examples of invasion effects on ecosystem Experimental studies supplied evidence that supports diversity come from freshwater habitats that have suf- both negative (e.g. Tilman 1997) and positive relation- fered historically from structural modifications. These ships (e.g. Robinson et al. 1995) between diversity and are among the most endangered ecosystems, and invasion. Nevertheless, there are also experimental biotic modifications imposed by the addition of inva- studies highlighting the lack of a relationship between sive species could lead to local extinction of freshwater diversity and invasion (e.g. Robinson & Dickerson species (Witte et al. 1992; Dodds & Whiles 2010). 1984). The presence of such contrasting results has Waterbodies such as lakes, ponds, pools and tempo- been termed ‘the invasion paradox’ (Fridley et al. rary streams can be viewed as small islands, which 2007) and seems to be related to spatial scale. In an seem to be very sensitive to invasions (Ricciardi & extensive review, Fridley et al. (2007) showed that MacIsaac 2011). The bullfrog (or American bullfrog) large-scale studies suggest a positive relationship Lithobates catesbeianus (Shaw 1802) is a member of the between invasive species and native richness. Such ‘hall of fame’ of freshwater invasive species. It is cited positive correlation indicates that richer communities as one of the 100 worst alien species in the world are able to accommodate introduced species, despite (Lever 2003) and is one of the causes of amphibian high numbers and abundance of native species, known declines in North America. Furthermore, the species as ‘biotic acceptance’ (Stohlgren et al. 2006). In con- appears to be an important vector of fatal amphibian trast, at smaller scales relationships are often negative, diseases such as chytridiomycosis (Daszak et al. 2004; especially in experimental studies (Fridley et al. 2007). Schloegel et al. 2010). Bullfrogs benefit from human- Nonetheless, a positive relationship between non- modified landscapes (Adams 1999; Zampella & native and native species might be due to sampling Bunnell 2000; D’Amore et al. 2010; Fuller et al. artefacts (Fridley et al. 2004). Apparently positive or 2011), the presence of non-native fish in fresh waters negative relationships might also result from analytical (Adams & Wasson 2000; Adams et al. 2003), and models that ignore important co-variables that might simple aquaculture enclosures that enable a continu- be related to invasive species abundance or richness ous source of propagules (Liu & Li 2009). and native species diversity (Taylor & Irwin 2004). Bullfrogs originate from eastern North America Factors such as propagule pressure, disturbance and and are distributed from Mexico to south Canada; other environmental gradients often function synergis- however, invasive populations occur in western North tically to explain invasion patterns (Elton 1958; Von America, Europe, Asia, and Central and South Ameri- Holle & Simberloff 2005). Therefore, in situ studies can countries where they were introduced for aquacul- that consider distinct spatial scales and address the ture practices (Santos-Barrera et al. 2009). In invaded relationship between native richness and invasive sites, they can be negatively correlated with native frog species are required for a thorough understanding of abundance and with the occurrence of certain native the diversity–invasibility hypothesis. This is especially frog species; they can also reduce tadpole survival true for approaches that consider causal links connect- (Kupferberg 1997; Kiesecker & Blaustein 1998; Kats ing environmental and community gradients. & Ferrer 2003; Wang & Li 2009) and are often impli- In many systems, invasions are a concrete reality, cated as a factor in the decline of native amphibians despite several decades of research and efforts for their and potential species loss (Kraus 2009). A negative prevention (Richardson 2011). In previously invaded relationship between bullfrogs and native anuran rich- ecosystems, we are challenged to estimate the invasive ness has been reported in China (Li et al. 2011). species effects on diversity rather than to understand Bullfrogs have been present in Brazil since the whether diversity will prevent invasions.The same pre- 1930s, and invasions have continued to occur and dictions about the relationship between diversity and expand since then, probably growing exponentially invasibility described above also hold for the relation- from the 1970s to the 1990s when bullfrog farms ship between invasive species effects on diversity. received governmental incentives (Lima & Agostinho Accordingly, at fine spatial scales negative relation- 1988). Brazil is highly suitable for bullfrog establish- ships are expected to occur, because it is at smaller ment, especially in the Atlantic Forest, and climate spatial scales that species interact. The problem faced change might make protected forest areas even more when studying an invasion that has reached the final prone to invasion by bullfrog populations (Nori et al. stages is that, without data collected prior to the inva- 2011; Loyola et al. 2012). Although the relationship sion, we cannot know if invasive species promoted between invasive bullfrogs and the diversity of native changes to communities in the past. Nevertheless, we communities in the Atlantic Forest is currently can study the relationship between current invasive unknown, we do know that bullfrogs are widespread in

Brazil and that the number of known localities is breeding (calling males, eggs, tadpoles), and 15 waterbodies increasing (Both et al. 2011b). without, alternating spatially.This site selection allowed us to In this study, we tested relationships between the collect data for invaded and non-invaded sites in each study abundance of the non-native American bullfrog and area.We considered sites that presented evidence of breeding native diversity in areas where the invasion had to be invaded, given that eggs, tadpoles or adults were present.Water bodies with no evidence of breeding are likely reached a fully invasive degree (sensu Blackburn et al. to be non-invaded. Juveniles are known to disperse from their 2011). Our primary goal was to test whether bullfrog natal waterbody and can occasionally be found in ephemeral abundance can predict richness patterns by assessing pools or when crossing roads at night (Both, pers. obs., how amphibians in Atlantic Forest communities vary 2010); therefore, the presence of juveniles in ponds is likely in richness in invaded areas. We expected a negative to be temporary and we did not consider it to be evidence relationship because we were investigating the relation- that a site has been invaded. ship between invasive species abundance and native We sampled a total of 90 waterbodies: 32 in A1, 30 in A2 community richness at a fine spatial scale. Further- and 28 in A3 (see Appendix S1).We surveyed each site twice more, we developed a model to understand the inter- for post-metamorphic individuals, tadpoles and egg clutches. play of environmental, spatial and species composition Collections were made over two 25-day periods in February/ gradients with the relationship between bullfrogs and March (late summer) and October/November (spring) of 2010. Sampling months were chosen to represent periods of native species richness. longer photoperiod, which is a predictor of high native amphibian breeding activity in subtropical regions (Canavero & Arim 2009), and also coincides with the breeding period of METHODS bullfrogs in southern Brazil (Kaefer et al. 2007). We sampled 4–6 waterbodies per day for eggs and tadpoles in the daytime and post-metamorphic individuals at night. Study areas We sampled tadpoles in distinct microhabitats with dip net sweeps (40 × 30 cm, mesh 0.02 mm), always using the same The study was conducted in three different areas of the collector. At least five sweeps were undertaken in each Atlantic Forest domain in southern Brazil: the central region microhabitat (covering approx. 1 m2).The effort was propor- of Rio Grande do Sul State and the western and eastern tional to pond size and heterogeneity (Shaffer et al. 1994). portions of Santa Catarina State (see Appendix S1 for Microhabitats were classified into marginal/shallow veg- coordinates). These areas are highly susceptible to bullfrog etated, marginal/shallow non-vegetated, centre/deep veg- invasions and are in regions where records of L. catesbeianus etated and centre/deep non-vegetated types. Marginal/ in Brazil are concentrated (Both et al. 2011b). Invasions date shallow microhabitas are generally <30–50 cm deep, while to at least 10 years ago in all cases. There were no active centre/deep habitats are generally >50 cm deep. Heterogene- bullfrog farms near the study areas. In study area A1 (central ity is positively correlated with waterbody depth and size; Rio Grande do Sul), the vegetation is characterized by sea- therefore, fewer microhabitats were sampled in small, sonal deciduous forest (IBGE (Instituto Brasileiro de shallow, non-vegetated waterbodies (approx. 4–6) than in Geografia e Estatística 2004). In western Santa Catarina, large, permanent, vegetated waterbodies with all four types of study area A2, two phytogeographic forms of Atlantic Forest microhabitats (approx. 20 microhabitats sampled). In the occur, including seasonal deciduous forest and mixed largest ponds we also sampled microhabitats midway ombrophilous forest (IBGE (Instituto Brasileiro de between the margin and the centre. The total number of Geografia e Estatística 2004; Lucas & Fortes 2008). In microhabitats sampled in each waterbody in a single survey eastern Santa Catarina, study area A3, the vegetation is ranged from 4–30. Samples were spaced at least 2 m apart. formed by dense ombrophilous forest (IBGE (Instituto Samples were taken between 09:00–19:00 h, always during Brasileiro de Geografia e Estatística 2004). The Atlantic daylight. Forest in the three study areas is highly fragmented. The Post-metamorphic surveys were begun 30 min after general land use is farming and cattle grazing, with larger sunset, and we searched for individuals along the perimeters forest fragments found in protected areas. In A1, Parque of breeding sites.We counted all post-metamorphic individu- Estadual da Quarta Colonia protects 1847.90 ha along the als, separating calling and non-calling individuals. Sampling border of the Dona Francisca hydroelectric dam. A2 includes effort was proportional to size and heterogeneity of the the Floresta Nacional de Chapecó, with 1606.3 ha divided in waterbody (Scott & Woodward 1994). two areas. Parque Nacional da Serra do Itajaí, with 57 374 ha, is located within A3. Waterbody and landscape descriptors

Waterbody selection and sampling We measured the distance from the waterbody to the nearest forest fragment and to the nearest road using GPS In spring 2009, we carried out initial surveys to select coordinates. We expanded these distances in second- and waterbodies for subsequent sampling, including natural third-order monomials to test for potential non-linear rela- marshes, ponds, stream sections with low water flow tionships with richness. Area and depth of each waterbody (streamside pools), and artificial ponds. In each area we were measured at each collection event. For depth, we took identified 15 waterbodies with evidence of L. catesbeianus the measurements in each of the microhabitats sampled for doi:10.1111/aec.12155 © 2014 The Authors Austral Ecology © 2014 Ecological Society of Australia BULLFROG INVASION AND ANURAN DIVERSITY 867

tadpoles and recorded the mean, maximum, and minimum and spatial variation. This analytical approach allowed us to depth. Water surface area ratio obtained from the two sam- depict the co-variation structure between predictors (Shipley pling events (summer area/spring area) was used as a 2000). First, we tested whether bullfrog abundance affects measure of pond permanence. We recorded hydrophyte native richness. Using path analysis, we then tested whether morphotype (floating and emergent macrophytes) richness in species richness is related to (i) waterbody, (ii) landscape, the pond. We visually estimated the hydrophyte coverage of and (iii) spatial descriptors, jointly with (iv) community com- waterbodies and classified it into three classes: <30%, position and (v) bullfrog abundance. In order to assess the 30–60% and >60%. Fish presence/absence was determined relative importance of all five sets of predictors, we built a by visual surveys or dip net captures or was based on infor- hierarchical theoretical model of potential causal relation- mation provided by local rural owners. We classified the ships explaining native species richness (Fig. 1). This model vegetation of waterbody banks into low grass, tall grass, assumes a hierarchical causal order between predictors, shrubs, and trees and used the sum of types present in a site where spatial descriptors have the highest causal order and all as a descriptor of the bank vegetation structure. other variables are endogenous (see causal links in Fig. 1). We tested this model using the analytical steps proposed by Brum et al. (2012). First, we performed a model selection procedure using each set of descriptors as predictors of native Spatial descriptors richness to identify the descriptors that are directly related with native richness. We then regressed all selected descrip- This study was performed at two major scales: between areas tors from all sets against richness to test for direct causal (A1, A2 and A3) and within areas.To account for this nested links. Obeying the causal hierarchy, we regressed all variables spatial arrangement, we considered two kinds of spatial directly linked with richness on their respective potential descriptors: the three large study areas represented by predictors (immediate higher hierarchy factors, Fig. 1), and dummy variables, and Moran’s eigenvector maps (MEMs). so on, until we reached the exogenous descriptors, or spatial MEMs provide orthogonal spatial variables ranging from descriptors in this case. For instance, if a waterbody descrip- broad, inter-area scales to the finest scales derived from the tor was directly related to local richness, we further geographic coordinates of the sites (Dray et al. 2006). The investigated which landscape and spatial descriptors could spatial arrangement in this study shows a high truncation explain this descriptor. If a landscape descriptor was = value (minimum distance connecting all points 334 km) selected, we inspected its relationships with high-order between study areas. We used a nested MEM model spatial descriptors. Fish presence was considered together (Declerck et al. 2011), where MEM variables are blocked with other waterbody descriptors. However, we investigated within study areas to describe spatial relationships at this further causal links associated with this variable (see Fig. 1) scale. We used the R function create.MEM.model (Declerck because it is a well-known filter for lentic communities et al. 2011) to build the MEM model using a matrix of (Wellborn et al. 1996). Beta regression coefficients of linear dummy variables that described the three regions as the models were taken as path coefficients. Model selection was connectivity criteria. The nested analysis resulted in seven based on corrected Akaike information criterion (Anderson spatial vectors describing the spatial arrangement of sites 2008). All regression models were performed in R (R within areas. These seven MEM vectors and the dummy Development Core Team 2012). variables of the three large study areas were used as spatial descriptors. RESULTS

Data analyses Across the three study areas we recorded 40 native amphibian species: 19 species in A1, 19 in A2 and 23 To remove the effects of sampling effort on richness values, in A3 (Table 1). Native amphibian richness ranged we regressed richness on the abundance of post- from 0–11 species, with an average of 3.9 species per metamorphic individuals, which showed a 96% concordance = with total richness, and used the residuals, henceforth called waterbody (SD 2.8).The ordination of species com- richness, in subsequent analyses.To describe species compo- position resulted in 37 PCoA axes.The first three axes sition across waterbodies, we calculated Jaccard’s index of represented 41% of the species composition gradients similarity for pairs of communities and performed a principal and were used in the regression models. The stability coordinate analysis (PCoA) (Legendre & Legendre 1998) of the first three ordination axes was validated by boot- using only occurrence (presence/absence) data. We chose strap resampling. The first two axes represented the occurrences instead of species abundances because they composition gradient from ombrophilous vegetation allowed us to combine post-metamorphic and tadpole to mixed ombrophilous vegetation to seasonal decidu- surveys, thereby dispensing with further weighting assump- ous forest amphibian communities (Fig. 2A). The tions about the importance of distinct life history stages to first principal coordinate (PC1) axis clearly varied the characterization of community composition.The stability of ordination axes was evaluated by bootstrap resampling from amphibian compositions restricted to dense (Pillar 1999). The PCoA analysis was performed in Multiv ombrophilous forest (e.g. Sphaenorhynchus sp. and (Pillar 2006). Bokermanohyla hylax) to those occurring in high fre- We used successive regression models to analyse the effects quencies across the study regions (e.g. Dendropsophus of bullfrogs on richness, taking into account environmental minutus and Physalaemus cuvieri) (Fig. 2B).

Fig. 1. Theoretical path model explaining relationships of native amphibian richness with community gradients, space, landscape, waterbody descriptors and American bullfrog abundance in invaded areas.

We detected bullfrogs at 62 of the 90 sampled richness was also only indirectly linked to richness waterbodies, 44 of which showed evidence of breeding through the composition gradients. Hydrophyte rich- (16 in A1, 16 in A2 and 12 in A3), indicating the ness directly determined bullfrog abundance and com- presence of established populations. At least eight sites munity composition gradients PC1 and PC2 (Fig. 3). in each study area lacked evidence of bullfrogs at any The final model accounted for 63% of native richness life stage. The linear model relating native amphibian variation. richness to bullfrog abundance showed a weak positive relationship (Table 2). Richness was also positively related to the first two principal coordinates of com- DISCUSSION munity composition, PC1 and PC2 (Fig. 2). Three MEMs were selected in the spatial model: MEM-2, Traditionally, the diversity–invasibility relationship has which described spatial structure within A2; MEM-3, been explored in order to investigate whether diversity which described spatial structure within A3; and will prevent invasions. In the face of increasing inva- MEM-7 describing spatial structure within A1. The sion rates and successful establishment of invaders, it model relating richness with local waterbody des- has become important to explore the effect of inva- criptors selected maximum depth, hydrophyte rich- sions on species diversity at a local scale. In this study, ness, bank vegetation structure, fish presence and we assessed whether invasion affects diversity in streamside pools as predictors (Table 2). already successfully occupied areas, expecting a nega- The path model showed that native amphibian rich- tive relationship between bullfrog abundance and ness was directly determined by the first two principal native amphibian richness. Contrary to our initial coordinates of community composition (PC1 and expectation, we found a weak positive relationship, PC2), maximum depth, and the binary descriptors for which was revealed to be indirectly linked to species fish presence, as well as by one spatial model, MEM-2 composition gradients. Local environmental gradients (Fig. 3).The composition gradients described by PC1 describing waterbody features directly explained all and PC2 showed the highest path coefficients, and biotic components of the model. They also showed an therefore are the main predictors of native amphibian additional and important influence of species compo- richness. Stream side waterbodies and two spatial sition gradients on native richness. Bullfrog abundance models (MEM-3 and MEM-4), did not show signifi- also responded to the same waterbody gradients. cant causal links with amphibian richness. Bullfrog Across the three Atlantic Forest areas, the influence of abundance was only indirectly related to amphibian space upon richness patterns was small.We found only species richness through its relationship with the com- weak relationships between micro-regions and lower munity composition gradient of PC1. Hydrophyte richness in A1 and A2, and higher richness in A3. doi:10.1111/aec.12155 © 2014 The Authors Austral Ecology © 2014 Ecological Society of Australia BULLFROG INVASION AND ANURAN DIVERSITY 869

Table 1. Amphibian composition from southern Atlantic Forest areas invaded by Lithobates catesbeianus

Species Abbreviation Post-metamorphics Tadpoles A1 A2 A3

Bufonidae Rhinella abei Rab 1 1 1 Rhinella fernandezae Rfe 1 1 Rhinella icterica Ric 1 1 1 1 Cycloramphidae Limnomedusa macroglossa Lma 1 1 Hylidae Bokermannohyla hylax Bhy 1 1 1 Dendropsophus minutus Dmin 1 1 1 1 Dendropsophus microps Dmic 1 1 1 Dendropsophus nahdereri Dnah 1 1 Dendropsophus nanus Dnan 1 1 1 Dendropsophus sanborni Dsa 1 1 1 Dendropsophus werneri Dwe 1 1 1 Hypsiboas albomarginatus Halb 1 1 1 Hypsiboas albopunctatus Halbp 1 1 Hypsiboas bishofﬁ Hbi 1 1 Hypsiboas faber Hfa 1 1 111 Hypsiboas pulchellus Hpu 1 1 1 Hypsiboas semilineatus Hse 1 1 1 Hypsiboas cf. semiguttatus 1 Hsm1 1 1 Hypsiboas cf. semiguttatus 2 Hsm2 1 1 Phyllomedusa tetraploidea Pte 1 1 1 Phyllomedusa distincta Pdi 1 1 Scinax alter Sal 1 1 Scinax fuscovarius Sfu 1 1 1 1 1 Scinax granulatus Sgr 1 1 111 Scinax perereca Spe 1 1 1 1 Sphaenorhynchus sp. Sphe 1 1 Hylodidae Crossodactylus sp. Cro 1 1 Leiuperidae Physalaemus gracilis Pgr 1 111 Physalaemus cuvieri Pcu 1 1 1 1 1 Physalaemus nanus Pna 1 1 1 Pseudopaludicola falcipes Pfa 1 1 1 1 Leptodactylidae Leptodactylus fuscus Lfu 1 1 1 1 1 Leptodactylus gracilis Lgr 1 1 1 1 Leptodactylus labyrynthicus Llab 1 1 Leptodactylus joly Ljo 1 1 Leptodactylus latinasus Llat 1 1 1 Leptodactylus latrans Llat 1 1 1 1 1 Leptodactylus mystacinus Lmy 1 1 1 Leptodactylus plaumani Lpl 1 1 Microhylidae Elachistocleis bicolor Ebi 1 1 1 1 Total 34 30 19 19 23

A1 = central region of Rio Grande do Sul, deciduous forest; A2 = western region of Santa Catarina, deciduous and mixed ombrophilous forest; A3 = east of Santa Catarina, ombrophilous forest.

However, only one spatial model was causally signiﬁ- expected to operate, and both would lead to negative cant when considering the entire co-variation struc- relationships (Elton 1958). However, it has been ture between all sets of predictors. shown that native species differ greatly in the way they At ﬁne spatial scales, negative relationships are are affected by invasive species. For instance, Thomaz expected because it is at these scales that individuals and Michelan (2011) studied the effects of an invasive actually interact. At this local scale, biotic resistance of macrophyte on several native macrophytes at distinct the native community or impact of invasive species are spatial scales and observed that only some of the native

ship is mediated by community composition. Bull- frogs were more abundant in certain species-rich communities. Bullfrog abundance was positively related to the PC1 community gradient, varying in a similar fashion in this gradient to common species like Physalaemus cuvieri and Dendropsophus minutus that are broadly distributed in South America (IUCN 2011). These species have been recognized as anthropogenically adapted (Santos et al. 2007), like bullfrogs. These results are in accordance with those of Bunnell and Zampella (2008), who found that bullfrogs are associated with distinct species compositions, being more closely associated with species that have a widespread distribution. However, bullfrogs do co-occur with other communities in the Atlantic Forest, albeit at lower abundance, and might affect the communities by other pathways. For instance, Both and Grant (2012) found evidence that bullfrogs have the potential to affect the acoustic niche of native amphibians: tree-frogs naïve to bullfrog vocalizations significantly altered advertisement call frequencies in response to bullfrog advertisement calls. In this case, bullfrogs might not cause an immediate decrease in richness but could affect female mate selection and result in important but less immediate fitness consequences (Both & Grant 2012). Our results highlight how all biotic components – species richness, composition, fish presence, and bullfrog abundance – respond directly or indirectly to waterbody descriptors. These waterbody descriptors act as local scale filters, structuring both native species diversity and bullfrog abundance. Hydrophyte richness and maximum depth were the main predictors describing waterbodies selected in our final Fig. 2. Ordination of amphibian species composition model. These predictors could be viewed as a across the three study areas. (A) First two axes of principal coordinate analysis (PCoA) revealing composition gradients measure of habitat heterogeneity, which undoubtedly across the study areas (PC1 and PC2). Triangles indicate plays an important role for both faunal biodiversity species composition from A1 (seasonal deciduous forest), and abundance. Heterogeneity has the potential to circles indicate compositions from A2 (mixed ombrophilous decouple trophic interactions, promoting greater forest), and squares from A3 (ombrophilous forest). (B) Cor- diversity (Kovalenko et al. 2012). Despite using a relation scatter plot for amphibian species with the first two simplistic measure of heterogeneity of the water PCoA axes. Amphibian species are identified by abbrevia- surface (richness of hydrophyte morphotypes), we tions (see Table 2). found that it was an important predictor of commu- macrophytes were affected, and, further, that these nity gradients. Hydrophytes could promote diversity negative associations are more common at smaller by providing shelter for tadpoles and post- scales. A frequently neglected issue is that at fine scales metamorphic individuals and calling sites or for adult species facilitation can occur, which would lead to males. Maximum depth is a gradient that contains positive relationships (Bruno et al. 2003). In our case, variation related to pond area, permanence, availabil- a structurally complex community might provide ity of other lower depths, and is also an important greater resources for bullfrogs, which are generalist resource for communities in small waterbodies. It predators (Bury & Whelan 1984). In turn, bullfrogs governs the coexistence of nektonic and benthic tad- might be able to explore diverse resources in poles in ponds and streams (Eterovick & Fernandes waterbodies, including anuran prey, without affecting 2001; Both et al. 2011a). Fish and bullfrog presence native diversity. are also regulated by this gradient, because only Here, we tested whether bullfrog abundance can waterbodies with a certain depth can support these predict anuran richness and found that this relation- organisms (Kushlan 1976; Bury & Whelan 1984). doi:10.1111/aec.12155 © 2014 The Authors Austral Ecology © 2014 Ecological Society of Australia BULLFROG INVASION AND ANURAN DIVERSITY 871

Table 2. Linear model relating native amphibian richness and bullfrog abundance (1) and best models relating native richness with three different sets of predictors: (2) community gradients, (3) waterbody descriptors and (4) spatial models

R2 AICc AICc wi

(1) Bullfrog abundance 0.05 405.17 – (2) Community gradients 0.51 348.24 0.75 PC1 + PC2 (3) Waterbody descriptors 0.31 384.94 0.27

Depthmax + bank vegetation structure* + hydrophyte richness + stream* + ﬁsh presence* (4) Spatial descriptors 0.14 401.22 0.10 MEM.2 + MEM.3 + MEM.7

The selection of best models was based on corrected Akaike information criterion (AICc) and Akaike weight (AICc wi). The selected models have the highest AICc wi, which describes the relative likelihood of the model, normalized across the set of all possible models to sum 1. Fixed factors are identiﬁed by (*). Principal coordinates of community composition (PC) were used as predictors in model 2 and Moran’s eigenvector maps (MEM) were used as predictors in model 4. Abbreviations in model (3):

Depthmax = maximum depth recorded for a given waterbody; bank vegetation structure = number of types of vegetation present in waterbody banks; stream = stream side pool (dummy variable describing type of waterbody).

Fig. 3. Causal relationships between selected spatial and environmental predictors, amphibian species composition gradients, bullfrog abundance and native amphibian richness. Standardized path correlation coefficient values follow each path. Solid arrows (—) represent positive relationships and dashed arrows (- - -) represent negative relationships. Only paths with P ≤ 0.05 are shown. U = non-determination coefficient (U = 1 − R2). Richness = native amphibian richness; Hydrophyte R = hydrophyte richness; Veg. types = number of structural vegetation types around the ponds; Depth = maximum depth recorded for a waterbody; bullfrogs = Lithobates catesbeianus abundance; Fish = fish presence/absence; PC1 = first principal coordinate of community composition; PC2 = second principal coordinate of community composition; MEM.2 = Moran’s eigenvector two, describing spatial structure within study area A1.

Habitat heterogeneity has been recognized as a key abundance with local waterbody descriptors observed factor determining the persistence of native frog here are encouraging, because they suggest a means populations in areas invaded by bullfrogs in North for habitat management to protect native frogs. In America. Adams et al. (2011) studied extinction this study, we found that waterbody depth was caus- probabilities of Rana aurora populations, testing ally linked with bullfrog abundance and species rich- whether they could be related to bullfrog invasion ness through positive and negative paths, respectively. and habitat modification. They did not find support- In addition, previous studies showed that waterbody ing evidence for an effect of bullfrogs on local extinc- modification favours bullfrogs and promotes changes tion probability for the native species. Their results in community composition (e.g. Bunnell & Zampella showed that vegetation cover and riparian vegetation 2008; Fuller et al. 2011). Waterbodies in areas of are the best predictors of low extinction risk on local special interest for conservation could be managed to scales for the native frog populations. The relation- favour native amphibians, or to prevent the establish- ship between species composition and bullfrog ment of bullfrogs.

Our study is not the first to note a positive relation- agree with recent studies suggesting that environmen- ship between amphibian community richness and tal gradients directly affect both native and invasive invasion.This pattern has been reported for native and species. Although we did not detect negative effects of invasive amphibian and reptile communities on bullfrogs on native richness, we do not conclude that broader spatial scales, suggesting the occurrence of such effects are absent in general. Disease transmis- biotic acceptance (Stohlgren et al. 2006). Poessel et al. sion, differential predation, and competition by inter- (2013) studied how native richness and alien amphib- ference are mechanisms that could plausibly operate ian species establishment are related in Europe and in invaded areas, even when bullfrogs have low North America and found positive relationships in abundance. both regions. They used invasive species richness as a measure of invasibility, and not single species abundance, as we used in the present study to access bullfrog effects. Nevertheless, they also concluded that ACKNOWLEDGEMENTS environmental gradients favouring higher native species richness also favour invasions. Additionally, it This study was supported by IGRE Associação Sócio is worth noting that only recent studies dealing with Ambientalista and received grants from Fundação o bullfrog invasion patterns highlight neutral relation- Boticário de Proteção a Natureza (0835_20092) and ships between bullfrogs and richness or emphasize the the Rufford Small Grants Foundation. The Conselho role of habitat filtering to promote or facilitate inva- Nacional de Desenvolvimento Científico provided a sions (e.g. D’Amore et al. 2010; Adams et al. 2011). student fellowship to CB and a grant (Proc. 476789/ Such results might be due to study spatial scale and/or 2009-5) and fellowship (Procs. 305473/2008-5, refinement of analytical approaches. Another possibil- 307001/2011-3) to TG. The Fundação de Amparo à ity is that over time, some communities might reach a Pesquisa do Estado de São Paulo provided a grant to new dynamic equilibrium state in which bullfrogs are TG (Proc. 2012/10000-5). ICMBio granted a collec- not a major threat. tion permit for this study (Proc. 23009-1).This manu- An important consideration when comparing the script benefited from comments by Sandra Hartz, results of the present study with those of previous Célio Haddad, Márcio Borges-Martins, Adriano studies in other regions is that the phylogenetic rela- Sanches Melo, and three anonymous reviewers.We are tionship between bullfrogs and native species is much thankful to all rural owners who granted access to closer in those regions than in the Atlantic Forest study sites and the Parque Nacional da Serra do Itajaí localities we studied. In Asia, Europe and North for logistical support. We also thank S. Declerck for America there are more ranid species. In China, where sharing the nested MEM function, V.F. Caorsi for her bullfrogs are known to negatively affect richness, native help with laboratory procedures, and D. Borges- species composition mostly consists of ranids (Li et al. Provette for help with the literature for tadpole 2011). In contrast, only a single native ranid species identification. occurs in the Atlantic Forest (Lithobates palmipes; Hillis & de Sá 1988), and it is restricted to the northern portion of the biome. No ranids occurred at any of the REFERENCES localities we studied, and the closest relative we found in sympatry with bullfrogs is the microhylid Adams M. J. (1999) Correlated factors in amphibian decline: Elachistocleis bicolor.The potential influence of related- exotic species and habitat change in Western Washington. J. ness upon invasions was recognized by Darwin (1859) Wildl. Manage. 63, 1162–71. in ‘The Origin of Species’. He suspected that a low Adams M. J., Pearl C. A. & Bury R. B. (2003) Indirect facilitation degree of relatedness should give some advantage to of an anuran invasion by non-native fishes. Ecol. Lett. 6, novel species in new habitats. This idea deserves 343–51. further investigation, considering the global scale of Adams M. J., Pearl C. A., Galvan S. & Mccreary B. (2011) invasions by bullfrogs and many other alien species. Non-native species impacts on pond occupancy by an anuran. J.Wildl. Manage. 75, 30–5. In conclusion, we found only a weak and indirect, Adams M. J. & Wasson K. (2000) Pond permanence and the positive relationship between native anuran richness effects of exotic vertebrates on anurans. Ecol.Appl. 10, 559– and bullfrog abundance, which is mediated by species 68. composition at the invaded sites. Our results indicate Anderson D. R. (2008) Model based Inference in the Life Sciences. biotic acceptance of bullfrogs in the Atlantic Forest.We Springer, New York. observed that bullfrogs were more abundant in some Blackburn T. M., Pyšek P., Bacher S. et al. (2011) A proposed unified framework for biological invasions. Trends Ecol. Evol. amphibian communities that include anthropogenic- 26, 333–9. adapted species. The main result is that local Both C., Cechin S. Z., Melo A. S. & Hartz S. M. (2011a) What waterbody descriptors predict native richness and controls tadpole richness and guild composition in ponds in composition and bullfrog abundance. These results subtropical grasslands? Austral Ecol. 36, 530–6. doi:10.1111/aec.12155 © 2014 The Authors Austral Ecology © 2014 Ecological Society of Australia BULLFROG INVASION AND ANURAN DIVERSITY 873

Both C. & Grant T. (2012) Biological invasions and the acoustic Fridley J. D., Stachowicz J. J., Naeem S. et al. (2007) The inva- niche: the effect of bullfrog calls on the acoustic signals of sion paradox: reconciling pattern and process in species white-banded tree frog. Biol. Lett. 8, 714–16. invasions. Ecology 88, 3–17. Both C., Lingnau R., Santos A. Jr. et al. (2011b) Widespread Fuller T. E., Pope K. L., Ashton D. T. & Welsh H. H. J. (2011) occurrence of the American bullfrog, Lithobates catesbeianus Linking the distribution of an invasive amphibian (Rana (Shaw, 1802) (Anura: Ranidae), in Brazil. S.Am. J. Herpetol. catesbeiana) to habitat conditions in a managed river system 6, 127–34. in northern California. Restor. Ecol. 19, 204–13. Brum F. T., Kindel A., Hartz S. M. & Duarte L. D. S. (2012) Hillis D. M. & de Sá R. O. (1988) Phylogeny and taxonomy of Spatial and phylogenetic structure drive frugivory in the Rana palmipes group (Salientia: Ranidae). Herpetol. Tyrannidae birds across the range of Brazilian Araucaria Monogr. 2, 1–26. forests. Oikos 121, 899–906. IBGE (Instituto Brasileiro de Geografia e Estatística) (2004) Bruno J. F., Stachowicz J. J. & Bertness M. D. (2003) Inclusion Mapa de vegetação do Brasil. [Cited 12 Apr 2012.] Avail- of facilitation into ecological theory. Trends Ecol. Evol. 18, able from URL: http://www.ibge.gov.br/home/geociencias/ 119–25. default_prod.shtm Bunnell J. F. & Zampella R. A. (2008) Native fish and anuran IUCN (2011) The IUCN Red list of threatened species.Version assemblages differ between impoundments with and 2011.2. [Cited 10 Nov 2011.] Available from URL: http:// without non-native centrarchids and bullfrogs. Copeia 2008, www.iucnredlist.org/ 931–9. Justus J. (2008) Complexity, diversity, stability. In: A companion Bury R. B. & Whelan J. A. (1984) Ecology and Management of the to the Philosophy of Biology (eds S. Sarkar & A. Plutynski) Bullfrog. US Department of the Interior, Fish and Wildlife pp. 321–50. Blackwell, Oxford. Service, Washington, DC. Kaefer Í. L., Boelter R. A. & Cechin S. Z. (2007) Reproductive Canavero A. & Arim M. (2009) Clues supporting photoperiod as biology of the invasive bullfrog. Ann. Zool. Fenn. 44, 435–44. the main determinant of seasonal variation in amphibian Kats L. B. & Ferrer R. P. (2003) Alien predators and amphibian activity. J. Nat. Hist. 43, 2975–84. declines: review of two decades of science and the transition Case T. J. & Bolger D. T. (1991) The role of introduced species to conservation. Divers. Distrib. 9, 99–110. in shaping the distribution and abundance of island reptiles. Kiesecker J. M. & Blaustein A. R. (1998) Effects of introduced Evol. Ecol. 5, 272–90. bullfrogs and smallmouth bass on microhabitat use, growth, Clavero M. & García-Berthou E. (2005) Invasive species are a and survival of native red-legged frogs (Rana aurora). leading cause of animal extinctions. Trends Ecol. Evol. 20, Conserv. Biol. 12, 776–87. 110. Kovalenko K. E., Thomaz S. M. & Warfe D. M. (2012) Habitat Connell J. H. & Slatyer R. O. (1977) Mechanisms of succession complexity: approaches and future directions. Hydrobiologia in natural communities and their role in community stability 685, 1–17. and organisation. Am. Nat. 111, 1119–44. Kraus F. (2009) Alien Reptiles and Amphibians: A Scientific Com- Darwin C. (1859) On the Origin of Species by Means of Natural pendium and Analysis. Springer, Dordrecht. Selection. Murray, London. Kupferberg S. J. (1997) Bullfrog (Rana catesbeiana) invasion of a Daszak P., Strieby A., Cunningham A. A. et al. (2004) California river: the role of larval competition. Ecology 78, Experimental evidence that the bullfrog (Rana 1736–51. catesbeiana) is a potential carrier of chytridiomycosis, an Kushlan J. A. (1976) Environmental stability and fish commu- emerging fungal disease of amphibians. Herpetol. J. 14, nity diversity. Ecology 57, 821–5. 201–7. Legendre P. & Legendre L. (1998) Numerical Ecology, 2nd edn. D’Amore A., Hemingway V. & Wasson K. (2010) Do a threat- Elsevier, Amsterdam, Boston, London, New York, Oxford, ened native amphibian and its invasive congener differ in Paris, San Diego, San Francisco, Singapore, Sydney,Toquio. response to human alteration of the landscape? Biol. Inva- Lever C. (2003) Naturalized Reptiles and Amphibians of theWorld. sions 12, 145–54. Oxford University Press, Oxford. Declerck S. A. J., Coronel J. S., Legendre P. et al. (2011) Scale Li Y., Ke Z., Wang Y. & Blackburn T. M. (2011) Frog commu- dependency of processes structuring metacommunities of nity responses to recent American bullfrog invasions. Can. J. cladocerans in temporary pools of High-Andes wetlands. Zool. 57, 83–92. Ecography 34, 296–305. Lima S. L. & Agostinho C. A. (1988) A Criação de Rãs, 2nd edn. Dodds W. & Whiles M. (2010) Freshwater Ecology: Concepts and Editora Globo, São Paulo. Environmental Applications of Limnology, 2nd edn. Academic Liu X. & Li Y. (2009) Aquaculture enclosures relate to the Press, San Diego. establishment of feral populations of introduced species. Dray S., Legendre P. & Peres-Neto P. R. (2006) Spatial model- PLoS ONE 4, e6199. ling: a comprehensive framework for principal coordinate Loreau M. & Mouquet N. (1999) Immigration and the mainte- analysis of neighbour matrices (PCNM). Ecol. Model. 196, nance of local species diversity. Am. Nat. 154, 427–40. 483–93. Loyola R. D., Nabout J. C., Trindade-Filho J. et al. (2012) Elton C. (1958) The ecology of Invasions by Animals and Plants, Climate change might drive species into reserves: a case 2000th edn. The University of Chicago Press, Chicago, study of the American bullfrog in the Atlantic Forest Biodi- London. versity Hotspot. Alytes 29, 61–74. Eterovick P. C. & Fernandes G. W. (2001) Tadpole distribution Lucas E. M. & Fortes V. B. (2008) Frog diversity in the Floresta within montane meadow streams at the Serra do Cipó, Nacional de Chapecó, Atlantic Forest of southern Brazil. southeastern Brazil: ecological or phylogenetic constraints? Biota Neotrop. 8, 51–61. J.Trop. Ecol. 17, 683–93. MacArthur R. (1970) Species packing and competitive equilib- Fridley J. D., Brown R. L. & Bruno J. F. (2004) Null models of rium for many species. Theor. Popul. Biol. 1, 1–11. exotic invasion and scale-dependent patterns of native and MacArthur R. H. & Wilson E. O. (1967) The Theory of Island exotic species richness. Ecology 85, 3215–22. Biogeography. Princeton University Press, Princeton.

MacArthur R. & Levins R. (1967) The limiting similarity, con- Shaffer H. B., Alford R. A., Woodward B. D. et al. (1994) Quan- vergence, and divergence of coexisting species. Am. Nat. titative sampling of amphibian larvae. In: Measuring and 101, 377–85. Monitoring Biological Diversity, Standard Methods for Amphib- Nori J., Urbina-Cardona J. N., Loyola R. D., Lescano J. N. & ians (eds W.R. Heyer, M. A. Donnelly, R.W. McDiarmid, L. Leynaud G. C. (2011) Climate change and American bull- A. C. Hayek & M. S. Foster) pp. 130–41. Smithsonian frog invasion: what could we expect in South America? PLoS Institution Press, Washington, DC. ONE 6, e25718. Shipley B. (2000) Cause and Correlation in Biology:A User’s Guide Pillar V. D. (2006) Multiv: multivariate exploratory analysis, To path Analysis, Structural Equations and Causal Inference. randomization testing and bootstrap resampling. [Cited 22 Cambridge University Press, Cambridge. Nov 2011.] Available from URL: http://ecoqua.ecologia. Simberloff D. (2011) Charles Elton: neither founder nor siren, ufrgs.br/arquivos/Software/MULTIV/ but prophet. In: FiftyYears of Invasion Ecology: The legacy of Pillar V. P. (1999) How sharp are classiﬁcations? Ecology 80, Charles Elton (ed. D. M. Richardson) pp. 11–24. Wiley- 2508. Blackwell, Oxford. Pimentel D., McNair S., Janecka J. et al. (2001) Economic and Soulé M. E. (1990) The onslaught of alien species, and other environmental threats of alien plant, animal, and microbe challenges in the coming decades. Conserv. Biol. 4, 233–40. invasions. Agric. Ecosyst. Environ. 84, 1–20. Stohlgren T. J., Jarnevich C., Chong G. W. & Evangelista P. H. Poessel S. A., Beard K. H., Callahan C. M., Ferreira R. B. & (2006) Scale and plant invasions: a theory of biotic Stevenson E. T. (2013) Biotic acceptance in introduced acceptance. Preslia 78, 405–26. amphibians and reptiles in Europe and North America (ed Taylor B.W. & Irwin R. E. (2004) Linking economic activities to M. Olalla-Tárraga). Glob. Ecol. Biogeogr. 22, 192–201. the distribution of exotic plants. Proc. Natl Acad. Sci. USA R Development Core Team (2012) R: a language and environ- 101, 17725–30. ment for statistical computing. [Cited 15 March 2012.] Thomaz S. M. & Michelan T. S. (2011) Associations between a Available from URL: www.R-project.org highly invasive species and native macrophytes differ across Ricciardi A. & MacIsaac H. J. (2011) Impacts of biological spatial scales. Biol. Invasions 13, 1881–91. invasions on freshwater ecosystems. In: FiftyYears of Invasion Tilman D. (1997) Community invasibility, recruitment limita- Ecology:The Legacy of Charles Elton (ed. D. M. Richardson) tion, and grassland biodiversity. Ecology 78, 81–92. pp. 211–24. Wiley-Blackwell, Oxford. Von Holle B. & Simberloff D. (2005) Ecological resistance to Richardson D. M. (2011) Fifty Years of Invasion Ecology. Wiley- biological invasion overwhelmed by propagule pressure. Blackwell, Oxford. Ecology 86, 3212–18. Robinson G. R., Quinn J. F. & Stanton M. L. (1995) Invasibility Wang Y. & Li Y. (2009) Habitat selection by the introduced of experimental habitat islands in a California winter annual American bullfrog (Lithobates catesbeianus) on Daishan grassland. Ecology 76, 786–94. Island, China. J. Herpetol. 43, 205–11. Robinson J.V. & Dickerson J. E. (1984) Testing the invulnerabil- Wellborn G. A., Skelly D. K. & Werner E. E. (1996) Mechanisms ity of laboratory island communities to invasion. Oecologia creating community structure across a freshwater habitat 61, 169–74. gradient. Annu. Rev. Ecol. Syst. 27, 337–63. Santos T. G., Rossa-feres D. C. & Casatti L. (2007) Diversidade Witte F., Goldschmidt T., Wanink J. et al. (1992) The destruc- e distribuição espaço-temporal de anuros em região com tion of an endemic species ﬂock: quantitative data on the pronunciada estação seca no sudeste do Brasil. Iheringia Ser. decline of the haplochromine cichlids of Lake Victoria. Zool. 97, 37–49. Environ. Biol. Fishes 34, 1–28. Santos-Barrera G., Hammerson G., Hedges B. et al. (2009) Zampella R. A. & Bunnell J. F. (2000) The distribution of Lithobates catesbeianus. : IUCN Red List of Threatened anurans in two river systems of a coastal plain watershed. Species IUCN. [Cited 23 Nov 2011.] Available from URL: J. Herpetol. 34, 210–21. http://www.iucnredlist.org Schloegel L. M., Ferreira C. M., James T.Y. et al. (2010) The North American bullfrog as a reservoir for the spread of SUPPORTING INFORMATION Batrachochytrium dendrobatidis in Brazil. Anim. Conserv. 13, 53–61. Scott J. & Woodward B. D. (1994) Standard techniques for Additional Supporting Information may be found in inventory and monitoring: surveys at breeding sites. In: the online version of this article at the publisher’s Measuring and Monitoring Biological Diversity, Standard web-site: Methods for Amphibians (eds W. R. Heyer, M. A. Donnelly, R. W. McDiarmid, L. C. Hayek & M. S. Foster) pp. 92–6. Appendix S1. Geographic coordinates of the 90 Smithsonian Institution Press, Washington, DC. sampled waterbodies.