Biological Conservation 277 (2023) 109845

Contents lists available at ScienceDirect

Biological Conservation
journal homepage: www.elsevier.com/locate/biocon

A retrospective overview of amphibian declines in Brazil's Atlantic Forest

Luís Felipe Toledo a, *, Sergio Potsch de Carvalho-e-Silva b,
Ana Maria Paulino Telles de Carvalho-e-Silva c, João Luiz Gasparini d, Délio Baêta e,
Raoni Rebouças a, Célio F.B. Haddad e, C. Guilherme Becker f, Tamilie Carvalho a, g
a
Laboratório de História Natural de Anfíbios Brasileiros (LaHNAB), Departamento de Biologia Animal, Instituto de Biologia, Universidade Estadual de Campinas,
13083-862 Campinas, SP, Brazil
b
Laboratório de Anfíbios e Répteis, Departamento de Zoologia, Instituto de Biologia, Universidade Federal do Rio de Janeiro, 21941-902 Rio de Janeiro, RJ, Brazil
c
Laboratório de Biossistemática de Anfíbios, Universidade Federal do Estado do Rio de Janeiro, 22290-240, Rio de Janeiro, RJ, Brazil
d
Pós-Graducação em Ciências Ambientais e Conservação, Instituto de Biodiversidade e Sustentabilidade, Universidade Federal do Rio de Janeiro, 27910-970, Macaé, RJ,
Brazil
e
Departamento de Biodiversidade e Centro de Aquicultura (CAUNESP), Universidade Estadual Paulista, 13506-900 Rio Claro, SP, Brazil
f
Department of Biology, The Pennsylvania State University, 16802, University Park, PA, USA
g
Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, MI 48109, USA

A R T I C L E I N F O A B S T R A C T

Keywords: Amphibians are facing population declines and extinctions across the globe, and megadiverse Brazil is no
6th mass extinction exception. One of the global hotspots of amphibian declines is Brazil's Atlantic Forest (AF), an ecoregion that in
Amphibian declines the past covered over 1,300,000 km2 of luxuriant coastal rainforests. Reports of historical declines were
Atlantic forest
concentrated in the southern and southeastern sections of AF, however we now propose that these reports did not
Conservation
Disease outbreaks
capture the whole extent and complexity of historical amphibian declines. Here, we conducted a refined review
Ghosts of the past of historical amphibian declines that have been linked to environmental factors, climatic anomalies, and the
Lazarus effect emergence of the amphibian chytrid fungus Batrachochytrium dendrobatidis (Bd). We reviewed data on 169
Population recovery populations of 106 species that have undergone population declines in the southern and southeastern Atlantic
Forest, more than doubling the number of population declines reported in previous studies. These numbers place
the Atlantic Forest among the ecoregions with the highest rates of amphibian declines and extinctions globally.
We provide novel spatiotemporal information of amphibian declines, underscoring that the peak of declines
happened in 1979, and that population recoveries, when they occurred, often took over 30 years. Our analyses
indicate that the volume of field collections accessioned in museums over time matched temporal patterns of
declines and extinctions, which indicates that historical declines might have impacted a far larger number of
amphibian populations and species. Our review helps guide targeted management programs for amphibian
surveys and Bd surveillance with practical implications for conservation.

1. Introduction (DAPTF) (Bishop et al., 2012). Population declines in several continents

have been initially linked to multiple environmental and anthropogenic
The so-called amphibian crisis is global and unprecedented (Scheele factors, such as habitat loss, environmental pollution, UV-B radiation,
et al., 2019). Although the first reports of amphibian declines date back introductions of alien species, and over-exploitation (Becker et al., 2007;
to the late 1950's (e.g., Conant, 1958), scientists only started to Collins and Storfer, 2003; Falaschi et al., 2020; Stuart et al., 2004).
acknowledge during the 1980's that declines were severe and wide­ In the late 1990s, the emergence of the amphibian chytrid fungus
spread across continents (Blaustein and Wake, 1990; Wake, 1991). The Batrachochytrium dendrobatidis (Bd) was identified as another potential
first reports of amphibian population declines mark the beginning of a cause of the observed amphibian crisis (Berger et al., 1998; Lips et al.,
global effort to study and manage population crashes, which led to the 2006; Longcore et al., 1999; Scheele et al., 2019), especially in pristine
establishment of the IUCN Declining Amphibian Populations Task Force tropical forests where other environmental stressors were less likely to

* Corresponding author.
E-mail address: [email protected] (L.F. Toledo).

https://doi.org/10.1016/j.biocon.2022.109845
Received 23 August 2022; Received in revised form 26 November 2022; Accepted 2 December 2022
Available online 14 December 2022
0006-3207/© 2022 Elsevier Ltd. All rights reserved.
L.F. Toledo et al. Biological Conservation 277 (2023) 109845

play an important role (Becker and Zamudio, 2011; Pounds et al., 2006). in a straight-line distance (see Smith and Green, 2005). Our classifica­
Bd was described in the late 1990's (Longcore et al., 1999), but the tion was based on a stepwise system that followed the sequential scheme
fungus was linked to amphibian declines in Central America and below:
Australia even before its description (Berger et al., 1998; e.g., Groff
et al., 1991; Taylor et al., 1999a, 1999b). In Brazil, despite studies 1) Did the population decline?
reporting Bd infections in amphibians across several Brazilian ecor­ a. Yes = considered in the present study; proceed to question 2. In
egions such as the Atlantic Forest (e.g., Becker and Zamudio, 2011; this case we considered the first year that a multi-year population
Lambertini et al., 2021), Caatinga (Benício et al., 2019), and the Amazon decline was recorded; or the year of decline if data from multiple
basin (Becker et al., 2016; Lambertini et al., 2017), Bd was only recently years was absent (YOD); or the year when the last individual was
linked with historical declines and extinctions (Carvalho et al., 2017). deposited in a museum collection; or the last year when the spe­
Recent studies have linked climatic anomalies with Bd-induced die- cies was detected in the wild; hereafter, last year observed (LYO).
offs in Brazilian amphibians, including droughts possibly inducing host Because YOD is the exact year when the decline started, or a year
aggregation and disease outbreaks near drying water bodies (Moura- very close to that, we chose to use it instead of LYO whenever
Campos et al., 2021). Furthermore, temperature increases have been possible.
linked to higher Bd loads and mortality in Brazilian frogs both under b. No = not considered in the present study.
laboratory (Neely et al., 2021) and natural settings (Rebouças et al., 2) Did the population disappear? (Individuals of that species could not
2021). Modeling analyses predict a reduction in habitat suitability for be found in the site where it previously occurred).
several Brazilian amphibian species (e.g., Costa et al., 2012; Haddad a. Yes = meaning that no individuals could be detected in the wild
et al., 2008; Zellmer et al., 2020), including those caused by sea-level despite subsequent targeted sampling events with significant
rise (de Oliveira et al., 2016a, 2016b), ultimately altering host den­ effort in that site. Proceed to question 3.
sity, community-level interactions (Gomes-Mello et al., 2021), and b. No = meaning that the number of individuals reduced consider­
pathogen transmission. ably, but at least a few individuals could be detected. Proceed to
The Atlantic Forest (AF) of Brazil has always been an emblematic question 4.
ecoregion due to its megadiverse amphibian fauna. Several studies have 3) Was the population rediscovered? (Individuals of a particular species
looked into multiple threats to the rich AF amphibian fauna (e.g.,Car­ were found after years where they previously occurred, but were
valho et al., 2017; Costa et al., 2012; Eterovick et al., 2005; Verdade missing for a given period despite significant survey efforts).
et al., 2012). Recently, Carvalho et al. (2017) conducted a thorough a. Yes = population rediscovered. For each rediscovered population,
retrospective survey of Bd in museum-preserved specimens and found we included the rediscovery date. In this case, we used the year in
that most of the 73 records of population declines (of 65 species) which the rediscovery happened, hereafter year of rediscovery
spatiotemporally matched clusters of Bd outbreaks. These striking (YORd); or the year of the first individual deposited in a collec­
numbers underscore the likely vulnerability of AF frogs to chy­ tion; or first year that the species was observed since the past
tridiomycosis compared to other tropical regions (Scheele et al., 2019). decline, hereafter first year observed (FYO). When the YORd was
Expert opinion acknowledges that previous surveys underestimated the known, we used it instead of FYO. Proceed to question 4.
real number of population declines, especially in the southern and b. No = population missing.
southeastern Brazil, and a significant number of amphibian species 4) Did the rediscovered population or the declined population recover?
ended up not listed as endangered due to lack of data, including several (The overall abundance of individuals of that population recovered
IUCN Data Deficient (DD) species. For instance, Greenspan et al. (2018) to previous numbers).
already added (in their supplementary material) three terrestrial a. Yes = population recovered. In this case, we used the year in
breeding species (Brachycephaloidea) to the list provided by Carvalho which the recovery began; hereafter year of recovery (YORc).
et al. (2017). In the present study we reviewed multiple sources of data b. No = population not recovered.
to provide a more comprehensive picture of the past amphibian declines
in the south and southeast political regions of Brazil where most records A summary of the historical notes on each population, including data
of amphibian declines were reported (e.g., Carvalho et al., 2017; on collection voucher specimens, references, and personal communi­
Greenspan et al., 2018; Heyer et al., 1988; Weygoldt, 1989). We also cations are reported in Table S1. This dataset will be a valuable
reviewed data on recent population recoveries or rediscoveries, which contribution to future ecological studies testing the impacts of global
could improve our understanding of the underlying causes of past change stressors on disease emergence and amphibian community
amphibian die-offs and provide a baseline for understanding amphibian resilience.
community resilience and management. We acknowledge that some of the populations we considered in
Table S1 could be influenced by some different source of errors. For
2. Methodsb.2)b.3)b.4)b. example, if the specific site of the population was unknown for a long
time, or incorrectly assigned in the original records, an increased time-
We summarized records of declines and extinctions of Atlantic Forest lag for its rediscovery or a perception of decline could have been
amphibians up until the year of 2020. We included records from compromised. In order to account for that, we made a separate set of
southeastern and southern Brazil: states of Espírito Santo (ES), Minas analyses without those species (n = 23 species; ~13 % of the total
Gerais (MG), Rio de Janeiro (RJ), São Paulo (SP), Paraná (PR), Santa dataset) for which we did not have the precise locality of its original
Catarina (SC), and Rio Grande do Sul (RS) – as these states combined population (only the municipality centroid for example), or we are
encompass the area where most of the past declines occurred (Carvalho confident that the geographic coordinates were incorrect (such as those
et al., 2017). Within this region, we i) reviewed primary literature and locations that had their type locality corrected). Additionally, as
dissertations from university repositories, summarized accession data different species have different detectability or some individuals could
from museum collections (using available information from the online have been misidentified in museum records, we did another set of an­
database SpeciesLink, added to the data directly from the curatorship of alyses excluding 15 % (which was set to be a higher percentage than the
Museu Nacional), and ii) interviewed amphibian experts in Brazil to previous error estimation, in order to be more conservative) of the
identify populations that have declined or disappeared from the wild species, randomly selected, from the raw dataset. Both sets of analyses
(Table S1). In order to avoid data inflation by clumping those close range are available in Appendix I and their results are consistent with those
metapopulations into the same unit, we considered distinct populations obtained using the full dataset.
those of the same species that were at least 15 km apart from each other To highlight which taxonomic group declined at a higher-than-

2
L.F. Toledo et al. Biological Conservation 277 (2023) 109845

expected rate if declines were evenly distributed across all anurans (Table S1). The families that had a relatively higher proportion of spe­
sampled in the south and southeast Brazil, we calculated the difference cies that declined were Cycloramphidae, Hylodidae and Phyllomedusi­
between the number of species that suffered declines within a family and dae (Fig. 3).
the estimated number of species that would suffer declines if they were The number of herpetologists holding faculty positions in the tar­
evenly distributed across all families (Fig. S1). The deviation of data geted states increased over the years, reaching a total of 63 researchers
points (observed) to the predicted points was plotted in a ranked bar who were actively studying and collecting frogs in southern or south­
chart. eastern Brazil as in 2014 (Fig. 4; Table S2). In general, the number of
In addition to the above classification, we compiled two additional collected specimens increased over the years, peaking in 2011 with
sets of historical data (up to 2020) as follows: i) the number of collected almost 4800 specimens sampled. Despite this general trend, there was a
specimens/year deposited in the main zoological collections of the south marked reduction in the number of specimens sampled in the 1970's,
and southeast Brazilian states (with the exception of Museu de Zoologia with the lowest number of individuals sampled (n = 239) in 1979
da Universidade de São Paulo – MZUSP, for which we could not retrieve (coinciding with the peak of reported population declines). After this
the data; Table S2) independently of species identity; and ii) the number period of declines, the number of individual frogs sampled remained low
of active and hired faculty herpetologists/year at Universities in the up until the year 2000 (for 21 years; Fig. 4A, D), when it started to in­
same Brazilian states (Table S3). We used linear regressions to test for crease to levels prior to the declines of the 1970's (Table S2). This
associations between number of herpetologists and number of sampled pattern indicated the presence of three temporal phases: a) pre-decline
frogs per year. peak (before 1976): when number of faculty actively studying am­
phibians was a positive predictor of amphibian collections (r2 = 0.35;
3. Results F(1,66) = 34.6; P < 0.001; Fig. 4B); b) decline peak (from 1976 to 1985),
when the largest declines were observed and the number amphibian
We compiled information of 169 populations of 106 amphibian specimens deposited in museum collections had a negative, non-
species that have undergone significant declines in the AF (Fig. 1; significant association with the number of active herpetologists (r2 =
Table S1). Among these species, 35 currently lack other viable pop­ 0.29, F(1,10) = 3.2; P = 0.111; Fig. 4C); and c) post-decline peak (after
ulations in situ or ex situ, and most populations that experienced de­ 1985), when again amphibians collections were positively associated
clines (n = 128) are still missing. Our data indicates that the peak of with an increasing number of active herpetologists (r2 = 0.79; F(1,29) =
amphibian declines in the Atlantic Forest occurred in 1979, and only 24 100.7; P < 0.001; Fig. 4D), which also included part of the 21-year
populations have recovered or have been rediscovered after that, with period when amphibian sampling remained low (Fig. 4).
an average of 33.8 years post-decline before recovery (Fig. 2; Table S1).
The taxonomic representation of species that declined was diverse, 4. Discussion
including direct developing (16 populations; 14 species; 2 families) and
indirect developing species (153 populations; 92 species; 10 families) Amphibian declines in Brazil have been recently analyzed and

Fig. 1. Distribution of amphibian declines in south (RS = Rio Grande do Sul, SC = Santa Catarina, and PR = Paraná) and southeast (SP = São Paulo, RJ = Rio de
Janeiro, ES = Espírito Santo, and MG = Minas Gerais) Brazil. Circle size indicates the number of population declines in a single site. Ecoregions are shaded in
different colors.

3
L.F. Toledo et al. Biological Conservation 277 (2023) 109845

Fig. 2. The upper graph shows the number of populations that declined (in red) and that were rediscovered or recovered (in green) over the years. The dashed lines
enclose the decline peak. The lower graph indicates the number of populations that declined, classified in each of the three proposed decline progression categories.
For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.

Fig. 3. Deviation of the number of species observed to suffer

declines in each anuran family, in relationship to the linear
regression between these and the relative number of anuran
species in each family in the geographical limits of this study
(extracted from Fig. S1). Green bars represent those that
deviated negatively in and red bars represent those that devi­
ated positively in relation to the linear regression. For inter­
pretation of the references to colour in this figure legend, the
reader is referred to the web version of this article.

discussed in both local and global-scale studies (Carvalho et al., 2017; there might be other populations to add to this list, our novel data
Scheele et al., 2019), but our novel dataset added nearly 100 records of emphasizes that amphibian population declines in southern and south­
population decline to the existing data, doubling the original number of eastern Brazil were among the most severe ones observed in the tropics.
amphibian declines in Brazil's AF (Carvalho et al., 2017). Our data On the other hand, it is possible that some species were not detected
suggests that these alarming numbers could be an underestimation of over several decades because field herpetologists did not know the exact
historical amphibian declines in the AF. For instance, we are aware that sampling location, or whether species experienced declines, became
other populations (besides those reported here) of Allobates olfersioides rare, and subsequently experienced population recovery. Another
in the state of Rio de Janeiro are missing, but we were unable to retrieve possible shortcoming of our dataset is the heterogeneity in species
the specific dates of declines (as there was no available data for some detectability (Tanadini and Schmidt, 2011). There are evident differ­
collected specimens), hindering their inclusion in our analyses. Thus, as ences in species conspicuousness in this study, which could increase the

4
L.F. Toledo et al. Biological Conservation 277 (2023) 109845

spatiotemporal clustering of Bd occurrence and historical population

declines (Carvalho et al., 2017). Alongside, other pathogens with more
localized impacts such as ranaviruses (e.g., Ranavirus spp.) and Severe
Perkinsea Infection (SPI) (Chambouvet et al., 2020; Lesbarrères et al.,
2012) could have also contributed to disease and declines in AF frogs (e.
g., Ruggeri et al., 2019). Large-scale climatic anomalies have been
recently linked to local chytrid outbreaks or cryptic amphibian declines
in Atlantic Forest frogs, including shifts in temperature (Rebouças et al.,
2021) and rainfall patterns (Moura-Campos et al., 2021). Thus, large-
scale climate change and diseases could be acting synergistically to
explain the observed historical declines. Future studies crossing data
from our newly compiled dataset with retrospective survey of chytrid in
museum preserved specimens and historical records of climatic vari­
ability could shed new light into causes, consequences and management
of amphibian declines in Brazil.
Interestingly, we found that some amphibian families suffered a
disproportionally higher impact, in terms of number of species that
declined, namely: Cycloramphidae, Hylodidae and Phyllomedusidae.
Almost all species that suffered declines in these families inhabit small
Fig. 4. The graphs (A-D) depict the number of amphibians sampled and
streams inside forests (Toledo et al., 2021), and these species often occur
deposited in zoological collections plotted against the number of active her­ in syntopy. Thus, it is possible that the specific habitats they share,
petologists holding faculty positions during the three decline periods: pre- coupled with some intrinsic physiological traits, may have been the
decline peak (highlighted in B), decline peak (highlighted in C), and post- optimal factors leading those species to such disproportional declines.
decline peak (highlighted in D), in south and southeast Brazil. The blue We propose that a reduction in the number of samples deposited in
points denote the 21 years period where samples were still reduced in com­ zoological collections, controlling for the number sampling effort by
parison with the following years (in yellow). For interpretation of the references active herpetologists, could serve as a proxy for historical declines (a
to colour in this figure legend, the reader is referred to the web version of scar left in museum datasets), also in other Brazilian ecoregions such as
this article. the Amazon and Cerrado. Such circumstance would consist of a case of
“ghosts of the past”, which have been pointed as one of the major aspects
chances of not detecting a rare, mute or cryptic species. However, when that are still necessary to explore in retrospective disease ecology studies
herpetologists are looking for a target species, they use targeted methods (James et al., 2015). The detection of such ghost scars could provide
that may reduce the chances of not detecting a rare species. For example, useful information to curb future outbreaks. For example, if similar
for Phantasmarana bocainensis, previous searches used eDNA (Lopes generalized amphibian diversity reduction is observed (in the Atlantic
et al., 2020), tadpole traps, and diurnal and nocturnal active search to Forest or other regions) (e.g., Rebouças et al., 2021), it may precede
find specimens (see details of sampling effort in Table S1); thus, subsequent population extirpations. Thus, immediate actions must be
reducing the chances of not detecting target species. We also acknowl­ taken in that region. It is important to note that the peak of historical
edge that different sampling methods and efforts were applied to amphibian declines (1979) coincided with the 1979 oil shock (Batista,
different populations over the course of the last decades. This is a 2014; Biasetto, 2016). During that time, fuel wasn't sold at gas stations
common confounding factor while reviewing historical data in conser­ on weekends, and the oil price was astronomically high (Biasetto, 2016),
vation studies (e.g., Carvalho et al., 2017; Scheele et al., 2019), and we significantly reducing field work activities at that time. Despite that, the
believe this could also have influenced the records provided in the number of collections remained low even after those restrictions.
present study. Therefore, the temporal detection of both declines and Therefore, this important sampling bias does not explain the observed
reencounters could be influenced by variable sampling methods. We historical declines.
attempted to account for these sources of potential error, and our results Another aspect we observed is that the lag time for a population to
remained unaltered after adding two distinct sources of bias in our recover could vary greatly, from a few years to more than 100 years.
dataset (see Figs. S2 and S3). However, the average lag time observed for population recovery (~33
Local extinctions are notoriously difficult (or impossible) to be years) was close to the length of time (~21 years) that the number of
verified over short periods of time, and population declines are often amphibian specimens accessioned in museums took to reestablish a
even more difficult to describe due the need of long-term standardized positive association with the number of active herpetologists at Brazilian
monitoring. Our novel dataset compiled from zoological collections and research institutions. Thus, we estimate that 20 to 30 years would be the
expert opinion point to a general trend of historical population declines mean timeframe needed for populations to recover from a severe
that temporally match those reported in previous studies (Carvalho reduction in abundance. Consequently, this is another aspect that could
et al., 2017; Scheele et al., 2019). These results indicate that the new be explored in historical collection data across the globe, possibly
dataset only solidifies the timeline of declines and suggests that a much corroborating our findings and detecting past declines, independently of
larger number of AF amphibian species might have suffered significant recent reports.
population declines that peaked during the late 1970's and early 1980's. Finally, we are undergoing a phase in which the “Lazarus effect” is
It is unlikely that localized stressors such as deforestation and environ­ evidenced, meaning that missing species are reappearing even if previ­
mental pollution could explain such large-scale temporally-clustered ously considered to be extinct or probably extinct. Vanished amphibian
declines, especially because a large fraction of population crashes was populations are being rediscovered, or some populations that declined
reported from protected parks with pristine forest cover such as Estação are recovering to their pre-decline numbers, not only in Brazil, but in
Biólogica de Boracéia, Parque Nacional do Itatiaia, and Estação Bio­ other regions in the Neotropics (e.g., Barrio-Amorós et al., 2020; Enciso-
lógica de Santa Lúcia (ES) (Heyer et al., 1988; Weygoldt, 1989). Alter­ Calle et al., 2017; Rodríguez-Contreras et al., 2008; Voyles et al., 2018)
natively, broader scale climatic anomalies could not be ruled out as and Australia (Scheele et al., 2017). Because the global amphibian de­
likely drivers of widespread declines. Thus far, chytridiomycosis is the clines occurred in the same timeframe as the one we are reporting here
only factor that has been empirically linked with past amphibian de­ for Brazil (see Scheele et al., 2019), several parts of the globe could be
clines in Brazil due to his widespread distribution and the undergoing similar phases of amphibian declines and recoveries.

5
L.F. Toledo et al. Biological Conservation 277 (2023) 109845

Rediscoveries and recoveries are certainly great news, indicating that Collins, J.P., Storfer, A., 2003. Global amphibian declines: sorting the hypotheses. Divers.
Distrib. 9, 89–98.
some populations and species are adapting to cope with enzootic Bd, but
Conant, R., 1958. A Field Guide to Reptiles and Amphibians of the United States and
our results indicate that these recoveries are happening at a slow pace Canada East of the 100th Meridian. Houghton Mifflin, Boston.
and at low magnitude (less species recovered than those that declined). Costa, T.R., Carnaval, A., Toledo, L.F., 2012. Mudanças climáticas e seus impactos sobre
This pattern indicates that it is likely that many populations will remain os anfíbios brasileiros. Rev. Biol. 8, 33–37.
Enciso-Calle, M.P., Viuche-Lozano, A., Anganoy-Criollo, M., Bernal, M.H., 2017.
missing and will eventually be declared officially extinct by the IUCN Rediscovery of Atelopus subornatus Werner, 1899 (Anura: Bufonidae), with a
and the local Red Lists. Data from The Brazilian Red List of threatened redescription of the tadpole. Zootaxa 4344, 160–162.
amphibians is showing this trend as well. The past lists (published in Eterovick, P.C., Carnaval, A.C.O.de Q., Borges-Nojosa, D.M., Silvano, D.L., Segalla, M.V.,
Sazima, I., 2005. Amphibian declines in Brazil: an overview. Biotropica 37, 166–179.
1989, 2003, and 2014) included only one formally extinct (EX) species Falaschi, M., Melotto, A., Manenti, R., Ficetola, G.F., 2020. Invasive species and
(only in the list of 2014; the lists of 1989 and 2003 did not assign any amphibian conservation. Herpetologica 76, 216–227.
species to the EX IUCN category) (Ministério do Meio Ambiente, 2003, Gomes-Mello, F., Provete, D.B., Heino, J., da Silva, F.R., 2021. Frog community
composition-environment relationships vary over time: are snapshot studies of
2014). However, the current Brazilian Red List (elaborated in 2018 and metacommunity dynamics useful? Basic Appl. Ecol. 56, 85–96.
published in 2022: Ministério do Meio Ambiente, 2022) is including two Greenspan, S., Lambertini, C., Carvalho, T., James, T., Toledo, L., Haddad, C., Becker, C.,
extinct and 15 species tagged as probably extinct, underscoring the 2018. Hybrids of amphibian chytrid show high virulence in native hosts. Sci. Rep. 8,
9600.
lagging trend of widespread declines and extinctions. Groff, J., Mughannam, A., McDowell, T., Wong, A., Dykstra, M., Frye, F., Hedrick, R.,
Supplementary data to this article can be found online at https://doi. 1991. An epizootic of cutaneous zygomycosis in cultured dwarf african clawed frogs
org/10.1016/j.biocon.2022.109845. (Hymenochirus curtipes) due to Basidiobolus ranarum. J. Med. Vet. Mycol. 29,
215–223.
Haddad, C.F.B., Giovanelli, J.G.R., Alexandrino, J., 2008. Global warming and its effects
Declaration of competing interest on amphibians distribution and decline. In: Biology & Climate Change in Brazil.
Rima Editora, São Carlos, pp. 195–206.
Heyer, W.R., Rand, A.S., da Cruz, C.A.G., Peixoto, O.L., 1988. Decimations, extinctions,
All authors declare no competing interests. and colonizations of frog populations in Southeast Brazil and their evolutionary
implications. Biotropica 230–235.
Data availability James, T.Y., Toledo, L.F., Rödder, D., da Silva Leite, D., Belasen, A.M., Betancourt-
Román, C.M., Jenkinson, T.S., Soto-Azat, C., Lambertini, C., Longo, A.V., Ruggeri, J.,
Collins, J.P., Burrowes, P.A., Lips, K., Zamudio, K.R., Longcore, J.E., 2015.
All data is available in the manuscript and in the supplements Disentangling host, pathogen, and environmental determinants of a recently
emerged wildlife disease: lessons from the first 15 years of amphibian
chytridiomycosis research. Ecol. Evol. 5, 4079–4097.
Acknowledgements
Lambertini, C., Becker, C.G., Bardier, C., da Silva Leite, D., Toledo, L.F., 2017. Spatial
distribution of batrachochytrium dendrobatidis in south american caecilians. Dis.
We thank José P. Pombal Jr. for the access to data from specimens Aquat. Org. 124, 109–116.
from Museu Nacional (MNRJ). Grants and fellowships were provided by Lambertini, C., Becker, C.G., Belasen, A.M., Valencia-Aguilar, A., Nunes-de-Almeida, C.
H.L., Betancourt-Roman, C.M., Rodriguez, D., da Silva Leite, D., Oliveira, I.S.,
São Paulo Research Foundation (FAPESP #2013/50741-7; #2016/ Gasparini, J.L., Ruggeri, J., Tamí, M., Jenkinson, T.S., James, T.Y., Zamudio, K.R.,
25358-3; #2019/18335-5), the National Council for Scientific and Toledo, L.F., 2021. Biotic and abiotic determinants of batrachochytrium
Technological Development (CNPq #312606/2017-6; #306623/2018- dendrobatidis infections in amphibians of the brazilian Atlantic Forest. Fungal Ecol.
49, 100995.
8; #302834/2020-6), the Coordination for the Improvement of Higher Lesbarrères, D., Balseiro, A., Brunner, J., Chinchar, V.G., Duffus, A., Kerby, J., Miller, D.
Education Personnel (CAPES - Finance Code 001), the National Science L., Robert, J., Schock, D.M., Waltzek, T., Gray, M.J., 2012. Ranavirus: past, present
Foundation (NSF #2120084), and by the Mohamed bin Zayed Species and future. Biol. Lett. 8, 481–483. https://doi.org/10.1098/rsbl.2011.0951.
Lips, K.R., Brem, F., Brenes, R., Reeve, J.D., Alford, R.A., Voyles, J., Carey, C., Livo, L.,
Conservation Fund (#202523589). Pessier, A.P., Collins, J.P., 2006. Emerging infectious disease and the loss of
biodiversity in a Neotropical amphibian community. Proc. Natl. Acad. Sci. 103,
References 3165–3170.
Longcore, J.E., Pessier, A.P., Nichols, D.K., 1999. Batrachochytrium dendrobatidis gen.
Et sp. Nov., a chytrid pathogenic to amphibians. Mycologia 91, 219–227.
Barrio-Amorós, C.L., Costales, M., Vieira, J., Osterman, E., Kaiser, H., Arteaga, A., 2020.
Lopes, C.M., Baêta, D., Valentini, A., Lyra, M.L., Sabbag, A.F., Gasparini, J.L., Dejean, T.,
Back from extinction: rediscovery of the harlequin toad Atelopus mindoensis Peters,
Haddad, C.F.B., Zamudio, K.R., 2020. Lost and found: frogs in a biodiversity hotspot
1973 in Ecuador. Herpetol. Notes 13, 325–328.
rediscovered with environmental DNA. Mol. Ecol. 30 (13), 3289–3298.
Batista, L., 2014. Alta do petróleo fez país viver crise nos anos 1970. O Estado de São
Ministério do Meio Ambiente, 2003. Lista das Espécies da Fauna Brasileira Ameaçadas de
Paulo.
Extinção. Instrução Normativa n◦ . 3, de 27 de maio de 2003. Diário Oficial da
Becker, C.G., Zamudio, K.R., 2011. Tropical amphibian population experience higher
República Federativa do Brasil, Brasília, DF.
disease risk in natural habitats. Proc. Natl. Acad. Sci. 108, 9893–9898.
Ministério do Meio Ambiente, 2014. Lista das Espécies da Fauna Brasileira Ameaçadas de
Becker, C.G., Fonseca, C.R., Haddad, C.F.B., Batista, R.F., Prado, P.I., 2007. Habitat split
Extinção. Portaria no 444, de 17 de dezembro de 2014. Diário Oficial União 245 (1),
and the global decline of amphibians. Science 318, 1775–1777.
121–126.
Becker, C.G., Rodriguez, D., Lambertini, C., Toledo, L.F., Haddad, C.F., 2016. Historical
Ministério do Meio Ambiente, 2022. Lista das Espécies da Fauna Brasileira Ameaçadas de
dynamics of batrachochytrium dendrobatidis in Amazonia. Ecography 39, 954–960.
Extinção. Portaria no 148, de 07 de junho de 2022. Diário Oficial União 108 (1), 74.
Benício, R.A., Carvalho, T., Barbosa, M.D., Costa, J.de M., da Silva, F.C., Fonseca, M.G.,
Moura-Campos, D., Greenspan, S.E., DiRenzo, G.V., Neely, W.J., Toledo, L.F., Becker, C.
2019. Worrying news for brazilian caatinga: prevalence of Batrachochytrium
G., 2021. Fungal disease cluster in tropical terrestrial frogs predicted by low rainfall.
dendrobatidis in amphibians. Trop. Conserv. Sci. 12, 1–6.
Biol. Conserv. 261, 109246 https://doi.org/10.1016/j.biocon.2021.109246.
Berger, L., Speare, R., Daszak, P., Green, D.E., Cunningham, A.A., Goggin, C.L.,
Neely, W.J., Greenspan, S.E., Stahl, L.M., Heraghty, S.D., Marshall, V.M., Atkinson, C.L.,
Slocombe, R., Ragan, M.A., Hyatt, A.D., McDonald, K.R., Hines, H.B., Lips, K.R.,
Becker, C.G., 2021. Habitat disturbance linked with host microbiome dispersion and
Marantelli, G., Parkes, H., 1998. Chytridiomycosis causes amphibian mortality
bd dynamics in temperate amphibians. Microb. Ecol. 1–10 https://doi.org/10.1007/
associated with population declines in the rain forests of Australia and Central
s00248-021-01897-3.
America. Proc. Natl. Acad. Sci. 95, 9031–9036.
de Oliveira, I.S., Rödder, D., Capinha, C., Ahmadzadeh, F., de Oliveira, A.K.C., Toledo, L.
Biasetto, B.H., 2016. The Poisoned Chalice: Oil and Macroeconomics in Brazil (1967-
F., 2016a. Assessing future habitat availability for coastal lowland anurans in the
2003). Georgetown University (PhD Dissertation).
brazilian Atlantic rainforest. Stud. Neotropical Fauna Environ. 51, 45–55.
Bishop, P., Angulo, A., Lewis, J., Moore, R., Rabb, G., Moreno, J.G., 2012. The
de Oliveira, I.S., Rödder, D., Toledo, L.F., 2016b. Potential worldwide impacts of sea
Amphibian extinction crisis - what will it take to put the action into the Amphibian
level rise on coastal-lowland anurans. North-West. J. Zool. 12, 91–101.
Conservation Action Plan? SAPIENS. In: Surveys and Perspectives Integrating
Pounds, J.A., Bustamante, M.R., Coloma, L.A., Consuegra, J.A., Fogden, M.P., Foster, P.
Environment and Society.
N., La Marca, E., Masters, K.L., Merino-Viteri, A., Puschendorf, R., Ron, S.R.,
Blaustein, A.R., Wake, D.B., 1990. Declining Amphibian Populations: A Global
Sánchez-Azofeifa, A., Still, C.J., Young, B.E., 2006. Widespread amphibian
Phenomenon?.
extinctions from epidemic disease driven by global warming. Nature 439, 161–167.
Carvalho, T., Becker, C.G., Toledo, L.F., 2017. Historical amphibian declines and
Rebouças, R., Dos Santos, M.M., da Silva Martins, A.G., Domingos, A.H.R., Santos, I.,
extinctions in Brazil linked to chytridiomycosis. Proc. R. Soc. B Biol. Sci. 284,
Toledo, L.F., 2021. Warming drives cryptic declines of amphibians in eastern Brazil.
20162254.
Biol. Conserv. 256, 109035.
Chambouvet, A., Smilansky, V., Jirků, M., Isidoro-Ayza, M., Itoïz, S., Derelle, E.,
Rodríguez-Contreras, A., Señaris, J.C., Lampo, M., Rivero, R., 2008. Rediscovery of
Monier, A., Gower, D.J., Wilkinson, M., Yabsley, M.J., Lukeš, J., Richards, T.A.,
Atelopus cruciger (Anura: Bufonidae): current status in the cordillera de la costa,
2020. Diverse alveolate infections of tadpoles, a new threat to frogs? PLoS Pathog.
Venezuela. Oryx 42, 301–304.
16, e1008107.

6
L.F. Toledo et al. Biological Conservation 277 (2023) 109845

Ruggeri, J., Ribeiro, L.P., Pontes, M.R., Toffolo, C., Candido, M., Carriero, M.M., Taylor, S.K., Williams, E.S., Mills, K.W., 1999a. Experimental exposure of Canadian toads
Zanella, N., Sousa, R.L.M., Toledo, L.F., 2019. Discovery of wild amphibians infected to Basidiobolus ranarum. J. Wildl. Dis. 35, 58–63.
with ranavirus in Brazil. J. Wildl. Dis. 55, 897–902. Taylor, S.K., Williams, E.S., Mills, K.W., 1999b. Mortality of captive Canadian toads from
Scheele, B.C., Skerratt, L.F., Grogan, L.F., Hunter, D.A., Clemann, N., McFadden, M., Basidiobolus ranarum mycotic dermatitis. J. Wildl. Dis. 35, 64–69.
Newell, D., Hoskin, C.J., Gillespie, G.R., Heard, G.W., Brannely, L., Roberts, A.A., Toledo, L.F., Dena, S., Segalla, M., Prado, C.P.A., Loebmann, D., Gasparini, J.L.,
Berger, L., 2017. After the epidemic: ongoing declines, stabilizations and recoveries Sazima, I., Haddad, C.F.B., 2021. Atlantic Forest Amphibians. Cellphone app.
in amphibians afflicted by chytridiomycosis. Biol. Conserv. 206, 37–46. Econature, Consultoria, Pesquisa e Educação Ambiental. Version 1.0.1.
Scheele, B.C., Pasmans, F., Skerratt, L.F., Berger, L., Martel, A., Beukema, W., Verdade, V.K., Valdujo, P.H., Carnaval, A.C., Schiesari, L., Toledo, L.F., Mott, T.,
Acevedo, A.A., Burrowes, P.A., Carvalho, T., Catenazzi, A., la Riva, I.D., Fisher, M.C., Andrade, G.V., Eterovick, P.C., Menin, M., Pimenta, B.V., Nogueira, C., Lisboa, C.S.,
Flechas, S.V., Foster, C.N., Frías-Álvarez, P., Garner, T.W.J., Gratwicke, B., De Paula, C., Silvano, D.L., 2012. A leap further: the Brazilian amphibian
Guayasamin, J.M., Hirschfeld, M., Kolby, J.E., Kosch, T.A., Marca, E.L., conservation action plan. Alytes 29, 1–4.
Lindenmayer, D.B., Lips, K.R., Longo, A.V., Maneyro, R., McDonald, C.A., Voyles, J., Woodhams, D.C., Saenz, V., Byrne, A.Q., Perez, R., Rios-Sotelo, G., Ryan, M.J.,
Mendelson, J., Palacios-Rodriguez, P., Parra-Olea, G., Richards-Zawacki, C.L., Bletz, M.C., Sobell, F.A., McLetchie, S., Reinert, L., Rosenblum, E.B., Rollins-
Rödel, M.-O., Rovito, S.M., Soto-Azat, C., Toledo, L.F., Voyles, J., Weldon, C., Smith, L.A., Ibáñez, R., Ray, J.M., Griffith, E.J., Ross, H., Richards-Zawacki, C.L.,
Whitfield, S.M., Wilkinson, M., Zamudio, K.R., Canessa, S., 2019. Amphibian fungal 2018. Shifts in disease dynamics in a tropical amphibian assemblage are not due to
panzootic causes catastrophic and ongoing loss of biodiversity. Science 363, pathogen attenuation. Science 359, 1517–1519.
1459–1463. Wake, D.B., 1991. Declining amphibian populations. Science 253, 860–861.
Smith, A.M., Green, D.M., 2005. Dispersal and the metapopulation paradigm in Weygoldt, P., 1989. Changes in the composition of mountain stream frog communities in
amphibian ecology and conservation: are all amphibian populations the Atlantic mountains of Brazil: frogs as indicators of environmental deteriorations?
metapopulations? Ecography 28, 110–128. Stud. Neotropical Fauna Environ. 24, 249–255.
Stuart, S.N., Chanson, J.S., Cox, N.A., Young, B.E., Rodrigues, A.S., Fischman, D.L., Zellmer, A.J., Slezak, P., Katz, T.S., 2020. Clearing up the crystal ball: understanding
Waller, R.W., 2004. Status and trends of amphibian declines and extinctions uncertainty in future climate suitability projections for amphibians. Herpetologica
worldwide. Science 306, 1783–1786. 76, 108–120.
Tanadini, L.G., Schmidt, B.R., 2011. Population size influences amphibian detection
probability: implications for biodiversity monitoring programs. Plos One 6 (12),
e28244.