Introduction

The Anthropocene is characterized by high levels of biodiversity loss1. At the UN Biodiversity Conference (COP15) held in December 2022 in Montreal almost 200 countries committed to reducing the extinction rate tenfold by 20502. Extreme declines have been noted in amphibians, which are among the most threatened vertebrates3,4,5,6,7. While populations are experiencing severe declines worldwide, strategies and tools have been identified to prevent further amphibian loss8. Extinction risk is unequally distributed across amphibians7. Some genera stand out as highly threatened, particularly those that contain many species with restricted ranges and high rates of population declines4. One of these ‘worst-cases’ are the Neotropical harlequin toads, genus Atelopus, which scientists have closely monitored since the early 1990s and of which several species were expected to be extinct4,6,9. In recent years, more than 30 Atelopus species have been rediscovered10. This raises hope and given the extreme declines in harlequin toads, they might be appropriate to explore where we stand in managing the amphibian crisis.

We analyse trends in population status of 131 harlequin toads from 2004 to 2022 and explore current and future threats and their mitigation in this highly imperiled genus. We find that neither population status of the species nor the threats have changed. Using harlequin toads as a ‘worst-case’ amphibian decline model, we conclude that the amphibian extinction crisis is still an emergency. Despite invaluable efforts in amphibian conservation3,8, we so far have not been able to reverse the massive declines. Implementation of mitigation strategies needs more attention.

Background

The global amphibian extinction crisis

In the late 1980s, researchers started to witness amphibian declines at alarming rates with sudden and rapid population crashes all over the globe11,12. The 2004 IUCN Global Amphibian Assessment revealed that about 32% of 5743 species were threatened with extinction and 34 species were already extinct5. The most up-to-date IUCN Red List of Threatened Species (version 2022-2) now indicates that almost 35% of 7486 evaluated species are threatened with extinction and 38 are extinct7. In addition to ‘traditional’ threats (e.g., habitat destruction or degradation), novel and synergistically acting threats have been identified including climate change and emerging infectious diseases5,7,13,14. Of particular importance are pathogenic skin fungi from the genus Batrachochytrium causing the infectious disease chytridiomycosis. Human-mediated spread of these fungi has resulted in mass mortality events in many amphibians worldwide6,15.

The IUCN Amphibian Conservation Action Plan, regularly updated since 2007, frames strategies and tools to escape the amphibian emergency. Among others, this includes research at all levels from ecology to systematics, evolutionary biology, increased monitoring and continuous status assessments, habitat protection, disease mitigation efforts, in-country capacity building, community-based work, and ex situ conservation breeding8.

While extinction threatens many amphibians, new species are continuously being discovered, with more than 150 species descriptions per year over the last two decades16. This makes assessing the amphibian crisis difficult as the true number of living amphibians is hard to estimate, and for this reason, it is challenging to assess our progress in overcoming the amphibian extinction crisis.

Dramatic declines of harlequin toads

More than 100 species of harlequin toads occur in tropical lowland and montane forests up to the paramos, from sea level to almost snowline in Central and South America (Supplementary Table 1). Scientists have closely monitored them since the early 1990s documenting the alarming situation of most of them to call for conservation action9,17,18,19,20. At least 27 species declined rapidly from 1984 to 2004 (Supplementary Table 2). By 2005, more than half of all species had not been seen for one decade or longer, and only 10 species had stable populations9. The latest IUCN Red List assessed 94 harlequin toads and listed two thirds (62) of these as Critically Endangered (Fig. 1), of which 39 are Possibly Extinct. Only one species is categorized as Least Concern7.

Fig. 1: Harlequin toads are a prime example of a worst-case scenario of amphibian declines, with most species threatened with extinction.
figure 1

Of 131 species, 94 have been assessed by the IUCN Red List:7 NE – Not Evaluated (37), DD – Data Deficient (9), LC – Least Concern (1), NT – Near Threatened (2), VU – Vulnerable (3), EN – Endangered (14), CR – Critically Endangered (62), EX – Extinct (3). The black line indicates that 39 CR species are possibly extinct. Representatives of different categories are shown, clockwise from above: Atelopus vogli, only known in preservative (EX), A. sp. “wampukrum” (NE), A. hoogmoedi (NE), A. spurrelli (NT), A. carrikeri (EN), A. ignescens (CR), A. longirostris (CR), and A. zeteki (CR), which is possibly extinct in the wild (photos: Christopher Heine (A. vogli), Jaime Culebras).

Full size image

While these numbers clearly demonstrate why Atelopus represent a ‘worst-case’ amphibian decline scenario, the fear of witnessing the unprecedented extinction of an entire species-rich amphibian genus21 has not yet materialized. Since 2005, only three species were added to the list of harlequin toads in decline (Supplementary Table 2). This is remarkable, given that Atelopus are among the best-studied and most sought-after amphibians in the Neotropics10,20,22. Instead, 30 species that had not been seen by scientists for many years have been rediscovered (Supplementary Table 3), suggesting a hypothetical reverse population trend in at least some species10. This also includes two species the IUCN Red List categorized as Extinct (Atelopus ignescens and A. longirostris; Fig. 1)23.

We define two phases: (i) the crisis with dramatic population declines, ca. 1984–2004; and (ii) post-2004, a hypothetical improvement of the population status in at least some species. Two features provide unique opportunities to explore this apparent change. This is different to many other ‘worst-case’ amphibian groups (e.g. Centrolene, Gastrotheca, Pseudophryne, Strabomantis, Telmatobius), why we consider Atelopus a potential model for them. First, population monitoring data from two periods led to the assessment of population status of all known species in 20049 and 2022. These are based on expert knowledge, which is common practice in conservation assessments24. As done in 2004, we assigned one of three population status categories to each species in 2022, resulting in two sets of standardized data 15 years apart (Fig. 2; Supplementary Table 1). The second feature making harlequin toads an exemplary group is that we take a high degree of accuracy in taxonomy for granted, which is essential for effective species conservation25. Several of the authors have contributed to the taxonomy of Atelopus for many years and have established the contemporary taxonomic architecture of the genus. With this unparalleled advantage, we have a comprehensive data set on this genus far beyond the available published information, so that species status data are not markedly behind the taxonomic progress, as in other amphibians. In total, we evaluated 131 Atelopus species, of which 37 (28.2%) are not yet assessed on the IUCN Red List (Fig. 1; Supplementary Table 1). Next to acknowledging the taxonomic progress (i.e., new and revalidated species), our 2022 database includes 31 species identified but not yet formally described (e.g. A. sp. “wampukrum”, Fig. 1).

Fig. 2: Harlequin toad population status 2004–2022.
figure 2

Change in population status of 131 Atelopus species over time, based on two standardized datasets with a span of more than 15 years between them. Not a single species has improved its population status. For details see Supplementary Table 1.

Full size image

Results and discussion

We did not detect change in the population status of the species assessed in 2004 and 2022 (Fig. 2), (Freeman-Halton extension of Fisher’s exact test for RxC table: p = 0.152; N = 131), including for a 2022 reduced dataset containing only the species listed in the 2004 assessment (p = 0.826; N = 94). Since 2004, no species have had an improvement in population status from shrinking to steady (Fig. 2), even the shrinking species that were rediscovered after 2004 remain shrinking. This is supported by the observation that specimens occur in densities apparently lower than in the past, so that it is hard to consider them ‘true’ recoveries with the demographic potential to contribute to population expansions. Most of the findings were new discoveries at formerly unknown localities10. Most historical localities, which have been regularly visited, have remained empty of harlequin toads for many years.

Harlequin toads with shrinking population status mostly occur in Andean areas and Central America, while none of the species from the lower Amazon basin, the Guianas, the Colombian Sierra Nevada de Santa Marta and part of the species from the Chocó have shrinking populations (13 species in total) (Fig. 3a). Alarmingly, 61 species (46.6%) have ‘year last seen’ (YLS) in 2004 or earlier. These species have population status data pending or shrinking (depending on the knowledge on the population status prior their YLS) and 37 have probably vanished (Supplementary Table 1).

Fig. 3: Threats to harlequin toads.
figure 3

a Climatic Stability Index (CSI, Shared Socioeconomic Pathway 8.5 scenario35) map of South America showing the distribution of harlequin toads and their current population status. b Number of species affected by habitat destruction and degradation or by the presence of the skin fungus Batrachochytrium dendrobatidis (Bd) or both arranged by population status. c Distribution of CSI values among species according to their population status. In b, c the number of species is given per population status. For details see Supplementary Table 1.

Full size image

Persisting threats

The main drivers of Atelopus declines are habitat destruction and degradation5,7 and the skin fungus Batrachochytrium dendrobatidis (Bd)6,26. The 2022 expert data demonstrate that habitat destruction and degradation are a threat to 93 (71.0%) species (Fig. 3b), of which 37 have YLS 2004 or earlier (Supplementary Table 1). We noted the presence of Bd in 50 (38.2%) species, and of the 42 species with Bd whose population status is categorized as shrinking or data pending (Fig. 3b), 16 have YLS 2004 or earlier (Supplementary Table 1).

Habitat destruction and degradation as well as Bd continue to threaten harlequin toads today. The situation is dire, considering how habitat change can dramatically affect Atelopus populations27. A vivid example is the recently rediscovered A. longirostris (Fig. 1) from the Intag Valley of Ecuador which is now the focus of a large copper mining project28. Regarding Bd, some populations compensate high mortality rates with a high reproductive output, which might cause life history shifts, e.g. in A. cruciger. That is, some Atelopus species can persist once the fungus becomes enzootic, using different strategies29,30,31,32. There are five species that continue to have steady populations despite the presence of Bd (see Supplementary Table 1) However, these are fragile systems, because effects of Bd are more complex due to co-stressors6.

Climate change as a future threat

Some studies suggest that unusual climatic conditions may play a role in Atelopus declines and might also exacerbate the effects of Bd26,33,34. Climatic change is associated with an increase in the frequency of extreme weather events. Climate change for the period from present (1970–2000) to the year 2100 can be assessed using the Climatic Stability Index, CSI35. CSI values are higher in lowlands (i.e. climate is unstable), where most harlequin toads with steady populations occur (Fig. 3a), suggesting that climate change might become an important threat to currently steady lowland Atelopus in the future (Fig. 3c). However, this does not preclude species living at high altitudes from the effects of climate change. Especially high-Andean taxa are potentially limited in their capacities to undergo range shifts as a response to warming. They may also become more exposed to an increased solar radiation.

The number of Atelopus species in protected areas has increased from 84 to 96 from 2004 to 2022 (64.1 to 73.3%; Supplementary Table 1). Given that the populations of 43 of these species are shrinking, it is evident that habitat protection alone is not sufficient. So far, multi-disciplinary approaches to improve the conservation status of Atelopus have been proposed, next to habitat protection, additionally including scientific research, monitoring, environmental education, and conservation breeding19. Recently, these were reinforced in the Atelopus Conservation Action Plan 2021–2041 along with elaborated targets for the coming decades. The plan was launched by the recently founded Atelopus Survival Initiative36, a multi-stakeholder collaborative and participative effort to stop the loss of this imperiled genus20.

Other initiatives, such as the El Valle Amphibian Conservation Center37, Project Golden Frog38 and the Panama Amphibian Rescue and Conservation Project39, have operated for more than 10 years now and have contributed to direct conservation efforts on Central American Atelopus species. They have promoted research to further understand the effects of Bd on population genetic structures40 and to explore the response of Bd to an altered skin microbiome15,41 or to skin secretions32,42. In addition, ex situ assisted reproductive technologies are being developed to preserve genetic diversity of some harlequin toads, for instance for Ecuadorian species43. Various local initiatives have been initiated recently to conserve particular species through community-based conservation action and habitat protection44,45, but their impact remains difficult to assess.

The special role of ex situ conservation breeding

For highly threatened species, captive assurance populations are strongly recommended as a potentially powerful tool allowing for reintroductions once the threat(s) can be mitigated. This is an extreme measure, especially for amphibian species that are vulnerable to Bd. In practice, ex situ conservation breeding allows us to buy time while appropriate mitigation and conservation strategies are developed and applied8. Thus, conservation breeding is a crucial action to enhance the chances of harlequin toad survival19,35,46. So far, captive assurance colonies have been established for 26 Atelopus species and scientists have now started to study reintroductions of captive-born individuals in A. limosus;47 studies on additional species are in progress (authors’ unpubl. data). So far, 16 species (13 with shrinking populations) have already successfully reproduced at least once at one institution, with offspring reaching maturity. However, only six Atelopus species have reproduced at two or more institutions with at least one institution successfully reproducing for more than one generation (Supplementary Table 4), showing that conservation breeding needs more efforts. Conservation breeding of A. zeteki (Fig. 1) has been a story of remarkable success38,46, with more than a thousand individuals being held in more than 50 institutions in three countries, independent from wild caught founders, for roughly 20 years. This species is considered shrinking and has likely already vanished in the wild. As a sharp contrast, in three Atelopus species, conservation breeding attempts have failed, with all individuals lost (Supplementary Table 4).

Captive breeding initiatives are regionally biased, with several in-country colonies covering mainly Ecuadorian and Panamanian species, while Brazil, Colombia and Venezuela have recently launched programs for one species each. Ex situ conservation efforts are lacking in other countries where harlequin toads occur, yet some species are or were bred in Europe and North America (Supplementary Table 4).

Conclusion

Our comparison of the 2004 and 2022 databases reveals that no harlequin toad went from status shrinking to steady. We reject the hypothesis that we have entered a hypothetical post-decline phase of improvement. Although reports on sudden and rapid population declines have lessened (Supplementary Table 2), many species continue to have shrinking populations, which implies that over time declines continue to happen more silently. This is reinforced by the observations that of the 29 species rediscovered since 2004, six have not been seen again for the past 10 years, despite targeted searches10. These harlequin toads either occur in low numbers or are possibly extinct emphasizing that rediscoveries are not equating recoveries. At the same time habitat destruction and Bd persist to threaten harlequin toads, despite that 96 (75%) of the species occur in protected areas and (at least) some Atelopus can somehow cope with Bd. In addition, future threats through changing climate are expected.

Using harlequin toads as a worst-case model for amphibian species with high extinction risk4,7, the situation of these amphibians has not improved over the past two decades. Despite invaluable efforts, we so far have not been able to reverse the massive declines. The amphibian extinction crisis is still an emergency.

The goal to massively reduce extinction rates within the next 20–30 years2 will require immense efforts and investments, which need to be massively scaled up. More in situ mitigation strategies need to be implemented, including effective habitat protection and management. This also addresses the development of innovative methods to mitigate the effects of infectious diseases (such as Bd). The expansion of ex situ capacities might address to strengthen the technical and scientific capacities for in country-captive breeding. Increased storage capacities for germplasm and genetic material as well as the implementation of genome resource banks are additional steps for safeguarding amphibian species with high extinction risk. In line with these strategies, programs aiming at reintroductions and translocations of species are a key step to mitigate the amphibian extinction crisis. Other strategies are to improve the baseline knowledge, to increase the visibility of the problem and to create mechanisms for multi-stakeholder collaboration and participation including local communities.

For harlequin toads, these aspects are described in detail in the Atelopus Conservation Action Plan 2021–204136. Given proper funding and support and the collaborative coalition of on-the-ground conservation organizations, zoos, governments and local communities, there is improvement in sight for the remaining harlequin toads20. For most other threatened amphibians such focal plans and coordinated conservation efforts are still lacking.

Methods

We used the original database of La Marca et al.9, which compiled information from experts on Atelopus population trends through February 2004. This database was updated with the revised taxonomy, as well as additional input and data from experts, and our study period was defined as March 2004 to December 2022 (Supplementary Table 1). We interviewed 105 experts (a diverse group of scientists and conservationists with considerable knowledge on Atelopus species), sometimes on multiple occasions, from October 2019 to December 2022. All 75 original database contributors to La Marca et al.9 were contacted, but we received updated information from only 13. The additional 92 contributing researchers were Atelopus taxonomists and conservationists. Original contributions included initial estimates on population status and general threats as well as field data on the species ecology and threats at the site level. Data was reviewed and summarized species-wise prior to a second round, where experts were able to provide corrected estimates, as has been proven successful for obtaining robust expert data48,49. Still, we are well-aware that expert opinions carry an unavoidable uncertainty and need to be considered carefully49,50. Most of the experts who contributed data also contributed to the writing of this paper and became co-authors.

In this study, we refer the term ‘species’ to described taxa that we consider valid and to forms that are undescribed and are expected to represent valid species. Also, we tentatively consider suggested subspecies as species here. According to this, we refer to 131 species of which 100 are formally described, while the 2004 database considered 94 taxa. Alphataxonomic problems remain for some harlequin toads, e.g. A. hoogmoedi which could be more than one species51.

Following La Marca et al[.9, for each species we updated the information on distribution, the known altitudinal range, and local endemism by number of known populations (1, 2, >2). We recorded: species population status; the year last seen (YLS); information on Bd presence; occurrence in any governmental or private protected area. In order to address threats that have been identified to play a key role in Atelopus declines9,34, we additionally assessed the potential future impact of climate change for each species. For detailed data see Supplementary Table 1. Details on how data were assessed are described below.

The population status was coded following definitions of La Marca et al.9. However, we change the terminology to avoid confusion with terms used by the IUCN Red List of Threatened Species. We used ‘steady’ ( = stable) when one or more populations are known to have persisted and no population has declined by more than 50%; ‘shrinking’ ( = declining) when at least one population has declined by more than 50%; ‘data pending’ ( = data deficient) when insufficient population trend data are available to judge whether a decline has occurred. This also includes species where unsystematic survey efforts were invested, but the species could not be detected. We additionally coded species as ‘probably vanished’ or ‘probably vanished in the wild’, when experts agreed that the particular species has possibly gone extinct (or extinct in the wild). However, as extinction is difficult to confirm, and as rediscoveries of formerly thought to be extinct species are common in the genus, this is not a separate status category. Freeman-Halton extension of Fisher’s exact tests (one-sided, with sequential Bonferroni-Holm correction) were performed to explore the population status change between the 2004 and 2022 databases using the following online web statistical calculator: https://astatsa.com/FisherTest/ (accessed 7 January 2023).

In addition to information on Bd presence through direct testing of Atelopus specimens (including diagnosis both by PCR or histology), we provide information on presumed Bd presence for Atelopus species that have not been tested for Bd. We consider likely presence of Bd in an Atelopus species when Bd infection is known in other amphibian species at Atelopus localities. For this purpose, we used Bd records available from the Aquatic Parasite Observatory (http://www.aquaticparasites.org/, accessed 8 June 2022) and the Amphibian Disease Portal (https://amphibiandisease.org/, accessed 28 September 2022). Employing ArcGIS Pro (ESRI), we then buffered (5 km) 776 georeferenced Atelopus records of 102 species. If a Bd record was within a buffer, we considered the Atelopus species to potentially have Bd.

To assess the potential future impact of climate change at Atelopus sites, we used the Climate Stability Index (CSI). It provides information on climate vulnerability for the time span Pliocene (3.3 Ma) to the year 2100 35. The CSI operates using the standard deviation over time at grid cells with resolution 2.5 arc-min. It is based on 9 general circulation models and 19 bioclimatic variables52 available from WorldClim 253. Of the different underlying climate change scenarios (Shared Socioeconomic Pathways, SSP), we chose, a priori, SSP5–8.5 from present (1970–2000) to the year 2100 for our study because it is among the most extreme (‘worst-case’) scenarios for Central and South America35. CSI data were downloaded from Figshare (https://doi.org/10.6084/m9.figshare.14672637, accessed 2 March 2022). CSI values range 0-1 (most to least stable).

With the goal of obtaining information on Atelopus conservation breeding, we contacted all existing institutions, to our knowledge, that hold or held captive assurance colonies of harlequin toads to compile a dataset on the species kept and breeding success (Supplementary Table 4). We received information from all seven institutions located in Atelopus range countries as well as from numerous North American and European institutions, partially coordinated through the multinational Project Golden Frog38.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.