Biol Invasions (2021) 23:69–81 https://doi.org/10.1007/s10530-020-02356-9 (0123456789().,-volV)( 0123456789().,-volV) ORIGINAL PAPER Non-native populations and global invasion potential of the Indian bullfrog Hoplobatrachus tigerinus: a synthesis for risk-analysis Nitya Prakash Mohanty . Angelica Crottini . Raquel A. Garcia . John Measey Received: 1 October 2019 / Accepted: 26 August 2020 / Published online: 9 September 2020 Ó Springer Nature Switzerland AG 2020 Abstract Invasive amphibians have considerable tigerinus as an invasive species to aid in risk analyses ecological and socio-economic impact. However, and management of existing populations. We review strong taxonomic biases in the existing literature the available knowledge on non-native populations of necessitate synthesizing knowledge on emerging H. tigerinus and model its potential distribution in the invaders. The Indian bullfrog, Hoplobatrachus tiger- non-native range and globally; finally, we evaluate its inus, a large dicroglossid frog (snout to vent length: up ecological and socio-economic impact using standard to 160 mm), is native to the Indian sub-continent. impact classification schemes. We confirm successful Despite the high likelihood of invasion success for H. invasions on the Andaman archipelago and Madagas- tigerinus, based on the species’ natural history traits car. The ensemble species distribution model, with and human use, the status of its non-native populations ‘good’ predictive ability and transferability, predicts and global invasion potential has not yet been tropical regions of the world to be climatically assessed. In this paper, we provide a profile of H. suitable for the species. Considering potential for propagule pressure, we predict the climatically suit- able Mascarene Islands, Malaysia and Indonesia, and Electronic supplementary material The online version of East Africa to likely be recipients of bridgehead this article (https://doi.org/10.1007/s10530-020-02356-9) con- invasions. We assign the species two impact scores: tains supplementary material, which is available to authorized both socio-economic and environmental scores were users. ‘moderate’ with ‘medium’ confidence levels in our N. P. Mohanty (&) Á R. A. Garcia Á J. Measey assessment. Finally, this synthesis outlines the inva- Centre for Invasion Biology, Department of Botany and sion process of the genus Hoplobatrachus, which is an Zoology, Stellenbosch University, Stellenbosch 7602, emerging group of amphibian invaders. South Africa e-mail: [email protected] Keywords Amphibian Á Madagascar Á Andaman N. P. Mohanty Nicobar Á Island invasive Á Risk-assessment Á Impact Andaman Nicobar Environment Team, scoring Á Environmental impact classification for Wandoor, South Andaman, Andaman and Nicobar Islands 744103, India Alien Taxa Á Socio-economic impact classification of Alien Taxa Á Pet trade A. Crottini CIBIO, Centro de Investigac¸a˜o em Biodiversidade e Recursos Gene´ticos, InBIO, Universidade do Porto, Campus Agra´rio de Vaira˜o, Rua Padre Armando Quintas, No 7, 4485-661 Vaira˜o, Portugal 123 70 N. P. Mohanty et al. Introduction tigerinus (see Novoa et al. 2020), the status of its non- native populations and its global invasion potential Invasive amphibians have a considerably high envi- have not yet been assessed. ronmental and socio-economic impact (Measey et al. Systematic literature reviews and species distribu- 2016; Kumschick et al. 2017; Bacher et al. 2018). tion modelling have been used to generate global Globally, 78 non-native species of amphibians are species profiles for several amphibian invaders (e.g. known to have at least one established or invasive Lithobates catesbeianus, Ficetola et al. 2007a, b; population (Capinha et al. 2017), whereas nearly 100 Xenopus laevis, Measey et al. 2012). Such assessments non-native amphibians could be considered to have can be particularly useful in understanding consistent established populations, with a level of uncertainty patterns of invasion dynamics of the species as well as (Kraus 2009; Measey et al. 2016). A recent review of its similarities to other invasions in terms of pathways, non-native occurrence of amphibians recorded 263 species traits, recipient ecosystems—a phenomenon species, including those traded and present in captivity referred to as ‘‘invasion syndromes’’ (Novoa et al. (van Wilgen et al. 2018), forecasting future invasions 2020). Risk assessments rely on published information of many more amphibian species. Introductions of from prior invasions, covering the species’ invasion amphibians have accelerated in recent decades potential based on species distribution modelling, (Seebens et al. 2017; Capinha et al. 2017). However, dispersal pathways, spread rates, potential impact and studies on amphibian invasions are heavily taxon- recommended management action (Kumschick et al. biased, with only three species (Rhinella marina, 2019). Potential impacts of species should ideally be Lithobates catesbeianus, and Xenopus laevis) respon- recorded using impact classification frameworks, sible for ca. 81% of all publications on non-native which are particularly useful to assign standardized amphibians (van Wilgen et al. 2018). A focus on scores to an invasive species. Based on dietary emerging and unassessed amphibian invaders is assessments of adult H. tigerinus on the Andaman therefore necessary. Islands, an Environmental Impact Classification of The Indian bullfrog, Hoplobatrachus tigerinus Alien Taxa (EICAT; Blackburn et al. 2014) score of (Daudin, 1802), a large dicroglossid frog (snout to ‘minor’ impact was assigned to the species by vent length: up to 160 mm), is native to the Indian sub- Mohanty and Measey (2018). Previously, a global continent (Dutta 1997). H. tigerinus is consumed evaluation by Kumschick et al. (2017) had resulted in regionally and was formerly a part of the international the same environmental impact score, whereas the ‘frog leg trade’ (Abdulali 1985; Oza 1990). Its large species was considered ‘data deficient’ under Socio- body size, association with human-modified land- Economic Impact Classification of Alien Taxa (SEI- scapes (e.g. paddy fields; Daniels, 2005), and use as a CAT; see supporting information in Bacher et al. food resource make H. tigerinus a likely candidate for 2018). However, a reassessment is deemed necessary human-mediated introduction outside its native range in the light of new studies on the ecological and socio- (Tingley et al. 2010; van Wilgen et al. 2018). Further, economic impact of the species. the species has a high establishment probability due to In this paper, we provide a global species profile of its fecundity (ca. 6000 eggs per clutch; Allen et al. the Indian bullfrog, Hoplobatrachus tigerinus,asan 2017) and ability to breed successfully in ephemeral invader, to aid in risk assessment and management of pools of human-modified habitats. Human consump- existing non-native populations. To this end, we aim to tion also makes the species likely to be moved locally (1) synthesize existing knowledge on invasion status, in the non-native range, thereby exacerbating its dispersal pathways and spread, impact, and manage- spread (Liu et al. 2014). The carnivorous tadpoles of ment action, (2) evaluate the global invasion potential H. tigerinus prey upon larvae of other anurans (Khan using species distribution models, and (3) update 1996; Grosjean et al. 2004), whereas post-metamor- standardized metrics of impact (EICAT and SEICAT) phic individuals consume a broad range of inverte- for H. tigerinus based on current knowledge of the brates and small vertebrates (Padhye et al. 2008), species’ environmental and socio-economic effects. making ecological impacts via predation a matter of concern in the non-native range. Despite the high likelihood of invasion success and impact for H. 123 Non-native populations and global invasion potential 71 Methods with the presence points, and did not find any strong patterns suggesting environmental sampling bias Non-native populations (Supplementary Information 1 Fig. 1). After removing duplicates (at a resolution of 2.5 arc min), a total of We searched for literature on non-native populations 153 presence points was retained from the native of Hoplobatrachus tigerinus on Google Scholar range. (September 2018) by employing a combination of Models can be trained with only native range data key words covering taxonomic variation (including when the environmental extent of the non-native range the previous taxonomic combinations ‘Rana tigerina’ is contained within that of the native range (Gallien and ‘Rana tigrina’) and invasion status (‘alien’, et al. 2010) and if native range data are reliable (Hattab ‘introduced’, ‘non-native’, ‘exotic’, ‘non-indigenous’ et al. 2017). We considered that these two require- and ‘invasive’). We also consulted local researchers ments were met in our case. We found the environ- and herpetologists, in both native and non-native mental space in the non-native range of H. tigerinus to ranges, for occurrence information where published be a subset of its native range (SI 1 Fig. 2). As H. literature was lacking (Ficetola et al. 2007a). Identified tigerinus is a commonly encountered frog with a wide non-native populations were evaluated for their inva- distribution in its native range (Daniels 2005), we sion stage (following Blackburn et al. 2011), dispersal considered the collated occurrence data to be of good pathways (of introduction and secondary transfers), quality. We also could not assume the non-native