Introduction
Amphibians are important both to the environment and to people (Hocking & Babbitt, Reference Hocking and Babbitt2014), but many are threatened with extinction, predominantly because of global threats posed by anthropogenic activities. Amphibians are the most imperiled vertebrate class, with > 40% of species categorized as threatened on the IUCN Red List (Luedtke et al., Reference Luedtke, Chanson, Neam, Hobin, Maciel and Catenazzi2023). Of the many threats to amphibians, climate and land-use changes are amongst the most significant (Cushman, Reference Cushman2006; Lawler et al., Reference Lawler, Shafer, Bancroft and Blaustein2010; Souza et al., Reference Souza, Jardim, Rodrigues, Batista, Rangel and Gouveia2019; Luedtke et al., Reference Luedtke, Chanson, Neam, Hobin, Maciel and Catenazzi2023). Both can have detrimental impacts on amphibian habitats, resulting in negative effects on all amphibian life stages (Cushman, Reference Cushman2006; Bickford et al., Reference Bickford, Howard, Ng and Sheridan2010). For example, amphibian eggs and larvae are particularly vulnerable to desiccation (Carey & Alexander, Reference Carey and Alexander2003), which can be worsened by climate and land-use changes. The impacts of both threats overlap spatially (Hof et al., Reference Hof, Araújo, Jetz and Rahbek2011) and could exacerbate other pervasive and pressing global threats to amphibians (Manzoor et al., Reference Manzoor, Griffiths and Lukac2021), and thus the evaluation of their effects is vital for global amphibian conservation.
Although amphibians are distributed globally, species richness is greatest in the tropics (AmphibiaWeb, Reference AmphibiaWeb2024), where anthropogenic threats such as climate and land-use changes can have severe impacts (Becker & Zamudio, Reference Becker and Zamudio2011; Hof et al., Reference Hof, Araújo, Jetz and Rahbek2011). However, conservation assessments focused on amphibians have been more common in North America, Australia and Europe (Winter et al., Reference Winter, Fiedler, Hochachka, Koehncke, Meiri and De la Riva2016; Cordier et al., Reference Cordier, Aguilar, Lescano, Leynaud, Bonino and Miloch2021). Amphibians on tropical islands appear to be less well evaluated yet are likely to be more vulnerable to climate and land-use changes because most have restricted ranges or are poor dispersers (Beebee, Reference Beebee1996). Furthermore, many already face harsh environmental conditions that are likely to be worsened by increasing anthropogenic activities (Foden et al., Reference Foden, Butchart, Stuart, Vié, Akçakaya and Angulo2013; Pörtner et al., Reference Pörtner, Roberts, Adams, Adler, Aldunce and Ali2022).
Oceanic islands have never been connected to continents (Dawson, Reference Dawson2016). They vary considerably in size, age and topography and tend to have high endemism through isolation (MacArthur & Wilson, Reference MacArthur and Wilson1967), and thus are considered model systems for studying ecology and biodiversity (Graham et al., Reference Graham, Gruner, Lim and Gillespie2017). They are particularly vulnerable to climate and land-use changes that lead to droughts, temperature fluctuations, hurricanes and socio-economic constraints (Robinson, Reference Robinson2020; Pörtner et al., Reference Pörtner, Roberts, Adams, Adler, Aldunce and Ali2022). The high endemism, isolation and small size of many tropical oceanic islands make amphibian populations especially vulnerable to global threats (Ríos-López & Heatwole, Reference Ríos-López and Heatwole2023). Conservation efforts on tropical oceanic islands require particular attention because the capacity to effectively manage their biodiversity can be limited (Burt et al., Reference Burt, Nuno and Bunbury2022). However, there is little peer-reviewed literature on the adaptive capacity of species on tropical oceanic islands (Robinson, Reference Robinson2017), and information on the responses of tropical oceanic island amphibians to climate and land-use changes is sparse. Many amphibian species are endemic to tropical oceanic islands, and some are categorized as Data Deficient on the IUCN Red List and/or are found within a small area of occupancy (IUCN, 2023; AmphibiaWeb, Reference AmphibiaWeb2024), further highlighting the need for improved conservation management. For example, 94% of amphibian species in Cuba and 84% of those in Jamaica are endemic (AmphibiaWeb, Reference AmphibiaWeb2024). A better understanding of the responses of such amphibians to growing global anthropogenic threats could provide much-needed insights to help tropical oceanic islands fulfil the United Nations Sustainable Development Goals (e.g. Goal 15: Life on Land; United Nations, 2023) and preserve amphibian biodiversity.
Given the urgent need to assess the effects of climate and land-use changes on amphibians on tropical oceanic islands, we ask: (1) Is there less research into the effects of climate and land-use changes on amphibians on tropical oceanic islands compared to continental areas? (2) What are the taxonomic gaps in our knowledge of amphibians and their responses to these global anthropogenic threats? (3) Are there gaps in our knowledge of the life stages of amphibians and their responses to these threats? (4) Has there been an even distribution of research into climate or land-use factors impacting amphibians, and how do species and populations respond to these factors?
Methods
Author RJA systematically searched the peer-reviewed literature published during 1 July 1998–30 June 2022 using Web of Science (Clarivate Analytics, Philadelphia, USA) supplemented with Google Scholar (Google, Mountain View, USA) on 11 November 2022. The search protocol is outlined in Fig. 1. The search terms included a combination of the threats and either ‘caecilian’, ‘anuran’, ‘frog’ or ‘toad’, but not ‘salamander’, as salamanders do not naturally occur on tropical oceanic islands (AmphibiaWeb, Reference AmphibiaWeb2024). We grouped the locations of studies as North America (the USA, Canada and Mexico), Europe (including the UK), Central and South America (inclusive of continental islands), Asia (inclusive of continental islands), Australia and New Zealand, Africa (including Madagascar) and tropical islands that are true oceanic islands (between the Tropic of Cancer 23.5°N and Tropic of Capricorn 23.5°S), with amphibian orders as documented in AmphibiaWeb (Reference AmphibiaWeb2024). For all studies that took place on tropical oceanic islands, we noted the location of each author's affiliation. If an author was listed on more than one paper, we only counted them once. We generated data visualizations using the package ggplot2 (Wickham, Reference Wickham2016) in R 4.1.2 (R Core Team, 2021) to illustrate the geographical scope and author affiliations of the studies on tropical oceanic islands.
Fig. 1 Search protocol used for a systematic review of the literature on experimental studies examining the effects of climate and land-use changes on amphibians published during 1 July 1998–30 June 2022.
In addition to recording the study location, species and life stage, we noted the factors assessed and the biological responses to these factors for each tropical oceanic island-based study. The factors analysed in these studies were temperature, precipitation, relative humidity and salinity in relation to climate change, and habitat disturbance in relation to land-use change. The responses were range expansion or contraction, body size change, changes in calling activity (e.g. call duration), changes in species diversity, detection and occupancy, changes in abundance, and changes in biochemistry (e.g. concentration of defence chemicals produced). We generated a radar chart using the R package fmsb (Nakazawa, Reference Nakazawa2022) to illustrate the degree of evenness in factors and responses.
Results
From the literature search we found 327 articles assessing amphibian responses to climate change (Supplementary Material 1) and 451 articles assessing amphibian responses to land-use change (Supplementary Material 2). These articles only included amphibians of the order Anura. Eighteen of the 778 articles reported research on five oceanic islands (Table 1), of which 13 were focused on climate change and six on land-use change (one study assessed both). North American anurans were represented by five times more studies (Fig. 2). The distribution of the scientists conducting the studies on oceanic islands was also skewed; c. 20% were affiliated with an institution on the island where the study took place, and nearly 80% with institutions elsewhere, mostly in the USA (Fig. 3). The 18 studies assessed 21 species of six families (Bufonidae, Eleutherodactylidae, Hylidae, Hyperoliidae, Leptodactylidae and Sooglossidae; Table 2). Two of 13 climate change studies of oceanic island amphibians comprised assessments of tadpoles from two families (Bufonidae and Leptodactylidae), with the remainder assessing adults or post-metamorphic individuals only (Table 1). All land-use change studies on tropical oceanic islands evaluated adults only. The influence of temperature was the most common factor assessed, followed by precipitation, habitat disturbance, humidity and salinity (Fig. 4). Range expansion or contraction was the most commonly assessed response to change, followed by changes in calling activity, body size, abundance, detection and occupancy, species diversity and biochemistry (Fig. 4).
Table 1 Studies published during 1 July 1998–30 June 2022 assessing the responses of amphibians on oceanic islands to climate and land-use changes.
1 LC, Least Concern; EN, Endangered; CR, Critically Endangered.
Fig. 2 Number of experimental studies evaluating amphibian responses to (a) climate change and (b) land-use change published during 1 July 1998–30 June 2022, illustrating there have been few studies on tropical oceanic islands compared to most continental areas.
Fig. 3 Locations of the affiliations of authors who published research evaluating amphibian vulnerability to climate and land-use changes on tropical oceanic islands during 1 July 1998–30 June 2022. Note there were no authors from Cuba or The Bahamas in the studies in these countries (Table 1).
Table 2 Taxonomic representation of research on oceanic islands, highlighting the limited number of amphibian families and species evaluated. Islands for which no species were evaluated are not listed.
1 All species assessed were native except for one in Puerto Rico (Rhinella marina).
2 From AmphibiaWeb (Reference AmphibiaWeb2024).
3 From Ríos-López & Heatwole (Reference Ríos-López and Heatwole2023; Bahamas, Cuba and Puerto Rico) and AmphibiaWeb (Reference AmphibiaWeb2024; São Tomé Island and Seychelles).
Fig. 4 Uneven research effort into (a) factors and (b) responses evaluated in climate and land-use change studies of tropical oceanic island amphibians published during 1 July 1998–30 June 2022. Each contour represents a single study for factors and responses.
Discussion
Our review revealed that studies of tropical oceanic islands formed only c. 2% of research on amphibian responses to the global threats of climate and land-use changes published during 1998–2022. Although the small sample size limits our ability to analyse these studies quantitatively, it is clear there are notable research gaps in terms of geographical scope, taxonomic representation and life stage. There is also unevenness in the factors and responses evaluated, with few studies focusing on biochemical responses or salinity as a factor. These gaps represent limitations to global amphibian conservation.
Geographical scope
For both climate and land-use changes, more studies emerged from Central and South America during 2010–2020 compared to the previous decade (Fig. 2). This region hosts the highest species richness of amphibians (AmphibiaWeb, Reference AmphibiaWeb2024). However, in Africa, which is also rich in amphibian species, these global threats remain poorly evaluated (Fig. 2), further highlighting spatial gaps hindering amphibian conservation efforts.
Globally, land-use change studies outnumbered climate change studies before 2010, but during 2016–2020 this trend reversed, with twice as many climate change studies published as those on land-use change (Fig. 2). Nonetheless, studies of amphibians on tropical oceanic islands remain limited. This is of concern given that these populations are vulnerable to the impacts of both climate and land-use changes (Wanger et al., Reference Wanger, Iskandar, Motzke, Brook, Sodhi and Clough2010; Courchamp et al., Reference Courchamp, Hoffmann, Russell, Leclerc and Bellard2014). Gathering more information on island amphibians is particularly important as current limitations to amphibian conservation on oceanic islands include a lack of knowledge of adaptation in these specific habitats (Robinson, Reference Robinson2020), a low number of local researchers, limited ecological data on the effects of global threats on amphibian survivorship (IUCN, 2023), and weak or limited policies specifically focusing on amphibians (Powell & Henderson, Reference Powell, Henderson, Ríos-López and Heatwole2023).
Geographical biases are common in scientific research (Skopec et al., Reference Skopec, Issa, Reed and Harris2020). For example, Blaustein et al. (Reference Blaustein, Urbina, Snyder, Reynolds, Dang and Hoverman2018) found geographical biases in research carried out on infectious diseases in amphibians, with most studies being in North America. This bias has also been noted in research on pollution, another global threat to amphibians (Schiesari et al., Reference Schiesari, Grillitsch and Grillitsch2007). These biases probably occur because larger, well-funded institutions such as those in North America employ more researchers and often focus on common, widely distributed species (Schiesari et al., Reference Schiesari, Grillitsch and Grillitsch2007). The use of common species as model systems is an established research tool in ecological assessments (Scheiner & Gurevitch, Reference Scheiner and Gurevitch2001). In contrast, many oceanic island amphibian species are endemic, uncommon or not widely distributed because of their microhabitat preferences (Ríos-López & Heatwole, Reference Ríos-López and Heatwole2023).
The remoteness of tropical oceanic islands could be another reason for the low number of studies assessing amphibian responses to climate and land-use changes. This may be a particular issue in the Pacific (for which we found no studies) but perhaps less so in the Caribbean (where most oceanic island-based studies occurred). Limited resources impeding local experts on developing oceanic islands could be another factor (Burt et al., Reference Burt, Nuno and Bunbury2022). Of the 21 amphibian species evaluated for their responses to climate or land-use changes on tropical oceanic islands, most were on Puerto Rico (Table 1), which is geographically and politically close to the USA. Historically there has been a disproportionate amount of ecology and conservation biology research carried out in Puerto Rico compared to other Caribbean islands (Vallès et al., Reference Vallès, Labaude, Bezault, Browne, Deacon and Guppy2021). The sixth Intergovernmental Panel on Climate Change report warned that small islands are likely to face significant effects of climate change, and these effects are likely to be compounded by increasing land-use change (Pörtner et al., Reference Pörtner, Roberts, Adams, Adler, Aldunce and Ali2022). We located studies conducted in only five tropical oceanic island countries. There are many other islands where, to our knowledge, there have been no published studies evaluating amphibian vulnerability to climate and land-use changes, despite the presence of native amphibians. For example, we found no studies from Dominica, Grenada, Jamaica or Tobago in the Caribbean, the archipelago of Mayotte in the Indian Ocean, or Fiji or the Solomon Islands in Oceania, yet each hosts endemic and threatened amphibians (AmphibiaWeb, Reference AmphibiaWeb2024). It is therefore imperative to develop research on amphibians on tropical oceanic islands, expanding on the work of Bickford et al. (Reference Bickford, Howard, Ng and Sheridan2010) in Southeast Asia and Oliver et al. (Reference Oliver, Bower, McDonald, Kraus, Luedtke and Neam2022) in Melanesia, to ensure there is adequate coverage of amphibians when it comes to understanding the impacts of global threats on oceanic islands.
Taxonomic representation
Taxonomic biases in amphibian research have been documented previously, particularly in countries with high scientific capacity (da Silva et al., Reference da Silva, Malhado, Correia, Ladle, Vital and Mott2020). It is more difficult to assess taxonomic biases for countries with sparse research output. We acknowledge that some countries are larger and have more amphibian species than some oceanic islands; for example, the USA, with a total area of 9.8 million km2, hosts 345 amphibian species, compared to the Seychelles, which has a total area of 455 km2 and hosts 14 amphibian species (AmphibiaWeb, Reference AmphibiaWeb2024). However, it is clear there is inadequate taxonomic representation of amphibians assessed on oceanic islands. For example, Cuba hosts 67 endemic amphibian species, yet the effects of climate and land-use changes appear to have been evaluated for only one species during 1998–2022 (Cobos & Alonso Bosch, Reference Cobos and Alonso Bosch2018). This limits the extent of our understanding of how amphibians of families with different life history traits are likely to be affected by these threats. There have been attempts to identify vulnerable amphibians based on a sample of species (Foden et al., Reference Foden, Butchart, Stuart, Vié, Akçakaya and Angulo2013; Pottier et al., Reference Pottier, Lin, Oh, Pollo, Rivera-Villanueva and Valdebenito2022), but these global assessments also have limited data on amphibians on oceanic islands.
Life stage assessment
The unevenness of knowledge on amphibian adults and larval stages globally has been reported (Vera Candioti et al., Reference Vera Candioti, Baldo, Grosjean, Pereyra and Nori2023). In our review only two of the 18 studies we located evaluated tadpoles. The lack of studies evaluating life stages other than post-metamorphic adults is a major gap for tropical oceanic islands. To put this into perspective, in the USA approximately three quarters of studies on global threats to amphibians evaluated the responses of multiple life stages to climate change (Supplementary Material 1). This disparity is of concern as the tadpoles of many anuran species on tropical oceanic islands require water bodies for survival (AmphibiaWeb, Reference AmphibiaWeb2024). With higher temperatures, higher rates of desiccation and sea-level rise projected as a result of climate change (Pörtner et al., Reference Pörtner, Roberts, Adams, Adler, Aldunce and Ali2022), and increasing changes in land use, amphibian tadpoles on tropical oceanic islands face increasing pressures. Studies on temperate anurans highlight the negative impacts tadpoles can face, such as reduced ranges (Perotti et al., Reference Perotti, Bonino, Ferraro and Cruz2018) and changes in growth and development (Norlin et al., Reference Norlin, Byström, Karlsson, Johansson and Liess2016; Colomer et al., Reference Colomer, Margalida, Sanuy, Llorente, Sanuy and Pujol-Buxó2021). Although some amphibians appear to be able to tolerate, and to an extent even benefit from, changes in the environment (Brüning et al., Reference Brüning, Krieger, Meneses-Pelayo, Eisenhauer, Ramirez Pinilla and Reu2018), negative impacts are likely to predominate (Murray et al., Reference Murray, Nowakowski and Frishkoff2021). These global threats would not only affect amphibians with larval stages but also those that undergo direct development (e.g. species of Eleutherodactylidae), for which temperature and moisture (through evaporative water loss) play important roles in egg survival (Blaustein et al., Reference Blaustein, Urbina, Snyder, Reynolds, Dang and Hoverman2018). Scheffers et al. (Reference Scheffers, Brunner, Ramirez, Shoo, Diesmos and Williams2013) found evidence to suggest that direct-developing frog metamorphs and adults may be more sensitive to increasing temperatures than amphibians with larval development.
Factors and responses
Amongst the 21 species of tropical oceanic island amphibians assessed, there was uneven research across the factors and responses evaluated (Fig. 4). Temperature was the dominant factor assessed in most studies. Temperature affects all aspects of ectotherm biology and is therefore a key factor to assess, especially as climate change is projected to increase global temperatures (Pörtner et al., Reference Pörtner, Roberts, Adams, Adler, Aldunce and Ali2022) and is a significant driver of amphibian extinction (Sodhi et al., Reference Sodhi, Bickford, Diesmos, Lee, Koh and Brook2008). We located a limited number of studies evaluating the effects of salinity (Fig. 4), but this is a factor requiring attention given that sea level rise will have a significant impact on small tropical oceanic islands (Pörtner et al., Reference Pörtner, Roberts, Adams, Adler, Aldunce and Ali2022). Rising sea levels expanding into freshwater swamps could be detrimental to amphibians inhabiting this habitat as many amphibian species have low salt tolerance (Auguste et al., Reference Auguste, Downie, Jowers, Lehtinen, Mohammed, Murphy, Ríos-López and Heatwole2023). In our review, range expansion or contraction was the most common response evaluated (Fig. 4), which is unsurprising as it is the most commonly observed response to climate change (Parmesan, Reference Parmesan2006). The paucity of studies evaluating biochemical responses (Fig. 2) is of concern as amphibians produce chemicals in their skin that have implications not only for the animals themselves, but also for human health (Mechkarska et al., Reference Mechkarska, Coquet, Leprince, Auguste, Jouenne, Mangoni and Conlon2018).
Other considerations
This review focuses on the impacts of climate and land-use changes on tropical oceanic island amphibians, but invasive species are another major threat (IUCN, 2023) and islands are disproportionately more vulnerable to invasive species than continental areas (Russell et al., Reference Russell, Meyer, Holmes and Pagad2017). Not only have human activities facilitated the introduction of invasive species to islands (Powell & Henderson, Reference Powell, Henderson, Ríos-López and Heatwole2023), but land-use and climate changes can also increase the spread of invasive species (Manzoor et al., Reference Manzoor, Griffiths and Lukac2021). For example, the invasive frog Eleutherodactylus johnstonei appears to be restricted to urban areas upon establishment in a new island country (Downie et al., Reference Downie, Greener, Auguste and Geerah2017) but expansion of urban development could facilitate its spread. Climate change can also affect the ranges of invasive species (Hellmann et al., Reference Hellmann, Byers, Bierwagen and Dukes2008). Invasive species, climate and land-use changes and other major global threats such as disease, exploitation and pollution, both individually and in synergy, will affect amphibian communities globally, including on tropical oceanic islands.
Conclusions and recommendations
Global studies assessing amphibian responses to climate and land-use changes have provided valuable information for conservation (Winter et al., Reference Winter, Fiedler, Hochachka, Koehncke, Meiri and De la Riva2016; Cordier et al., Reference Cordier, Aguilar, Lescano, Leynaud, Bonino and Miloch2021) but our findings highlight the need for additional research on tropical oceanic islands. We make the following recommendations: (1) improve training opportunities for researchers based on islands, (2) improve the equitable distribution of resources, (3) improve inclusion of island-based researchers in local studies, and (4) promote conservation actions that can be implemented in the short to medium term.
Training opportunities
One of the factors contributing to the low number of studies on tropical oceanic islands is the paucity of resources and training opportunities for amphibian conservationists. Authors from across the Neotropics have advocated for the removal of barriers such as limited funding, exclusion from international research leadership and poor dissemination of knowledge to biologists living and working in the Global South (Vallès et al., Reference Vallès, Labaude, Bezault, Browne, Deacon and Guppy2021; Soares et al., Reference Soares, Cockle, Inzuwnza, Ibarra, Miño and Zuluaga2023). Provision of training, funded by wealthier countries, to herpetologists, conservation practitioners and students would allow them to use their experience to build local conservation capacity for amphibian research (Wanger et al., Reference Wanger, Motzke, Saleh and Iskandar2011; Verdade et al., Reference Verdade, Valdujo, Carnaval, Schiesari, Toledo and Mott2012).
More equitable distribution of resources
Research on amphibian conservation has largely been published from the wealthier countries of the Northern Hemisphere. Initiatives in countries where research has been neglected are needed to address the geographical and taxonomic gaps in amphibian research. For example, in South Africa, conservation of amphibians has received less funding than conservation of other vertebrate groups (Tarrant et al., Reference Tarrant, Kruger and Du Preez2016). However, even in data-rich countries such as the USA there are disparities in the funding of amphibian conservation (Gratwicke et al., Reference Gratwicke, Lovejoy and Wildt2012), and this is of even greater concern for Small Island Developing States (Soares et al., Reference Soares, Cockle, Inzuwnza, Ibarra, Miño and Zuluaga2023). Improvements in local wildlife policies and political will are needed for amphibian conservation.
Greater inclusion of island-based researchers
Extensive work is required across tropical oceanic islands and other tropical areas, especially Africa (Fig. 2), to address geographical, taxonomic, life stage and ecological research gaps. However, this research must be conducted ethically: the dominance of so-called parachute science can be detrimental to tropical ecology (Ocampo-Ariza et al., Reference Ocampo-Ariza, Toledo-Hernández, Librán-Embid, Armenteras, Vansynghel and Raveloaritiana2023). Our review shows that the few studies that have been conducted on tropical oceanic islands were predominantly authored by researchers based elsewhere (Fig. 3).
Short- to medium-term efforts
Conservation interventions in the short to medium term will be important for improving amphibian conservation on islands, especially where local expertise is currently limited. Examples of actions that could be applied by tropical oceanic island governments are detailed in the Amphibian Conservation Action Plan (Wren et al., Reference Wren, Borzee, Marcec-Greaves and Angulo2024) and the Brazilian Amphibian Conservation Action Plan (Verdade et al., Reference Verdade, Valdujo, Carnaval, Schiesari, Toledo and Mott2012). Ways in which authorities can protect amphibians from the effects of climate and land-use changes include improving public awareness of these threats (e.g. via schools and through public outreach events) and protecting key habitats. Other potential actions include captive breeding, ex situ conservation and biobanking (Gascon et al., 2007; Verdade et al., Reference Verdade, Valdujo, Carnaval, Schiesari, Toledo and Mott2012). These measures may be vital for protecting threatened species that are difficult to study in situ, and will require collaboration between local wildlife authorities and external organizations that already have experience with the appropriate techniques.
Author contributions
Study design, literature compilation, data visualization: RJA; writing: all authors.
Acknowledgements
RJA was supported by The University of the West Indies Postgraduate Scholarship. We thank two anonymous reviewers for their constructive comments.
Conflicts of interest
None.
Ethical standards
No specific approval was required for this review. Our research abided by the Oryx guidelines on ethical standards.
Data availability
All references used for this review are in the Supplementary Material.
Open access
