Hidden declines in a seemingly secure Australian frog species

Amphibian populations in upland areas have experienced disproportionately high rates of decline worldwide, yet the status of many remains poorly understood due to limited systematic surveys and long-term monitoring. In eastern Australia, the Pseudophryne bibronii complex (P. bibronii and P. dendyi) was once widespread and common, but its current status is uncertain. We aimed to quantify contemporary occupancy, abundance, and disease prevalence in this group by surveying 70 historically occupied sites spanning an elevational gradient from near sea level to 1700 m above sea level. Sites were surveyed up to six times over two breeding seasons, and adult males were swabbed to determine infection rates with the fungal pathogen Batrachochytrium dendrobatidis (Bd). Occupancy declined sharply with elevation, from 71% at low elevations to 28% in upland areas, with abundance counts showing a similar pattern. Bd prevalence increased from 6.7% at low elevations to 29.2% in upland sites, with the highest infection intensities also recorded at higher elevations. These findings implicate Bd as a potential contributor to upland declines, possibly alongside other environmental pressures. Our results demonstrate that common species can undergo substantial, unrecognized contractions in the absence of targeted monitoring and provide an example of time-lagged declines potentially associated with chytrid fungus.

Understanding species population trajectories is a fundamental challenge in biodiversity conservation, particularly in the context of accelerating global change. Despite widespread concern about species loss, most conservation assessments are underpinned by limited or incomplete data on population trends (Lindenmayer et al. 2012; Wintle et al. 2019; Oliver et al. 2021). Even among species formally listed as threatened, many lack systematic, long-term monitoring, undermining assessments of extinction risk, and prioritization of conservation actions (Proença et al. 2017; Scheele et al. 2019b). For example, in Australia, population trends are not monitored for approximately one-third of listed threatened vertebrates, and many that are monitored are not tracked over meaningful time scales (Legge et al. 2018; Scheele et al. 2019b). As a result, declines may go unnoticed, particularly in taxa with patchy distributions, cryptic behavior, or complex life histories (Donaldson et al. 2016; Scheele et al. 2019b). The rapid decline of many amphibian species over the past four decades illustrates the critical value of systematic surveys and monitoring. Since the early 1980s, amphibians have experienced the most rapid declines of any vertebrate group, with 42% of species now listed as threatened under the IUCN Red List (Luedtke et al. 2023). These declines are driven by multiple threats including disease, habitat loss, invasive species, and climate change (Berger et al. 1998; Skerratt et al. 2007; Hof et al. 2011; Scheele et al. 2019a; Grant et al. 2020; Luedtke et al. 2023; Crawford-Ash et al. 2025). Species for which long-term monitoring exists show varied long-term trajectories: some continue to decline rapidly (Murray et al. 2011; Scheele et al. 2019a; Geyle et al. 2021), while others appear to have stabilized or recovered (Retallick et al. 2004; Newell et al. 2013; Crawford-Ash et al. 2024), even in modified environments (Callaghan et al. 2021). However, threatening processes can also have long-term, less obvious effects such as suppressing abundance, constraining distributions, and reducing resilience to other environmental stressors (Murray et al. 2009; Phillott et al. 2013; Heard et al. 2015, 2024; Belasen et al. 2022; Scheele et al. 2023). Moreover, 23% of amphibians on the IUCN Red List remain Data Deficient (Luedtke et al. 2023), and many more lack consistent, long-term monitoring. Australia is a global hotspot for amphibian declines, with 49 species listed as threatened nationally and eight considered at high risk of extinction (Gillespie et al. 2020; Geyle et al. 2021). While trajectories of some Australian species are well documented, many remain poorly understood due to a lack of systematic survey or monitoring. Two such species are Bibron’s toadlet ( Pseudophyrne bibronii ) and Dendy’s toadlet ( P. dendyi ) — small terrestrial frogs distributed collectively across much of south-eastern Australia (Anstis 2017). Both species were historically widespread and considered common within their ranges, but local extinctions of upland populations have been suspected in recent decades (Osborne 1990; Gillespie et al. 1995; Hunter et al. 2018). These species are closely related and share similar ecological traits and may represent a species complex (Anstis 2017). Because they are difficult to differentiate by call or visual inspection, and genetic boundaries remain unresolved, we refer to them collectively as the P. bibronii complex here. The P. bibronii complex may be particularly vulnerable to decline. Their traits mirror those of other Pseudophryne species that have declined severely, such as P. corroboree and P. pengilleyi, which are both highly susceptible to Bd (Kosch et al. 2019) and rely on breeding sites with a narrow band of hydrological and thermal characteristics (Berger et al. 1998; Osborne et al. 1999; Hunter et al. 2010). The P. bibronii complex occurs across a broader elevational gradient than P. corroboree and P. pengilleyi ; however, upland populations of the P. bibronii complex likely face greater vulnerability than their lowland counterparts due to colder temperatures that are associated with increased Bd impact (Woodhams & Alford 2005; Scheele et al. 2023) and climate-driven changes to the hydroperiod of ponds and seeps that are vital for reproduction (Scheele et al. 2012, 2016a). In this study, we quantified contemporary occupancy and abundance for the P. bibronii complex through resurveys of historically occupied sites across a broad elevational gradient in south-eastern Australia. We also quantified Bd prevalence and infection intensity across the gradient, and the degree of sympatry with the common eastern froglet ( Crinia signifera ); a known Bd reservoir host linked to declines of other upland frogs, including P. corroboree and P. pengilleyi (Hunter et al. 2010; Scheele et al. 2017; Brannelly et al. 2018). We hypothesized that P. bibronii complex occupancy and abundance would decline with increasing elevation, consistent with patterns in other eastern Australian frogs (Hero et al. 2005; Hunter et al. 2018). We also predicted Bd prevalence and intensity would increase with elevation, where cooler, more moist conditions favor pathogen persistence (Woodhams et al. 2003; Scheele et al. 2023). Lastly, we expected C. signifera abundance to negatively affect P. bibronii complex persistence and abundance, especially at higher elevations (Scheele et al. 2017). Study area We compiled over 1,000 historical occurrence records of the P. bibronii complex across south-eastern Australia from museum collections and survey databases, restricting records to those collected prior to 1990 and located within the IBRA Australian Alps (AUA), South East Coastal Plain (SCP), and South East Corner (SEC) bioregions (Atlas of Living Australia 2025; DCCEEW 2025) (Fig. 1). Sites were selected for field surveys based on continued site integrity (i.e., not destroyed), accessibility, and geographic coverage. Many historical records carried high spatial uncertainty. We prioritized sites with high spatial precision (≤1 km) but included sites with coarser data (up to 5 km uncertainty) where records were linked to locality names or defined areas of suitable breeding habitat. For records with precise coordinates, surveys were centered within 1 km of the original point and targeted suitable Pseudophryne breeding habitat (bogs, seepage lines and ephemeral pools). For less precise records, we searched within a 5 km radius to identify appropriate breeding sites before establishing the site (Fig. 1). Male frogs in the P. bibronii complex have high nest site fidelity (Byrne & Silla 2023), however, survey sites were spaced at least 1 km apart to reduce the likelihood of surveying the same individuals across locations. In total, 70 sites were surveyed across the Australian Capital Territory (ACT), New South Wales (NSW), and Victoria (VIC) (Fig. 1), spanning 46 high-elevation sites (1000–1700 m asl), 10 mid-elevation sites (500–999 m asl), and 14 low-elevation sites (10–499 m asl). Field surveys We conducted audio-visual surveys across two consecutive breeding seasons (February–April in both 2023 and 2024). Each site was surveyed up to three times per year, for six visits total. Surveys were spaced 7–10 days apart and conducted under varying weather conditions and times of day (including both day and night) to maximize detection probability based on known activity patterns (Byrne & Silla 2023). At each visit, we surveyed all suitable breeding habitat within a 50 m radius for 15 minutes using call playback from the FrogID app (Rowley et al. 2019) broadcast at ~5-minute intervals. The observer recorded detection (calling or not calling) and estimated the number of calling males using ordinal categories (0, 1 = 1–5, 2 = 6–20, 3 = 21–50, 4 = 51–100 individuals). Surveys were led by the authors (JCA, WO, DS, SP), with one or two volunteers assisting with call detection. Weather data (air temperature, wind speed and relative humidity) were recorded during each survey with a Kestrel 3000 weather meter (Kestrel Instruments, Boothwyn, Pennsylvania, USA), and cloud cover was estimated visually. Additional climate variables were extracted from the NASA POWER database on a site-by-survey basis (Sparks et al. 2024): daily precipitation, weekly accumulated precipitation, next-day precipitation, and average daily temperature. Site-level habitat attributes We recorded abiotic and biotic variables hypothesized to influence P. bibronii occupancy, including suspected decline drivers (Table 1). Elevation was extracted manually from Google Earth Pro using precise GPS coordinates. Fire severity data from the 2019/2020 ‘Black Summer’ bushfires were obtained from the Australian Google Earth Engine Burnt Area Map (AUS GEEBAM) Fire Severity Dataset (Roff & Aravena 2020). Average canopy cover (%), site moisture (0–4 scale), and habitat disturbance (binary) were assessed visually during surveys and averaged across the six survey visits for each site. Sites lacking trees were assigned 0% canopy cover. Habitat disturbance was noted when sites showed signs of soil and vegetation disturbance from introduced mammals, human activity, or other obvious modifications. We also recorded categorical count estimates for C. signifera at each site at each survey (0, 1 = 1–5, 2 = 6–20, 3 = 21–50, 4 = 51–100 individuals). Data processing All data processing and analyses were conducted in R version 4.4.1 (R Core Team 2024). Missing wind speed and humidity values for five ACT sites were imputed using the regional mean. Continuous covariates were z-transformed (mean-centered, divided by two SD) to aid convergence and interpretation (Gelman & Hill 2007). Binary and categorical variables were not standardized. Data availability statement The data that support the findings of this study are openly available in “Pseud_Occupancy” at https://github.com/JordannCA/Pseud_Occupancy. Occupancy modelling We used single-season occupancy models implemented with the unmarked R package (version 1.5.0; Fiske & Chandler 2011) to estimate occupancy (ψ) and detection probability ( p ) for the P. bibronii complex and to evaluate covariate effects (Table 1). A single-season framework was used for this study as we did not observe changes in site status between years, meeting the closure assumption (Mackenzie et al. 2002). Occupancy probability was modelled as a function of elevation, fire severity, canopy cover, and the site-level maximum ordinal category of C. signifera counts (0, 1 = 1–5, 2 = 6–20, 3 = 21–50, 4 = 51–100 individuals). Detection probability was modelled with nine survey-level covariates: daily temperature, next-day precipitation, wind speed, humidity, soil moisture, cloud cover, habitat disturbance, and a cosine transformation of survey time (for diel patterns; see equation below). Next-day precipitation was retained after model comparison as the precipitation variable with the strongest support. This was selected over same-day and weekly precipitation measures to reduce collinearity and because weather conditions preceding rainfall may influence calling behavior and detection in this species. To model diel patterns in detection, we used a two-parameter cosine function: \(\text{\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ log}\left(\frac{p_{i}}{1-p_{i}}\right)=\ \alpha+\ \beta_{1}\times cos\left(\frac{2\pi\times\text{hour}_{i}}{24}\right)+\beta_{2}\ \times sin\left(\frac{2\pi\times\text{hour}_{i}}{24}\right),\ \)Eq. 1 where α is the intercept, and β₁ and β₂ are coefficients capturing diel variation. With detection fixed, we fitted 15 combinations of the four site covariates (elevation, fire severity, canopy cover, C. signifera maximum count category) and ranked models by AICc. Final coefficients were obtained by model-averaging across models with ΔAICc < 2 ( MuMIn package version 1.48.4; Bartoń 2025). Count modelling We analyzed variation in P. bibronii complex counts (representing abundance) using cumulative-link mixed models (CLMM; ordinal package version 2023.12-4.1; Christensen 2019) with a logit link. Counts were binned into ordinal categories from 1-4 (1 = 1–5, 2 = 6–20, 3 = 21–50, 4 = 51–100 individuals), with surveys with zero detections excluded. Site ID was included as a random intercept to account for repeated measures. We constructed 16 candidate models based on covariates aligned with the occupancy analysis: elevation, fire severity, diel activity (night vs day), next-day precipitation and the maximum count category for C. signifera . To test for a stronger reservoir host effect of C. signifera at higher elevations, we included an elevation × Crinia interaction in the candidate model set. Models were ranked by AICc, and coefficients were model averaged across the ΔAICc < 2 subset. Disease sampling We swabbed 45 adult males from the P. bibronii complex across 13 sites spanning low, mid, and high elevations. Swabbing followed the protocols of Boyle et al. (2004), with 30 strokes across ventral body, groin, and hind feet. Frogs were weighed and measured (SVL) for body condition. Swabs were analyzed via qPCR at CESAR laboratory (Melbourne) according to the protocols of Boyle et al. (2004) to detect Bd and quantify infection load (as ‘zoospore equivalents’).

We surveyed 70 sites across south-eastern Australia, stratified by elevation, where the P. bibronii complex historically occurred (Fig. 1). Detection varied markedly across elevation bands: 71.4% of low-elevation sites (10–499 m asl), 10% of mid-elevation sites (500–999 m asl) and 28.3% of high-elevation sites (1000–1700 m asl) (Table S1). Counts of P. bibronii complex were grouped into four ordinal categories (1 = 1–5, 2 = 6–20, 3 = 21–50, 4 = 51–100 individuals). At occupied high-elevation sites, the lowest count category (1–5 individuals) dominated, whereas the highest category (51–100) occurred only at low elevation sites. The 21–50 category was likewise most frequent at low-elevation sites (Table S2). Occupancy modelling Of the 15 candidate models, two received substantial support (ΔAICc < 2; Table S3) and were model-averaged (Table S4). Elevation had a negative effect on occupancy (β = -1.42 ± 0.62 SE, p-value = 0.02; 95% CI: -2.63 – -0.20; Table S4), with predicted occupancy declining with increasing elevation (Fig. 2). In contrast, the Crinia signifera site-level maximum category was positively associated with occupancy (β = 1.47 ± 0.64, p-value = 0.02; 95% CI: 0.22 – 2.72; Table S4). Fire severity showed a weak positive trend (β = 0.31 ± 0.54, p-value= 0.57; 95% CI: 0.33 − 2.01) and was retained in one of the two top models but was not significant. Detection probability was best explained by diel activity patterns, with a significant cos-transformed hour effect (β = 2.02 ± 0.78, p-value = 0.01; 95% CI: 0.50 – 3.55; Table S5). Predicted detection peaked after nightfall, declined through the morning, reached a trough around midday, and increased into the evening (Fig. 3). Other detection covariates including temperature, next-day precipitation, humidity, wind, cloud cover, soil moisture, and habitat disturbance had weak, non-significant effects. Count modelling Of the 16 candidate models, two received substantial support (ΔAICc < 2; Table S6) and were model-averaged (Table S7). Site elevation and survey timing (night versus day) were the strongest predictors of counts of P. bibronii complex (Table S7). P. bibronii complex counts declined with elevation (β = -1.52 ± 0.58 SE, p-value = 0.01; 95% CI: -2.65 – -0.39; Table S7). Surveys conducted at night were also associated with higher counts (β = 1.33 ± 0.58, p-value = 0.02; 95% CI: 0.20 – 2.47), consistent with higher detection probabilities after dark (as above). Next-day precipitation and fire severity were retained in the top two models but were not significant predictors of counts. No other predictors were included in the top models, and interaction terms did not improve model performance. Predicted probabilities from the top-ranked models indicated that high and mid-elevation sites had the highest likelihood of recording the smallest count category (1–5 individuals), while high counts (≥21 individuals) were most probable at low elevations (Fig. 4). Batrachochytrium dendrobatidis sampling We swabbed 45 adult males from 13 sites: low elevation (n = 3 sites, 15 individuals), mid elevation (n = 2 sites, 6 individuals), high elevation (n = 8 sites, 24 individuals). Overall Bd prevalence was 18% (8 of 45 individuals positive), increasing with elevation from 6.7% at low elevation sites to 16.7% at mid elevations and 29.2% at high elevations (Table 2). The single low-elevation positive had a low load (8.17 zoospore equivalents), whereas seven high-elevation positives averaged 164.58 zoospore equivalents. The highest load came from the sole mid-elevation positive (906.33 zoospore equivalents).

Overlooked declines in a seemingly secure species Our results reveal an elevation-linked contraction of the P. bibronii complex from historically occupied sites in south-eastern Australia: frogs were detected at 71.4% of lowland sites, but only 28.3% of upland sites and 10% of mid-elevation sites. Abundance also declined with elevation, with generally low counts (1–5 calling males) at the few occupied upland sites, compared to higher counts (21–50 individuals) at the majority of lowland sites. Our results provide empirical evidence supporting reports (Osborne 1986; Howard et al. 2010; Anstis 2017; Collins 2022; Byrne & Silla 2023) of an elevation-linked decline in the P. bibronii complex. More broadly, our results contribute to a body of research showing that high-elevation amphibians are among the most extinction-prone vertebrates globally (Guirguis et al. 2023). In Australia, all seven potentially extinct taxa are upland species, while the most threatened extant species, such as the southern and northern corroboree frogs ( P. corroboree, P. pengilleyi ), spotted tree frog ( Litoria spenceri ), Baw Baw frog ( Philoria frosti ), and Kroombit tinker frog ( Taudactylus pleione ), are also upland species (Hunter et al. 2018; Scheele et al. 2019a; Geyle et al. 2021). Potential drivers of decline While direct evidence is lacking, we suggest that the decline we document is consistent with a potential role of Bd. Although our Bd sampling was limited, prevalence was highest at upland sites (29.2%) and lowest at lowland sites (6.7%), consistent with the pattern of decline severity. Several mechanisms could underlie increased Bd impacts at higher elevations. First, cooler upland temperatures favor Bd (Piotrowski et al. 2004; Sopniewski et al. 2022; Scheele et al. 2023), and second, longer frog maturation times at higher elevations reduce population capacity to persist despite increased mortality (Scheele et al. 2024). For example, age at maturation in corroboree frogs ( P. corroboree and P. pengilleyi ) increases with elevation, rising from one to four years across a gradient of 950–1600 m asl, which reduces population resilience to amplified adult mortality and increases the likelihood of local extinction (Scheele et al. 2024). Given the substantial elevation range of our sites (10 to ~1700 m asl), we suggest that slower maturation at high elevations, combined with higher Bd prevalence, are key mechanisms underpinning the elevational pattern of decline documented here. Reservoir hosts can amplify Bd prevalence in sympatric species (Brannelly et al. 2018; Hudson et al. 2019; Burns et al. 2021; Wilber et al. 2022). In P. pengilleyi, C. signifera presence has been linked to increased Bd impact (Scheele et al. 2017; Brannelly et al. 2018). We predicted that counts of C. signifera would be negatively related to occupancy probabilities for frogs in the P. bibronii complex, with the possibility of an interactive effect with elevation, because the P. bibronii complex is likely to be less resilient to Bd at higher elevations (as above). Additionally, C. signifera typically has lower Bd infection prevalence at lowland sites (~10%) but very high prevalence at upland sites (Scheele et al. 2016b; Brannelly et al. 2018; Crawford-Ash & Rowley 2021). Contrary to this expectation, occupancy probability for P. bibronii complex showed a positive association with counts of C. signifera . We interpret this relationship as likely reflecting shared environmental preferences, particularly at low elevation sites, where both species are more common, rather than facilitation at higher elevations where Bd risk is greater. Models provided no support for an interaction between elevation and C. signifera counts as a predictor of P. bibronii complex occupancy (ΔAICc > 2; Table S6). Furthermore, C. signifera counts showed no clear association with elevation (Spearman ρ = −0.12, 95% bootstrap CI −0.35 to 0.12; p = 0.33; Table S8; Fig. S1). Accordingly, our data do not indicate suppression of P. bibronii and P. dendyi by C. signifera at the scales analyzed here; however, context-specific effects particularly in high-elevation habitats cannot be ruled out. One potential reason for the differences between our results and those documented for P. pengilleyi is that C. signifera were largely ubiquitous across our sites (detected at 62 of 70 sites), whereas some P. pengilleyi sites lacked C. signifera, and those sites had significantly lower Bd prevalence (Scheele et al. 2017). Unlike some eastern Australian frogs that collapsed abruptly with the emergence of Bd in the 1980s (Berger et al. 1998, 1999; Osborne et al. 1999; Hero & Morrison 2004), declines in the P. bibronii complex appear more gradual. Reports from the 1990s documented loss of P. bibronii from many former Canberra sites (Osborne 1990; Rauhala 1997) and found few records in Victoria, mostly restricted to lower elevations in the Alps (Gillespie et al. 1995). P. dendyi were historically documented in higher elevations sites (~800–1000 m) but considered insufficiently known (Rauhala 1997). Since then, there has been little systematic survey effort and a lack of long-term monitoring, and the rate of decline has remained uncertain. Our re-surveys of historical sites confirm that important declines have occurred, but that the species still persists across much of its historical range in our study area, even in the upland regions in which it appears most exposed to Bd impacts. One explanation for the slower pace and patchier nature of declines experienced by the P. bibronii complex compared with other susceptible species are ecological characteristics that restrict Bd transmission. Adults of both P. bibronii and P. dendyi are highly terrestrial and lay eggs in terrestrial nests that are washed into nearby water bodies with flooding rains, sometimes weeks after breeding (Gillespie et al. 1995). Therefore, the adults are likely to have reduced contact with aquatic habitats and motile zoospores of Bd, reducing rates of contagion and infection prevalence (Kriger et al. 2007; Ruggeri et al. 2018). It is notable that infection prevalence among the breeding males swabbed during this study was low, at 18% overall. Another factor that could have contributed to the decline of the P. bibronii complex is recruitment failure during drought. While droughts are not an unusual occurrence in the study region, their severity is increasing due to rising temperatures and long-term reductions in autumn rainfall in south-eastern Australia, which has been linked to reduced runoff and moisture availability (Wasko et al. 2024). Species in the P. bibronii complex are autumn breeders, with slow-developing larvae dependent on stable, shallow pools (Anstis 2017). Notably, mid-elevation areas in our study, which had the lowest occupancy rates, lie within a known rain-shadow zone (Dey et al. 2019) and may be especially vulnerable to recruitment failure during droughts. Other research has also documented drought-associated amphibian declines in parts of the study region (Scheele et al. 2012; Evans et al. 2020) and we recommend further research focus on the hydrology of breeding sites, and how this factor shapes population persistence through time. Fire and anthropogenic disturbance are major threats to some amphibians (Rowley et al. 2020; Beranek et al. 2023; Heard et al. 2023). Given the severity of the 2019-2020 ‘Black Summer’ fires in south-eastern Australia, and their negative impacts on a broad range of fauna (Legge et al. 2022), we predicted a negative relationship between fire and frog occupancy. However, there was no significant relationship between fire severity and occupancy for the P. bibronii complex. Working in coastal woodlands and heathlands, Westgate et al. (2018) reported context-dependent fire effects on P. bibronii occurrence , with occurrence increasing with time since fire at sites characterized by low breeding-site density and frequent burning. It is plausible that frogs in the P. bibronii complex respond positively to fire more generally, if early successional conditions are favorable for adult or larval survival, and/or reduce the impacts of Bd (for example, due to negative effects of shading on temperature regimes; see Heard et al. 2015; Roznik et al. 2015; Rumschlag & Boone 2020). Nevertheless, fire severity in our study was derived from a remotely sensed raster layer and did not necessarily capture fine-scale impacts of fire at each burnt site. As such, field-based assessments will be important to clarify how P. bibronii and P. dendyi respond to fire regimes. The risk of overlooking gradual declines Research on amphibian declines has tended to focus on acute, rapid declines (Fisher & Garner 2020; Luedtke et al. 2023; Crawford-Ash et al. 2025), with gradual declines potentially going unrecognized, especially for species that remain common in parts of their range. Without historical baselines or long-term monitoring, these declines may be missed (Oliver et al. 2021). Cryptic amphibian species and those in remote montane or tropical regions are underrepresented in monitoring efforts (Silva et al. 2020; Angulo et al. 2024), and targeted surveys, such as undertaken in this study, can help close these gaps and improve detection of declines in under monitored species. Well-designed survey protocols that maximize detection are an important characteristic of robust monitoring (Scheele et al. 2019b). Our results showed detection probability peaked after nightfall, and while rainfall or moisture variables were not significant predictors of occupancy or counts, next-day rainfall was retained in the top count model, suggesting possible behavioral shifts before rainfall events. The weak negative relationship between probability of detecting calling males and next-day rainfall (Table S5) indicates that surveys immediately before rain may underestimate true occupancy or counts to some degree. Additionally, surveys should span the full environmental and geographic range of a species. Although our study covered most of the range of P. dendyi and a portion of the southern range of P. bibronii, further work is needed to assess whether the scale and pattern of declines documented here are replicated elsewhere across the range of these frogs. While the P. bibronii complex remains widespread and relatively common at low elevations, declines and population losses have occurred in mid and higher elevation regions. Our results highlight the need for targeted, range-wide surveys to detect localized declines before they become irreversible. Recovery in upland sites will depend on better understanding of the threatening processes therein. We recommend focusing on the potential role of Bd in the declines that we document, including reservoir species dynamics, as well as quantifying the effects of fire severity and breeding site hydrology to examine potential species responses to drought and changing climatic conditions. Studies of this nature will be vital for identifying management interventions that can stem further population losses and recover depressed populations, and either facilitate natural recolonization of locally extinct populations or support reintroduction efforts.

We thank Nick Clemann and Phil Byrne for generous advice at project inception that helped to shape our study design and field methods. We also thank all the volunteers who assisted with our fieldwork, particularly Mitchell Hodgson, Stephen Mahony and Rebecca Seeto. We also acknowledge our funding received from the Holsworth Wildlife Research Endowment (S20100I5). Angulo, A., et al. (2024). Common themes and challenges. In S. Wren, A. Borzée, R. Marcec‑Greaves, & A. Angulo (Eds.), Amphibian conservation action plan: A status review and roadmap for global amphibian conservation (IUCN SSC Occasional Paper No. 57). Gland, Switzerland: IUCN. Anstis, M. (2017). Tadpoles and frogs of Australia . New Holland Publishers. Atlas of Living Australia. (2025). Atlas of Living Australia database . https://www.ala.org.au/ Bartoń, K. (2025). MuMIn: Multi‑model inference [R package]. CRAN. https://CRAN.R-project.org/package=MuMIn Belasen, A. M., Amses, K. R., Clemons, R. A., Becker, C. G., Toledo, L. F., & James, T. Y. (2022). Habitat fragmentation in the Brazilian Atlantic Forest is associated with erosion of frog immunogenetic diversity and increased fungal infections. Immunogenetics, 74 (4), 431–441. Beranek, C. T., et al. (2023). Severe wildfires promoted by climate change negatively impact forest amphibian metacommunities. Diversity and Distributions, 29, 785–800. Berger, L., et al. (1998). Chytridiomycosis causes amphibian mortality associated with population declines in the rain forests of Australia and Central America. Proceedings of the National Academy of Sciences, 95, 9031–9036. Berger, L., Speare, R., & Hyatt, A. (1999). Chytrid fungi and amphibian declines: Overview, implications and future directions. In Declines and disappearances of Australian frogs (pp. 23–33). Environment Australia, Canberra. Boyle, D. G., Boyle, D. B., Olsen, V., Morgan, J. A. T., & Hyatt, A. D. (2004). Rapid quantitative detection of chytridiomycosis ( Batrachochytrium dendrobatidis ) in amphibian samples using real‑time TaqMan PCR assay. Diseases of Aquatic Organisms, 60, 141–148. Brannelly, L. A., Webb, R. J., Hunter, D. A., Clemann, N., Howard, K., Skerratt, L. F., Berger, L., & Scheele, B. C. (2018). Non‑declining amphibians can be important reservoir hosts for amphibian chytrid fungus. Animal Conservation, 21, 91–101. Burns, T. J., Scheele, B. C., Brannelly, L. A., Clemann, N., Gilbert, D., & Driscoll, D. A. (2021). Indirect terrestrial transmission of amphibian chytrid fungus from reservoir to susceptible host species leads to fatal chytridiomycosis. Animal Conservation, 24, 602–612. Byrne, P. G., & Silla, A. J. (2023). Lifetime breeding‑site and nest‑site fidelity in a declining terrestrial toadlet: Evidence for a win‑stay/lose‑shift strategy. Frontiers in Conservation Science, 4 . 1226658. Callaghan, C. T., Liu, G., Mitchell, B. A., Poore, A. G. B., & Rowley, J. J. L. (2021). Urbanization negatively impacts frog diversity at continental, regional, and local scales. Basic and Applied Ecology, 54, 64–74. Christensen, R. H. B. (2019). ordinal: Regression models for ordinal data [R package]. CRAN. https://CRAN.R-project.org/package=ordinal Collins, S. (2022). The role of disease and environment in driving amphibian population change (Doctoral dissertation, University of New England). https://hdl.handle.net/1959.11/53235 Crawford‑Ash, J., Erens, J., Martel, A., Noble, D. W. A., Pasmans, F., & Scheele, B. C. (2025). Defying decline: Very low chytrid prevalence in tadpoles, yet high infection in adults in a naturally recovering frog species. Animal Conservation, 28 (4), 567–581. Crawford‑Ash, J., Evans, M. J., Carvalho, T., Rowley, J. J. L., Garner, T. W. J., Muths, E., & Scheele, B. C. (2025). Evolution of research on global amphibian declines. Conservation Biology, e70146. Crawford‑Ash, J., & Rowley, J. J. L. (2021). Bad neighbours: Amphibian chytrid fungus Batrachochytrium dendrobatidis infection dynamics in three co‑occurring frog species of southern Sydney, Australia. Diseases of Aquatic Organisms, 143, 101–108. Department of Climate Change, Energy, the Environment and Water (DCCEEW). (2025). Australia’s bioregions (IBRA) . https://www.dcceew.gov.au/environment/land/nrs/science/ibra [accessed October 13, 2025] Dey, R., Lewis, S. C., Arblaster, J. M., & Abram, N. J. (2019). A review of past and projected changes in Australia’s rainfall. WIREs Climate Change, 10, e577. Donaldson, M. R., Burnett, N. J., Braun, D. C., Suski, C. D., Hinch, S. G., Cooke, S. J., & Kerr, J. T. (2016). Taxonomic bias and international biodiversity conservation research. FACETS, 1, 105–113. Evans, M. J., Scheele, B. C., Westgate, M. J., Yebra, M., Newport, J. S., & Manning, A. D. (2020). Beyond the pond: Terrestrial habitat use by frogs in a changing climate. Biological Conservation, 249, 108712. Fisher, M. C., & Garner, T. W. J. (2020). Chytrid fungi and global amphibian declines. Nature Reviews Microbiology, 18, 332–343. Fiske, I., & Chandler, R. (2011). unmarked: An R package for fitting hierarchical models of wildlife occurrence and abundance. Journal of Statistical Software, 43, 1–23. Gelman, A., & Hill, J. (2007). Data analysis using regression and multilevel/hierarchical models . Cambridge University Press. Geyle, H. M., et al. (2021). Red hot frogs: Identifying the Australian frogs most at risk of extinction. Pacific Conservation Biology, 28, 211–223. Gillespie, G. R., Osborne, W. S., & McElhinney, N. A. (1995). The conservation status of frogs in the Australian Alps: A review . https://nswdpe.intersearch.com.au/nswdpejspui/handle/1/9071 [accessed September 20, 2025] Gillespie, G. R., Roberts, J. D., Hunter, D., Hoskin, C. J., Alford, R. A., Heard, G. W., Hines, H., Lemckert, F., Newell, D., & Scheele, B. C. (2020). Status and priority conservation actions for Australian frog species. Biological Conservation, 247, 108543. Grant, E. H. C., Miller, D. A. W., & Muths, E. (2020). A synthesis of evidence of drivers of amphibian declines. Herpetologica, 76, 101–107. Guirguis, J., Goodyear, L. E. B., Finn, C., Johnson, J. V., & Pincheira‑Donoso, D. (2023). Risk of extinction increases towards higher elevations across the world’s amphibians. Global Ecology and Biogeography, 32, 1952–1963. Heard, G. W., Bolitho, L. J., Newell, D., Hines, H. B., Norman, P., Willacy, R. J., & Scheele, B. C. (2023). Drought, fire, and rainforest endemics: A case study of two threatened frogs impacted by Australia’s “Black Summer.” Ecology and Evolution, 13, e10069. Heard, G. W., Scroggie, M. P., Hollanders, M., & Scheele, B. C. (2024). Age truncation due to disease shrinks metapopulation viability for amphibians. Journal of Animal Ecology, 93, 1670–1683. Heard, G. W., Thomas, C. D., Hodgson, J. A., Scroggie, M. P., Ramsey, D. S. L., & Clemann, N. (2015). Refugia and connectivity sustain amphibian metapopulations afflicted by disease. Ecology Letters, 18, 853–863. Hero, J.‑M., & Morrison, C. (2004). Frog declines in Australia: Global implications. Herpetological Journal, 14, 175–186. Hero, J.‑M., Williams, S. E., & Magnusson, W. E. (2005). Ecological traits of declining amphibians in upland areas of eastern Australia. Journal of Zoology, 267, 221–232. Hof, C., Araújo, M. B., Jetz, W., & Rahbek, C. (2011). Additive threats from pathogens, climate and land‑use change for global amphibian diversity. Nature, 480, 516–519. Howard, K., Cleeland, C., & Clemann, N. (2010). Assessment of the status of the threatened Bibron’s toadlet and southern toadlet in areas affected by the Kilmore East‑Murrindindi fires: Black Saturday, Victoria 2009—Natural Values Fire Recovery Program . Department of Sustainability and Environment. Hudson, M. A., et al. (2019). Reservoir frogs: Seasonality of Batrachochytrium dendrobatidis infection in robber frogs in Dominica and Montserrat. PeerJ, 7, e7021. Hunter, D., Clemann, N., Coote, D., Gillespie, G., Hollis, G., Scheele, B., Philips, A., & West, M. (2018). Frog declines and associated management response in south‑eastern mainland Australia and Tasmania. In Status of conservation and decline of amphibians: Australia, New Zealand and Pacific Islands (pp. 38–58). CSIRO Publishing. Hunter, D. A., Speare, R., Marantelli, G., Mendez, D., Pietsch, R., & Osborne, W. (2010). Presence of the amphibian chytrid fungus Batrachochytrium dendrobatidis in threatened corroboree frog populations in the Australian Alps. Diseases of Aquatic Organisms, 92, 209–216. Kosch, T. A., Silva, C. N. S., Brannelly, L. A., Roberts, A. A., Lau, Q., Marantelli, G., Berger, L., & Skerratt, L. F. (2019). Genetic potential for disease resistance in critically endangered amphibians decimated by chytridiomycosis. Animal Conservation, 22, 238–250. Kriger, K. M., Pereoglou, F., & Hero, J.‑M. (2007). Latitudinal variation in the prevalence and intensity of chytrid ( Batrachochytrium dendrobatidis ) infection in eastern Australia. Conservation Biology, 21, 1280–1290. Legge, S., et al. (2022). The conservation impacts of ecological disturbance: Time‑bound estimates of population loss and recovery for fauna affected by the 2019–2020 Australian megafires. Global Ecology and Biogeography, 31, 2085–2104. Legge, S., Lindenmayer, D. B., Robinson, N. M., Scheele, B. C., Southwell, D. M., & Wintle, B. A. (2018). Monitoring threatened species and ecological communities . CSIRO Publishing. Lindenmayer, D. B., et al. (2012). Value of long‑term ecological studies. Austral Ecology, 37, 745–757. Luedtke, J. A., et al. (2023). Ongoing declines for the world’s amphibians in the face of emerging threats. Nature, 622, 308–314. MacKenzie, D. I., Nichols, J. D., Lachman, G. B., Droege, S., Royle, J. A., & Langtimm, C. A. (2002). Estimating site occupancy rates when detection probabilities are less than one. Ecology, 83, 2248–2255. Murray, K. A., Retallick, R. W. R., Puschendorf, R., Skerratt, L. F., Rosauer, D., McCallum, H. I., Berger, L., Speare, R., & VanDerWal, J. (2011). Assessing spatial patterns of disease risk to biodiversity: Implications for the management of the amphibian pathogen, Batrachochytrium dendrobatidis . Journal of Applied Ecology, 48, 163–173. Murray, K. A., Skerratt, L. F., Speare, R., & McCallum, H. (2009). Impact and dynamics of disease in species threatened by the amphibian chytrid fungus, Batrachochytrium dendrobatidis . Conservation Biology, 23, 1242–1252. Newell, D. A., Goldingay, R. L., & Brooks, L. O. (2013). Population recovery following decline in an endangered stream‑breeding frog ( Mixophyes fleayi ) from subtropical Australia. PLOS ONE, 8(3). e58559. Oliver, R. Y., Meyer, C., Ranipeta, A., Winner, K., & Jetz, W. (2021). Global and national trends, gaps, and opportunities in documenting and monitoring species distributions. PLOS Biology, 19, e3001336. Osborne, W. (1990). Declining frog populations and extinctions in the Canberra region. Bogong, 11, 4–7. Osborne, W., Hunter, D., & Hollis, G. (1999). Population declines and range contraction in Australian alpine frogs. In Declines and disappearances of Australian frogs (pp.145-157). Environment Australia. Osborne, W. S. (1986). Frogs of the Canberra region. Bogong, 7, 10–12. Phillott, A. D., Grogan, L. F., Cashins, S. D., McDonald, K. R., Berger, L., & Skerratt, L. F. (2013). Chytridiomycosis and seasonal mortality of tropical stream‑associated frogs 15 years after introduction of Batrachochytrium dendrobatidis . Conservation Biology, 27, 1058–1068. Piotrowski, J. S., Annis, S. L., & Longcore, J. E. (2004). Physiology of Batrachochytrium dendrobatidis, a chytrid pathogen of amphibians. Mycologia, 96 (1), 9–15. Proença, V., et al. (2017). Global biodiversity monitoring: From data sources to Essential Biodiversity Variables. Biological Conservation, 213, 256–263. Rauhala, M. A. (1997). Frog fauna of the Australian Capital Territory: A review . ACT Parks and Conservation Service. Retallick, R. W. R., McCallum, H., & Speare, R. (2004). Endemic infection of the amphibian chytrid fungus in a frog community post‑decline. PLOS Biology, 2 (11), e351. Roff, A., & Aravena, R. (2020). Australian Google Earth Engine burnt area map: A rapid, national approach to fire severity . Australian Government. Available from http://rgdoi.net/10.13140/RG.2.2.13434.52167 [accessed September 20, 2025]. Rowley, J. J. L., Callaghan, C. T., & Cornwell, W. K. (2020). Widespread short‑term persistence of frog species after the 2019–2020 bushfires in eastern Australia revealed by citizen science. Conservation Science and Practice, 2 (11), e287. Rowley, J. J. L., Callaghan, C. T., Cutajar, T., Portway, C., Potter, K., Mahony, S., Trembath, D. F., Flemons, P., & Woods, A. (2019). FrogID: Citizen scientists provide validated biodiversity data on frogs of Australia. Herpetological Conservation and Biology, 14, 155–170. Roznik, E. A., Sapsford, S. J., Pike, D. A., Schwarzkopf, L., & Alford, R. A. (2015). Natural disturbance reduces disease risk in endangered rainforest frog populations. Scientific Reports, 5, 13472. Ruggeri, J., de Carvalho‑e‑Silva, S. P., James, T. Y., & Toledo, L. F. (2018). Amphibian chytrid infection is influenced by rainfall seasonality and water availability. Diseases of Aquatic Organisms, 127, 107–115. Rumschlag, S. L., & Boone, M. D. (2020). Amphibian infection risk changes with host life stage and across a landscape gradient. Journal of Herpetology, 54 (3), 347–354. Scheele, B. C., et al. (2019a). Amphibian fungal panzootic causes catastrophic and ongoing loss of biodiversity. Science, 363, 1459–1463. Scheele, B. C., et al. (2019b). Continental‑scale assessment reveals inadequate monitoring for threatened vertebrates in a megadiverse country. Biological Conservation, 235, 273–278. Scheele, B. C., Driscoll, D. A., & Fischer, J. (2012). Decline of an endangered amphibian during an extreme climatic event. Ecosphere, 3 (11), 1–15. Scheele, B. C., Heard, G. W., Cardillo, M., Duncan, R. P., Gillespie, G. R., Hoskin, C. J., Mahony, M., Newell, D., Rowley, J. J. L., & Sopniewski, J. (2023). An invasive pathogen drives directional niche contractions in amphibians. Nature Ecology & Evolution, 7, 1682–1692. Scheele, B. C., Hunter, D. A., Banks, S. C., Pierson, J. C., Skerratt, L. F., Webb, R., & Driscoll, D. A. (2016a). High adult mortality in disease‑challenged frog populations increases vulnerability to drought. Journal of Animal Ecology, 85, 1453–1460. Scheele, B. C., Hunter, D. A., Brannelly, L. A., Skerratt, L. F., & Driscoll, D. A. (2017). Reservoir‑host amplification of disease impact in an endangered amphibian. Conservation Biology, 31, 592–600. Scheele, B. C., Webb, R. J., Hua, X., & Hollanders, M. (2024). Variation in amphibian maturation rates influences population vulnerability to disease‑induced declines. Animal Conservation, 27, 600–610. da Silva, A. F., Malhado, A. C. M., Correia, R. A., Ladle, R. J., Vital, M. V. C., & Mott, T. (2020). Taxonomic bias in amphibian research: Are researchers responding to conservation need? Journal for Nature Conservation, 56, 125829. Skerratt, L. F., Berger, L., Speare, R., Cashins, S., McDonald, K. R., Phillott, A. D., Hines, H. B., & Kenyon, N. (2007). Spread of chytridiomycosis has caused the rapid global decline and extinction of frogs. EcoHealth, 4, 125–134. Sopniewski, J., & Scheele, B. C., & Cardillo, M. (2022). Predicting the distribution of Australian frogs and their overlap with Batrachochytrium dendrobatidis under climate change. Diversity and Distributions, 28, 1255–1268. Sparks, A. H., et al. (2024, December 18). nasapower: NASA POWER API client [R package]. CRAN. Retrieved March 13, 2025, from https://cran.r-project.org/web/packages/nasapower/index.html Wasko, C., Stephens, C., Peterson, T. J., Nathan, R., Pepler, A, Hettiarachchi, S., Vogel, E., Johnson, F., & Westra, S. (2024). Understanding the implications of climate change for Australia’s surface water resources: Challenges and future directions. Journal of Hydrology, 645, 132221. Westgate, M. J., MacGregor, C., Scheele, B. C., Driscoll, D. A., & Lindenmayer, D. B. (2018). Effects of time since fire on frog occurrence are altered by isolation, vegetation and fire frequency gradients. Diversity and Distributions, 24, 82–91. Wilber, M. Q., et al. (2022). Once a reservoir, always a reservoir? Seasonality affects the pathogen maintenance potential of amphibian hosts. Ecology, 103, e3759. Wintle, B. A., et al. (2019). Global synthesis of conservation studies reveals the importance of small habitat patches for biodiversity. Proceedings of the National Academy of Sciences, 116, 909–914. Woodhams, D. C., & Alford, R. A. (2005). Ecology of chytridiomycosis in rainforest stream frog assemblages of tropical Queensland. Conservation Biology, 19 (5), 1449–1459. Woodhams, D. C., Alford, R. A., & Marantelli, G. (2003). Emerging disease of amphibians cured by elevated body temperature. Diseases of Aquatic Organisms, 55 (1), 65–67. Woodruff, D. S. (1977). Male postmating brooding behavior in three Australian Pseudophryne (Anura: Leptodactylidae). Herpetologica, 33, 296–303 Tables Table 1. Covariates used in models of the probability of site occupancy (ψ) and detection during surveys ( p ) for the Pseudophryne bibronii complex. Covariates were selected based on their potential to influence occurrence and detectability, as described by ‘ecological relevance’. Variables were measured in the field or obtained from spatial datasets (AUS GEEBAM) and climate datasets (NASA POWER). | Elevation | Occupancy | Elevation of the survey site (m), extracted from GPS coordinates. | Climatic conditions, hydrology, and disease prevalence often vary along elevational gradients, affecting persistence. | | Fire severity | Occupancy | Fire severity from AUS GEEBAM raster. | Severe burns can cause mortality, changes to vegetation, microclimate, and breeding habitat quality. | | Canopy cover | Occupancy | Average tree canopy cover (%) across surveys, visually estimated. | Influences temperature regulation, and moisture retention. | | Maximum count category of C. signifera | Occupancy | Highest count category observed (0, 1=1–5, 2=6–20, 3=21–50, 4=51–100). | Potential reservoir for Bd; may influence occupancy via increased disease transmission. | | Time of survey (hour) | Detection | Hour of day survey was conducted (0–23). | Captures diel variation in calling activity. | | Daily temperature | Detection | Temperature at time of survey (°C). | Affects calling behavior and detectability. | | Next-day precipitation | Detection | Total precipitation (mm) on the day after the survey. | Linked to calling activity and breeding site conditions. | | Wind | Detection | Wind speed at the time of survey (m/s). | High wind may reduce calling behavior or interfere with detection. | | Humidity | Detection | Relative humidity (%) during survey. | Affects calling behavior and frog activity. | | Moisture level | Detection | Categorical moisture rating from dry- high (0–4). | Reflects site wetness at survey time. | | Cloud cover | Detection | Estimated cloud cover (%) during survey. | May affect microclimate and calling. | | Habitat disturbance | Detection | Binary index (1=disturbed, 0=undisturbed). | May affect calling or habitat quality. | Table 2 . Batrachochytrium dendrobatidis (Bd) infection rates in Pseudophryne bibronii complex populations across three elevation bands. Zoospore equivalents are untransformed. | Low (10-499 m asl) | 3 | 15 | 1 | 6.7 | | Mid (500-999 m asl) | 2 | 6 | 1 | 16.7 | | High (1000-1700 m asl) | 8 | 24 | 7 | 29.2 | Figure Legends Fig. 1. Study area and survey outcomes for Pseudophryne bibronii complex across the Australian Capital Territory, Southern Tablelands of New South Wales and north-eastern Victoria, Australia. Background colors show elevation (0–2,000 m asl). Points denote sites surveyed in this study (green filled = detected; white filled = not detected). Grey crosses are historical records. Fig. 2. Model-averaged relationships between predicted occupancy probability of the Pseudophryne bibronii complex and two site-level covariates: (A) elevation and (B) Crinia signifera maximum count category (0, 1–5, 6–20, 21–50, 51–100 individuals). Shaded areas represent 95% confidence intervals. Fig. 3. Effect of time of day on detection probability for the Pseudophryne bibronii complex from the top-ranked occupancy model. The solid line is the model-estimated mean and the shaded ribbon is the 95% confidence interval. Fig. 4. Predicted probability of each count category for the Pseudophryne bibronii complex from the averaged cumulative link mixed models (CLMM; models within ΔAICc < 2). Count categories are: 1–5, 6–20, 21–50, 51–100 individuals. (A) Predictions across elevation, holding other predictors at their mean or reference values. (B) Predictions by diel period (day vs night), holding other predictors constant. Shaded areas or error bars represent 95% confidence intervals. Fig 1. Fig. 2 Fig. 3 Fig. 4 Supplementary Material File (supplementary material_10_oct.docx) - Download - 63.20 KB Information & Authors Information Version history Copyright This work is licensed under a Non Exclusive No Reuse License. Collection

Authors Metrics & Citations Metrics Article Usage 345views 190downloads Citations Download citation Jordann Crawford-Ash, Stephanie Pulsford, Will Osborne, et al. Hidden declines in a seemingly secure Australian frog species. Authorea. 17 November 2025. DOI: https://doi.org/10.22541/au.176336538.89763970/v1 DOI: https://doi.org/10.22541/au.176336538.89763970/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu.