• chytridiomycosis;
  • Crinia georgiana ;
  • frog declines;
  • Myobatrachidae;
  • south-western Australia


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  1. The amphibian chytrid fungus Batrachochytrium dendrobatidis is a major cause of frog declines globally.

  2. Two recent Maxent models predict high environmental suitability for B. dendrobatidis in south-western Australia despite severe summer drought and many frog species with direct development or breeding in ephemeral water bodies: features often associated with absence of B. dendrobatidis.

  3. We determined B. dendrobatidis occurrence and intensity of infection, in 15 populations of Crinia georgiana, a frog that breeds in ephemeral ponds, to assess (i) current validity of claims about environmental suitability, (ii) risk of decline against recent suggestions of 100% prevalence and chytrid loads of > 10 000 zoospore equivalents per frog as predictors of imminent decline and (iii) assess apparent change in prevalence over 20 years.

  4. Batrachochytrium dendrobatidis occurred on 40–100% of frogs at all sites and in water samples at 11 sites but infection levels were generally well below 10 000 zoospore equivalents per frog. Based on immediate and 5-year climate averages, higher rainfall, more rain days and temperatures > 3 °C but < 30 °C were positively correlated with infection levels – consistent with the known physiology and growth patterns of B. dendrobatidis. Overall infection levels have changed little from 1992 to 2008.

  5. Synthesis and applications. Predicted and realized high environmental suitability for B. dendrobatidis is not correlated with frog decline in south-western Australia. Innate or acquired immunity, chytrid strain type and limited opportunities for chytrid growth may all explain absence of chytrid impacts. Occurrence of Batrachochytrium dendrobatidis is not a current critical management issue for conservation managers in south-western Australia despite a known presence since 1985. Models predicting high environmental suitability for chytrid in Mediterranean climate zones should be interpreted cautiously in the absence of documentation of current, rather than historic, chytrid loads and a clear evaluation of any occurrence of chytrid on frog survival.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

The amphibian chytrid fungus, Batrachochytrium dendrobatidis Longcore, Pessier & Nichols (1999), known world-wide, is associated with catastrophic frog population crashes in many areas of the world (Skerrat et al. 2007; Lips et al. 2008; Wake & Vredenburg 2008) and is one of the most intensively studied causes for anuran declines (Rohr et al. 2008; Fisher, Garner & Walker 2009). Many B. dendrobatidis–associated declines have occurred at high altitude and/or in tropical locations (Lips et al. 2006; Murray et al. 2011a) but B. dendrobatidis has a global range in habitats from wet, tropical rain forest to Mediterranean climates (Ron 2005; Fisher, Garner & Walker 2009).

The occurrence and impact of B. dendrobatidis on frogs are tempered by time spent in contact with conspecifics and in stream zones (e.g. Rowley & Alford 2007a,b), by thermoregulatory behaviour (Richards-Zawacki 2010; Puschendorf et al. 2011) and by habitat, with a lower incidence in frogs sampled from ephemeral water bodies and in species with terrestrial egg deposition where adults never enter water (Kriger & Hero 2007). Batrachochytrium dendrobatidis is absent in many (e.g. Lips et al. 2006; Pearl et al. 2007; Hauselberger & Alford 2012), but not all (Longo & Burrowes 2010; Longo, Burrowes & Joglar 2010), direct developing species. These features limiting B. dendrobatidis incidence are also noted in patterns of frog decline with several analyses suggesting species with an aquatic life stage are at greater risk (e.g. Bielby et al. 2008; Smith, Lips & Chase 2009; Murray et al. 2011b) with the implication that aquatic environments favour B. dendrobatidis persistence or infection.

Murray et al. (2011a) and Liu, Rohr & Li (2013) both used an array of climate variables to predict environmental suitability for B. dendrobatidis and Liu, Rohr & Li (2013) also included a proxy for vegetation status and anuran species richness, to define a fundamental niche for B. dendrobatidis. Liu, Rohr & Li (2013) also included metrics affecting B. dendrobatidis spread: predictors of ‘propagule pressure’ – the presence of non-native anuran species, trade, trade in frog legs and human impact. These two studies concur in predicting high to very high environmental suitability for B. dendrobatidis in south-western Australia, where there are no non-native anurans (Tyler & Doughty 2009), no trade in frog legs and moderate human impact (see current maps at suggesting an over-riding impact of habitat or climate suitability.

The area of Mediterranean climate in south-western Australia supports 29 frog species (see Table S1, in Supporting Information), but despite B. dendrobatidis records from 15 species in this region dating back to 1985 (Aplin & Kirkpatrick 2000, 2001; Table S1, Supporting Information, Murray et al. 2010), there are no records of species decline or extinction attributable to infection with B. dendrobatidis (Hero et al. 2006). Aplin & Kirkpatrick (2001) did report sick and dying frogs infected with B. dendrobatidis, but we are not aware of any evidence that this is a current problem in south-western Australia. To understand B. dendrobatidis risk in south-western Australia, we need data in three areas: (i) current B. dendrobatidis occurrence given existing data are up to 15 or more years old (Murray et al. 2010, 2011a), (ii) data are based on techniques (histology) that underestimate occurrence and cannot quantify load (Skerrat et al. 2008), (iii) an understanding of what proportion of frogs are infected, what B. dendrobatidis loads they carry and whether these are close to threshold levels associated with precipitous decline of host populations (cf. Vredenburg et al. 2010) and (iv) B. dendrobatidis strain identity given only some strains are hypervirulent (Farrer et al. 2011).

We worked with Crinia georgiana, a frog species with many attributes that are predictors of decline: high environmental suitability for chytrid, aquatic life stages, intermediate range size and some use of bog and/or soak breeding habitats (Bielby et al. 2008; Murray et al. 2011a,b; Liu, Rohr & Li 2013) and because C. georgiana shares many of these ecological attributes with many other frog species in this region (see Table S1, Supporting Information).

We quantified both the incidence and prevalence of B. dendrobatidis in 15 populations of C. georgiana spread across the whole range of the species. We also assessed how immediate and longer-term climate affected B. dendrobatidis infection levels to determine whether B. dendrobatidis biology follows expected patterns based on its known physiology and impact of weather on patterns of abundance in other regions (cf. Kriger, Pereoglou & Hero 2007). We discuss mechanisms that may allow B. dendrobatidis to persist in Mediterranean climates with little impact. We then argue that without a better understanding of mechanisms of occurrence, persistence and resistance to B. dendrobatidis infection, models of environmental suitability for B. dendrobatidis may be poor predictors of risk of decline or extinction of frogs in strongly seasonal climate zones and consequently of limited use to conservation managers.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Frog Species

Crinia georgiana is widespread in south-western Australia in coastal swamps and in ephemeral shallow seepages and drainage systems in forests from Perth to east of Esperance (Tyler & Doughty 2009). Crinia georgiana breeds in shallow water (1–4 cm deep) in winter (Seymour et al. 2000). Frogs may breed in permanent water bodies, but out of 10 sites reported by Dziminski et al. (2010) and 15 in this study, only one is likely to contain water over summer. Crinia georgiana has been captured in pit traps during surveys of terrestrial vertebrates in south-western forests (Wayne, Liddelow & Williams 2011) indicating frogs are not tied to water bodies outside the breeding season but there are no reports of the natural history of this species outside the breeding season.

In July–September 2008, we surveyed for B. dendrobatidis in 15 populations of C. georgiana spread across its range (Fig. 1, see Table S2, Supporting Information) covering most of the area of predicted environmental suitability for B. dendrobatidis in south-western Australia (Liu, Rohr & Li 2013; fig. 2; Murray et al. 2011a; fig. 4f). We sampled up to 40 frogs per population (sampling protocols, including sample sizes, in Skerrat et al. 2008) as 29·3% of all C. georgiana samples reported by Murray et al. (2010) scored positive for B. dendrobatidis.


Figure 1. Sampling locations for Batrachochytrium dendrobatidis on Crinia georgiana. Sites are numbered by latitude. Inset indicates general location in south-western Australia. For site details, sampling dates and sample sizes, see Table S1 (Supporting Information). Note: site 8 is far to the east.

Download figure to PowerPoint

Collection and Chytrid Sampling

Frogs were collected and handled wearing new latex gloves for each frog and placed in individual, new, 20 × 20 cm plastic, zip lock bags to eliminate cross-contamination. Frogs were all sampled by K. Riley with individual, sterile cotton swabs (Medical Wire and Equipment, MW−100). Each frog was sampled with 10 unidirectional strokes in the following locations: (1) ventral surface from groin to throat; (2) both sides, armpit to groin; (3) underside of both legs, ankle to groin; and (4) five times across the underside of each foot: 70 strokes per swab following the protocol outlined by Kriger, Hero & Ashton (2006). Vredenburg et al. (2010) used 30 strokes on the abdomen, feet and limbs to form a single ‘swab’ per frog. Crinia georgiana averaged 24·3–32·7 mm snout–vent length across populations (see Table S2, Supporting Information) but Rana species studied by Vredenburg et al. (2010) were about double that size (67 and 59 mm, Vredenburg et al. 2007). Differences in sampling techniques will affect estimates of total B. dendrobatidis spores per frog but these will also be affected by frog size and therefore area sampled per stroke. Swabs were returned to their original tubes and stored at 4 °C immediately upon return from the field for subsequent DNA analysis to detect B. dendrobatidis.

DNA Analysis

We used quantitative DNA analysis protocols outlined by Boyle et al. (2004) to detect B. dendrobatidis DNA and quantify intensity of load with modifications suggested by Kriger, Hero & Ashton (2006) (see Appendix S1, Supporting Information).

The presence of B. dendrobatidis on a swabbed frog does not necessarily mean a frog is infected (Kriger et al. 2007; Smith 2007). Zoospores may be detected on the skin surface: for example, because the frog has been in contact with an infected individual or zoospores are present in the water or on the substrate (Kriger et al. 2007). Consequently, we sampled 10 ml of pond water at breeding sites to detect trace DNA from B. dendrobatidis that might account for incidental detections on frog skin (Ficetola et al. 2008).

Climate and Weather

We analysed the relationship between climate and levels of infection with B. dendrobatidis at two levels: (a) short-term climate: over the 15, 30 and 45 days prior to sampling, we aggregated total rainfall and scored maximum and minimum daily temperatures following Kriger, Pereoglou & Hero (2007) and (b) long-term climate: we scored annual rainfall, highest monthly rainfall, average February maximum temperature (the hottest month in this region), annual proportion of days with maximum temperature > 30 °C, annual proportion of days with minimum temperature < 3 °C and averaged rain days per annum. Long-term climate data were averaged over 5 years (2003–2007). These annual temperature ranges represent approximate limits to B. dendrobatidis growth in the laboratory (Piotrowoski, Annis & Longcore 2004), and we presumed higher annual rainfall, highest monthly rainfall and more rain days all mean either a greater likelihood of maintaining standing water in ponds for longer or more humid or moist substrate conditions if ponds do not contain water. We used the preceding 5 years of record as south-western Australia has experienced a long-term decline, 10–20%, in annual, and June–August rainfall in the period 1970–2005 (Taschetto & England 2009). For climate data sources and treatment of missing values, see Appendix S2 (Supporting Information). For temperature and rainfall data by site, see Tables S3 and S4 (Supporting Information).

We conducted a principal components analysis (PCA) to eliminate correlations between climate variables and base our analyses of correlations with climate on derived principal component (PC) scores for PCs with eigenvalues > 1. Principal component scores were calculated (i) from a correlation matrix because of the different measured variables included and (ii) without rotation. We correlated PC scores with B. dendrobatidis load variables to avoid assuming any causal relationships between climate and B. dendrobatidis infection levels. Analyses were conducted in StatistiXL (

We estimated distance to a permanent water body (e.g. permanent streams, farm dams) for each sampling site using on-ground inspection and imagery on Google Earth. Permanent water might act as a reservoir for fungus (Murray et al. 2011a) but these were not sites commonly used for breeding by this species or many other frog species in this region (see Table S1, Supporting Information).

Historical Data on Chytrid Occurrence

We assessed historical occurrence of B. dendrobatidis on C. georgiana using data records in Murray et al. (2010) for the years 1991–1993 and 1996–1999. We scored two data sets: (a) proportion of all records positive for years with 7 or more records and (b) site-specific records with 10 or more frogs sampled.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

We swabbed 574 frogs from 15 sites across the range of C. georgiana: 30–43 frogs per site (Table S2, Supporting Information). Site 1 was an urban drainage sump with adjacent natural swamp land. The sump contains water year-round but the swamp land dries over summer. Other sites were shallow, ephemeral ponds with no drainage linking them to permanent water bodies. None had flowing water. Permanent water bodies occurred at or close to all sites: farm dams (10), large dams for human use (2) and natural water bodies (2). The distance to permanent water bodies ranged from 10 to 2210 m.


Batrachochytrium dendrobatidis was detected on 432 frogs of 574 (75·3%) frogs sampled across all 15 sites. Infection rates for individual sites ranged from 40·5% to 100%, with only six sites having more than 90% of individuals carrying B. dendrobatidis (Table 1). The load of B. dendrobatidis on individual frogs ranged from 0 to 31 052 zoospore equivalents. Average site intensity ranged from 53·6 to 4827 zoospore equivalents per frog (Table 1). Percentage of frogs infected and mean zoospores/frog were positively correlated across sites, and this was marginally significant (r13 = 0·509, = 0·052).

Table 1. Occurrence of Batrachochytrium dendrobatidis on Crinia georgiana and in water samples across south-western Australia. Site details in Table S2 (Supporting Information). For 13 sites, minimum infection was zero. At two sites with 100% infection, lowest levels were 1 zoospore per frog (Albany) and 97 zoospores per frog (Augusta)
SiteFrogs infected (%)Zoospores per frogWater
MeanMaximumSEzsp 10 ml−1
  1. zsp, zoospores.

2.75·0473·5910 192276·250
4.94·02371·0317 834576·684·7
5.55·8504·3411 326281·090
7.92·51622·0114 024466·040·474
10.1004827·0031 0526038·562·87

Batrachochytrium dendrobatidis zoospores were detected in water at nine sites (Table 1) but there was no correlation between levels in water samples and incidence (r13 = −0·132, = 0·639) or prevalence (r13 = 0·0863, = 0·760). If sites with no detections in water samples are excluded – there were still no significant correlations (zoospores/frog, r7 = 0·397,% frogs infected r7 = 0·011). There were no significant correlations between distance to permanent water and intensity (r13 = −0·223) or prevalence (r13 = −0·084) of infection or average snout–vent length and intensity (r13 = −0·016) or prevalence (r13 = 0·480) of infection.

Historical Data

Batrachochytrium dendrobatidis was recorded in 1992–1993 and 1996–1999 (Table 2). Overall prevalence was lower in all years than in this study, expected given differences in detection techniques, but prevalence at two sites was higher than at some sites sampled in 2008 (compare data in Tables 1 and 2).

Table 2. Current and historical data on B. dendrobatidis occurrence in C. georgiana. WA covers all records from Western Australia but only includes years with a total sample size > 6. Site lists all historic sites from a single year with a sample size > 9. Data are from the data base in Murray et al. (2010) and this study
 Year% positiveSample size
Historical records
WA (this study)200880·3574
Site-specific records
10 km E Kalamunda19925010
Boulder Rock19924520
Ashendon Road19992825
Brookton Highway199966·715
WA (this study)200840·5–10036–43

Weather in the Year of Sampling

Principal components analysis on the six proximate climate variables generated three principal components with eigenvalues > 1 that collectively explained 92% of the variance in recent weather variables (see Table S5, Supporting Information). Both intensity (r13 = 0·658, < 0·01) and prevalence (r13 = 0·654, < 0·01) were significantly, positively correlated with PC2, which was positively loaded by recent rainfall and minimum temperature (Table S5, Supporting Information): cooler, wet conditions predict higher prevalence and intensity of infection.

Five-Year Climate

Principal components analysis on the six long-term weather variables gave two principal components with eigenvalues > 1, which collectively explained 83% of the total variation (see Table S6, Supporting Information). Prevalence was negatively correlated with PC1 score (r13 = −0·597, < 0·05). PC1 was positively loaded by maximum February temperatures, annual proportion of days > 30 °C, days < 3 °C and negatively by rain days per annum (see Table S6, Supporting Information): very cold (<3 °C) and very high (>30 °C) temperatures reduced load of B. dendrobatidis but more rain days increased load.

For both short- and long-term climate analyses, neither intensity nor prevalence was correlated significantly with day of the year: our ‘random’ variable included to detect spurious correlations.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Despite suggestions and demonstrations that frogs breeding in temporary water bodies have low levels of infection (Kriger & Hero 2007; Skerrat et al. 2008), our data show high percentages of frogs infected and moderate loads of B. dendrobatidis at all sites sampled despite all but one of these drying annually over summer. The levels of infection per frog averaged 1–2 orders of magnitude below levels reported as causing precipitous declines in frog populations (c. 10,000 zoospore equivalents per frog: Vredenburg et al. (2010)). Direct comparison is difficult as (i) we used 70 versus 30 strokes per frog (see 'Materials and methods') and (ii) C. georgiana is smaller than the Rana species studied by Vredenburg et al. (2010). Critically, Vredenburg et al. (2010) showed a threshold level of infection that seriously compromised survival but that level may vary with tolerance to infection and frog size. Given there is no evidence of decline in C. georgiana, it is not clear what that threshold might be for this species. Vredenburg et al. (2010) also reported prevalence levels of c. 100% associated with precipitous decline. This was matched in only two C. georgiana populations (Table 2). Our data are consistent with B. dendrobatidis occurring commonly on C. georgiana but having little impact on the conservation status of this species where there are no reported declines (Hero et al. 2006). This claim is further supported by multiple studies on this species over 11 years at one site, Kangaroo Gully where B. dendrobatidis infections were first reported in 1992 (Murray et al. 2010; studies summarized in Roberts & Byrne 2011) and across south-western Australia (e.g. Dziminski et al. 2010).

Batrachochytrium dendrobatidis was detected in 60% of water samples – coincidentally comparable to detection rates in water samples from infected sites in the USA (59·5%, Hossack et al. 2009) and Spain (64%, Walker et al. 2007). Incidental occurrence on the skin of frogs may account for some of our positive records but there was no correlation between levels of B. dendrobatidis detected in water samples and average zoospores per frog across populations suggesting accidental contact is not a major source of records on frogs. That argument assumes a constant sampling effort for frogs of all sizes but larger frogs may have a larger area sampled even using a standard protocol, and there is considerable variation in body size within and between populations of C. georgiana (Table S2, Supporting Information; Smith, Withers & Roberts 2003). Populations with larger frogs did not have higher infection rates (Table 1 and see Table S2, Supporting Information).

Incidence and the percentage of frogs infected both correlated with short-term and long-term climate data as anticipated from existing literature. In common with Kriger, Pereoglou & Hero (2007), we found prevalence and abundance were both positively correlated with PC2 reflecting higher rainfall over the preceding 15–45 days and warmer, minimum temperatures. This is consistent with B. dendrobatidis levels increasing over winter as rainfall accumulates and with known data about B. dendrobatidis growth patterns where low temperatures inhibit growth (e.g. Piotrowoski, Annis & Longcore 2004; Rohr et al. 2008). We did not find any inhibitory effect of high temperature as the maximum temperature at any site in the 45 days preceding sampling was 20·5 °C, well below temperatures that inhibit B. dendrobatidis growth (Piotrowoski, Annis & Longcore 2004; Rohr et al. 2008). B. dendrobatidis was less common at sites with high average summer temperatures (>30 °C) and more cold days (<3 °C) but more likely to occur at sites with more rain days per year: sites that are more likely to stay moist for longer or allow more frog activity consistent with other broad-scale climate analyses (e.g. Ron 2005; Rödder, Veith & Lötters 2008; Fisher, Garner & Walker 2009; Lötters et al. 2009). In suitable conditions (high rainfall, frequent rain days, temperatures within the known temperature tolerances for B. dendrobatidis), prevalence and incidence of B. dendrobatidis may rise in populations of C. georgiana. In years with higher winter rainfall or rainfall extending into spring and summer B. dendrobatidis infections might develop to levels where they cause deaths.

Crinia georgiana has maintained moderate to high infection levels with B. dendrobatidis since at least 1991 (Table 2; Murray et al. 2010), which is unexpected given earlier reports where high ESBd was strongly and positively associated with extinction and decline by Murray et al. (2011a, their figure 3a, and logistic models reported in the text). This is a common pattern for frogs in south-western Australia where there are historical records for 14 species in the Murray et al. (2010) data base, most from the 1990s, and recent data based on DNA detection methods for 8 of those species (see Table S1, Supporting Information) with no evidence of decline or extinction due to disease (Hero et al. 2006). Frogs in this region appear to coexist successfully with B. dendrobatidis infection but B. dendrobatidis also occurs on many species with atypical life cycles: many species that never enter water or where tadpoles are the only aquatic stage (see Table S1, Supporting Information).

The absence of discernible impact on population densities and regional persistence in C. georgiana and other frog species in this region leads us to three plausible explanations for persistence but lack of impact of B. dendrobatidis in this region and four models that might allow repeated infection with limited postinfection impact.

  1. Frogs may be naturally resistant to B. dendrobatidis infection. The mechanism is unknown but the occurrence of B. dendrobatidis on a broad taxonomic spread of species [three frog families: Pelodryadidae (one genus), Myobatrachidae (two genera) and Limnodynastidae (three genera)] suggests all species might be resistant. Microbial populations on the skin have been implicated in resistance to B. dendrobatidis (Daskin & Alford 2012), and cross-taxon immunity to B. dendrobatidis can be generated by a single bacterial species (Harris et al. 2009). Most species may possess MHC antigens generating resistance to either initial infection or that can control infection levels once initiated (Savage & Zamudio 2011).
  2. The strains of B. dendrobatidis occurring in south-western Australia may not be lethal. Farrer et al. (2011) reported three major strains: Bd GPL, with a pan-global distribution, Bd CAPE from southern Africa and Bd CH from Switzerland. Infection trials showed the Bd CAPE strain had little impact on survival but the Bd GPL strain, the common strain in North, Central and South America, caused high levels of mortality (Farrer et al. 2011).
  3. Climate may vary enough within and across seasons for most species to maintain nonlethal, fungal loads (Longo, Burrowes & Joglar 2010; Daskin, Alford & Puschendorf 2011) – the context dependent, symbiosis argument developed by Daskin & Alford (2012). This requires models for persistence of B. dendrobatidis in periods when climate is inimical to survival – considered below.
    1. Di Rosa et al. (2007) reported encysted B. dendrobatidis in the skin of Rana lessonae (but see Kilpatrick, Briggs & Daszak 2009 for an alternative view). If these cysts are activated rapidly when frogs enter water or are exposed to wetter conditions in winter, dermal cysts could be a general explanation for persistence of B. dendrobatidis over summer when most frogs in this region are not associated with water bodies of any type.v
    2. Encysted forms of B. dendrobatidis may persist in soil when ponds dry in spring and reinfect each winter with current drier winters reducing the potential for growth to lethal levels. Flooding or saturation of soils the following autumn–winter may lead to persistent infection sources in the water column or in soil water. Alternation of sexual and asexual life stages has been reported in other chytrid fungi and has been suggested as a mechanism for B. dendrobatidis to survive dry periods (Morgan et al. 2007). Clonal and sexual reproduction have been inferred from genetic studies on chytrid species (Morgan et al. 2007), and sexual reproduction can lead to the production of resistant meiosporangia in other Chytridiomycota (Morgan et al. 2007).
    3. Batrachochytrium dendrobatidis may reinvade newly formed temporary water bodies from adjacent permanent water bodies: for example, carried by birds or dispersing invertebrates. Batrachochytrium dendrobatidis was not detected on tropical, aquatic crustaceans (Rowley et al. 2007) but in laboratory trials can survive on damp bird feathers (Johnson & Speare 2005). Acceptance of this model requires a demonstration that B. dendrobatidis occurs consistently in permanent water bodies in this region and that there are appropriate vectors for dispersal moving between infected and clean water bodies.
    4. Batrachochytrium dendrobatidis may persist in moist soils over summer and proliferate when sites are flooded. Johnson & Speare (2005) reported B. dendrobatidis survival in soils that were 33% water by mass. Persistence of B. dendrobatidis in moist soils could account for infection in south-west frog species but we should also expect this in temporary pond breeders in other regions or in species with terrestrial eggs in very wet environments, which is not true in Australia (Kriger & Hero 2007; Hauselberger & Alford 2012).

Management and Conservation Issues

If B. dendrobatidis can survive in a resting phase in seasonally arid areas or whole faunas are resistant, then sampling protocols to detect B. dendrobatidis developed by Skerrat et al. (2008) that preferentially focus on frog species breeding in streams or permanent water bodies are not appropriate. In climate zones like south-western Australia with strongly seasonal climates, understanding the population dynamics and mechanisms for persistence of B. dendrobatidis in and outside the breeding season and assessing the true potential for decline are needed before sensible sampling or hygiene protocols can be developed.

Crinia georgiana is sympatric with three threatened frog species, Geocrinia alba, G. vitellina and Spicospina flammocaerulea (Australian Government 2012). The calling season of C. georgiana overlaps with the breeding season of all three threatened species – all spring and early summer breeders (Tyler & Doughty 2009). The persistence of B. dendrobatidis in these sites may be enhanced if C. georgiana is a reservoir that enhances the persistence of B. dendrobatidis into spring and summer, potentially affecting all three species. However, only G. alba shows any evidence of decline and this is related to extreme habitat fragmentation, resultant small population size in fragments and chance extinctions exacerbated by very limited dispersal and fire rather than any known disease impact (Hero et al. 2006). Experimental infection of these threatened species with locally derived lineages of B. dendrobatidis could be used to assess that risk.

Aplin & Kirkpatrick (2000, 2001) reported summaries of B. dendrobatidis infection in frogs from south-western Australia (with no supporting data or analysis). Except for Geocrinia alba (see above), none of the frog species reported as chytrid positive by Murray et al. (2010) and discussed by Aplin & Kirkpatrick (2000, 2001) has shown signs of broad-scale decline in Western Australia. This general claim is also supported by our analysis of current and historical chytrid infection levels in C. georgiana (Tables 1 and 2). Under current climate and strain conditions, B. dendrobatidis may be a benign infection.

Our new data confirm the broad expectations of the Murray et al. (2011a) and Liu, Rohr and Li (2103) and similar, earlier models in Lötters et al. (2009): high environmental suitability for B. dendrobatidis is reflected in widespread occurrence on one anuran host in south-western Australia. However, high ESbd is not an indicator of B. dendrobatidis causing decline or extinction in this region. Assessing true risk from B. dendrobatidis infection may be habitat or climate zone specific. The clear shift from wet winters to hot dry summers in Mediterranean climate zones may commonly limit B. dendrobatidis growth or survival, and this may be true in other regions with similar strong seasonal shifts in climate (Longo, Burrowes & Joglar 2010). Knowing a site is environmentally suitable for B. dendrobatidis is of no value to on-ground management of frog populations unless suitability can be matched to an understanding of conditions that truly put frog populations at risk.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

K. Riley's Honours research was funded by the School of Animal Biology, UWA, Department of Environment & Conservation, WA, and The Peter Rankin Trust Fund for Herpetology. Sharron Perks helped with fieldwork. Thanks to an anonymous reviewer and K. Murray for comments on earlier versions of this manuscript.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Aplin, K. & Kirkpatrick, P. (2000) Chytridiomycosis in southwest Australia: historical sampling documents the date of introduction, rates of spread and seasonal epidemiology, and sheds new light on chytrid ecology. Getting the Jump on Amphibian Disease (eds K. Moore & R. Speare), pp. 24. Rainforest CRC, Cairns.
  • Aplin, K. & Kirkpatrick, P. (2001) In pursuit of the frog fungus. Landscape, 16, 1016.
  • Australian Government 2012. (accessed June 25, 2012)
  • Bielby, J., Cooper, N., Cunningham, A.A., Garner, T.W.J. & Purvis, A. (2008) Predicting susceptibility to future declines in the world's frogs. Conservation Letters, 1, 8290.
  • 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, 141148.
  • Daskin, J.H. & Alford, R.A. (2012) Context-dependent symbioses and their potential roles in wildlife diseases. Proceedings of the Royal Society B-Biological Sciences, 279, 14571465.
  • Daskin, J.H., Alford, R.A. & Puschendorf, R. (2011) Short-term exposure to warm microhabitats could explain amphibian persistence with Batrachochytrium dendrobatidis. PLoS One, 6, e26215.
  • Di Rosa, I., Simoncelli, F., Fagotti, A. & Pascolini, R. (2007) Ecology: the proximate cause of frog declines? Nature, 447, E4E5.
  • Dziminski, M.A., Roberts, J.D., Beveridge, M. & Simmons, L.W. (2010) Among-population covariation between sperm competition and ejaculate expenditure in frogs. Behavioral Ecology, 21, 322328.
  • Farrer, R.A., Weinert, L.A., Bielby, J., Garner, T.W.J., Balloux, F., Clare, F., Bosch, J., Cunningham, A.A., Weldon, C., du Preez, L.H., Anderson, L., Kosakovsky Pond, S.L., Shahar-Golan, R., Henk, D.A. & Fisher, M.C. (2011) Multiple emergences of genetically diverse amphibian-infecting chytrids include a globalized hyper-virulent recombinant lineage. Proceedings of the National Academy of Science of USA, 108, 1873218736.
  • Ficetola, G.F., Miaud, C., Pompanon, F. & Taberlet, P. (2008) Species detection using environmental DNA from water samples. Biology Letters, 4, 423425.
  • Fisher, M.C., Garner, T.W.J. & Walker, S.F. (2009) Global emergence of Batrachochytrium dendrobatidis and amphibian chytridiomycosis in space, time and host. Annual Review of Microbiology, 63, 291310.
  • Harris, R.N., Brucker, R.M., Walke, J.B., Becker, M.H., Schwantes, C.R., Flaherty, D.C., Lam, B.A., Woodhams, D.C., Briggs, C.J., Vredenburg, V.T. & Minbiole, K.P.C. (2009) Skin microbes on frogs prevent morbidity and mortality caused by a lethal skin fungus. The ISME Journal, 3, 818824.
  • Hauselberger, K.F. & Alford, R.A. (2012) Prevalence of Batrachochytrium dendrobatidis infection is extremely low in direct-developing Australian microhylids. Diseases of Aquatic Organisms, 100, 191200.
  • Hero, J.M., Morrison, C., Gillespie, G., Roberts, J.D., Newell, D., Meyer, E., McDonald, K., Lemckert, F., Mahony, M., Osborne, W., Hines, H., Richards, S., Hoskin, C., Clarke, J., Doak, N. & Shoo, L. (2006) Overview of the conservation status of Australian frogs. Pacific Conservation Biology, 12, 315320.
  • Hossack, B.R., Muths, E., Anderson, C.W., Kirshtein, J.D. & Corn, P.S. (2009) Distribution limits of Batrachochytrium dendrobatidis: a case study in the Rocky Mountains, USA. Journal of Wildlife Diseases, 45, 11981202.
  • Johnson, M.L. & Speare, R. (2005) Possible modes of dissemination of the amphibian chytrid Batrachochytrium dendrobatidis in the environment. Diseases of Aquatic Organisms, 65, 181186.
  • Kilpatrick, A.M., Briggs, C.J. & Daszak, P. (2009) The ecology and impact of chytridiomycosis: an emerging disease of amphibians. Trends in Ecology and Evolution, 25, 109118.
  • Kriger, K.M. & Hero, J.M. (2007) The chytrid fungus Batrachochytrium dendrobatidis is non-randomly distributed across amphibian breeding habitats. Diversity and Distributions, 13, 781788.
  • Kriger, K.M., Hero, J.M. & Ashton, K.J. (2006) Cost efficiency in the detection of chytridiomycosis using PCR assay. Diseases of Aquatic Organisms, 71, 149154.
  • 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, 12801290.
  • Kriger, K.M., Ashton, K.J., Hines, H.B. & Hero, J.M. (2007) On the biological relevance of a single Batrachochytrium dendrobatidis zoospore: a reply to Smith (2007). Diseases of Aquatic Organism, 73, 257260.
  • Lips, K.R., Brem, F., Brenes, R., Reeve, J.D., Alford, R.A., Voyles, J., Carey, C., Livo, L., Pessier, A.P. & Collins, J.P. (2006) Emerging infectious disease and the loss of biodiversity in a Neotropical amphibian community. Proceedings of the National Academy of Science of USA, 103, 31653170.
  • Lips, K.R., Diffendorfer, J., Mendelson, J.R. III & Sears, M.W. (2008) Riding the wave: reconciling the roles of disease and climate change in amphibian declines. PLoS Biology, 6, 441454.
  • Liu, X., Rohr, J.R. & Li, Y. (2013) Climate, vegetation, introduced hosts and trade shape a global wildlife pandemic. Proceedings of the Royal Society of London, B, 280, 20122506.
  • Longcore, J.E., Pessier, A.P. & Nichols, D.K. (1999) Batrachochytrium dendrobatidis gen. et sp. nov., a chytrid pathogenic to amphibians. Mycologia, 91, 219227.
  • Longo, A.V. & Burrowes, P.A. (2010) Persistence with chytridiomycosis does not assure survival of direct-developing frogs. EcoHealth, 7, 185195.
  • Longo, A.V., Burrowes, P.A. & Joglar, R.L. (2010) Seasonality of Batrachochytrium dendrobatidis infection in direct-developing frogs suggests a mechanism for persistence. Diseases of Aquatic Organisms, 92, 253260.
  • Lötters, S., Kielgast, J., Bielby, J., Schmidtlein, S., Bosch, J., Veith, M., Walker, S.F., Fisher, M.C. & Rödder, D. (2009) The link between rapid enigmatic amphibian decline and the globally emerging chytrid fungus. EcoHealth, 6, 358372.
  • Morgan, J.A.T., Vredenburg, V.T., Rachowicz, L.J., Knapp, R.A., Stice, M.J., Tunstall, T., Bingham, R.E., Parker, J.M., Longcore, J.E., Moritz, C., Briggs, C.J. & Taylor, J.W. (2007) Population genetics of the frog-killing fungus Batrachochytrium dendrobatidis. Proceedings of the National Academy of Sciences of the USA, 104, 1384513850.
  • Murray, K., Retallick, R., McDonald, K.R., Mendez, D., Aplin, K., Kirkpatrick, P., Berger, L., Hunter, D., Hines, H.B., Campbell, R., Pauza, M., Driessen, M., Speare, R., Richards, S.J., Mahony, M., Freeman, A., Phillott, A.D., Hero, J.M., Kriger, K., Driscoll, D., Felton, A., Puschendorf, R. & Skerratt, L.F. (2010) The distribution and host range of the pandemic disease chytridiomycosis in Australia, spanning surveys from 1956–2007. Ecology, 91, 1557.
  • Murray, K.A., Retallick, R.W.R., Puschendorf, R., Skerratt, L.F., Rosauer, D., McCallum, H.I., Berger, L., Speare, R. & VanDerWal, J. (2011a) Assessing spatial patterns of disease risk to biodiversity: implications for the management of the amphibian pathogen, Batrachochytrium dendrobatidis. Journal of Applied Ecology, 48, 163173.
  • Murray, K.A., Rosauer, D., McCallum, H. & Skerrat, L.F. (2011b) Integrating species traits with extrinsic threats: closing the gap between predicting and preventing species declines. Proceedings of the Royal Society of London, B, 278, 15151523.
  • Pearl, C.A., Bull, E.L., Green, D.E., Bowerman, J., Adams, M.J., Hyatt, A. & Wente, W.H. (2007) Occurrence of the amphibian pathogen Batrachochytrium dendrobatidis in the Pacific Northwest. Journal of Herpetology, 41, 145149.
  • Piotrowoski, J.S., Annis, S.L. & Longcore, J.E. (2004) Physiology of Batrachochytrium dendrobatidis, a chytrid pathogen of amphibians. Mycologia, 96, 915.
  • Puschendorf, R., Hoskin, C.J., Cashins, S.D., McDonald, K., Skerratt, L.F., Vanderwal, J. & Alford, R.A. (2011) Environmental refuge from disease-driven amphibian extinction. Conservation Biology, 25, 956964.
  • Richards-Zawacki, C.L. (2010) Thermoregulatory behaviour affects prevalence of chytrid fungal infection in a wild population of Panamanian golden frogs. Proceedings of the Royal Society B-Biological Sciences, 277, 519528.
  • Roberts, J.D. & Byrne, P.G. (2011) Polyandry, sperm competition and the evolution of anuran amphibians. Advances in the Study of Behavior, 43, 153.
  • Rödder, D., Veith, M. & Lötters, S. (2008) Environmental gradients explaining the prevalence and intensity of infection with the amphibian chytrid fungus: the host's perspective. Animal Conservation, 11, 513517.
  • Rohr, J.R., Raffel, T.R., Romansic, J.M., McCallum, H. & Hudson, P.J. (2008) Evaluating the links between climate, disease spread, and amphibian declines. Proceedings of the National Academy of Sciences of the USA, 105, 1743617441.
  • Ron, S.R. (2005) Predicting the distribution of the amphibian pathogen Batrachochytrium dendrobatidis in the New World. Biotropica, 37, 209221.
  • Rowley, J.J.L. & Alford, R.A. (2007a) Behaviour of Australian rainforest stream frogs may affect the transmission of chytridiomycosis. Diseases of Aquatic Organisms, 77, 19.
  • Rowley, J.J.L. & Alford, R.A. (2007b) Movement patterns and habitat use of rainforest stream frogs in northern Queensland, Australia: implications for extinction vulnerability. Wildlife Research, 34, 371378.
  • Rowley, J.J.L., Hemingway, V.A., Alford, R.A., Waycott, M., Skerratt, L.F., Campbell, R. & Webb, R. (2007) Experimental infection and repeat survey data indicate the amphibian chytrid Batrachochytrium dendrobatidis may not occur on freshwater crustaceans in northern Queensland, Australia. EcoHealth, 4, 3136.
  • Savage, A.E. & Zamudio, K.R. (2011) MHC genotypes associate with resistance to a frog-killing fungus. Proceedings of the National Academy of Sciences of the USA, 108, 1670516710.
  • Seymour, R.S., Roberts, J.D., Mitchell, N.J. & Blaylock, A.J. (2000) Influence of environmental oxygen on development and hatching of aquatic eggs of the Australian frog, Crinia georgiana. Physiological and Biochemical Zoology, 73, 501507.
  • Skerrat, 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, 125134.
  • Skerrat, L.F., Berger, L., Hines, H.B., McDonald, K.R., Mendez, D. & Speare, R. (2008) Survey protocol for detecting chytridiomycosis in all Australian frog populations. Diseases of Aquatic Organisms, 80, 8594.
  • Smith, K.G. (2007) Use of quantitative PCR assay for amphibian chytrid detection: comment on Kriger et al. (2006a, b). Diseases of Aquatic Organisms, 73, 253255.
  • Smith, K.G., Lips, K.R. & Chase, J.M. (2009) Selecting for extinction: non-random disease-associated extinction homogenizes amphibian biotas. Ecology Letters, 12, 10691078.
  • Smith, M.J., Withers, P.C. & Roberts, J.D. (2003) Reproductive energetics and behavior of an Australian myobatrachid frog Crinia georgiana. Copeia, 2003, 248254.
  • Taschetto, A.S. & England, M.H. (2009) El Nino Modoki impacts on Australian rainfall. Journal of Climate, 22, 31673174.
  • Tyler, M.J. & Doughty, P. (2009) Field Guide to Frogs of Western Australia. Western Australian Museum, Perth, Australia.
  • Vredenburg, V.T., Bingham, R., Knapp, R., Morgan, J.A.T., Moritz, C. & Wake, D. (2007) Concordant molecular and phenotypic data delineate new taxonomy and conservation priorities for the endangered mountain yellow-legged frog. Journal of Zoology, 271, 361374.
  • Vredenburg, V.T., Knapp, R.A., Tunstall, T.S. & Briggs, C.J. (2010) Dynamics of an emerging disease drive large-scale amphibian population extinctions. Proceedings of the National Academy of Sciences of the USA, 107, 96899694.
  • Wake, D.B. & Vredenburg, V.T. (2008) Are we in the midst of the sixth mass extinction? A view from the world of amphibians. Proceedings of the National Academy of Sciences of the USA, 105(Suppl 1), 1146611473.
  • Walker, S.F., Salas, M.B., Jenkins, D., Garner, T.W.J., Cunningham, A.A., Hyatt, A.D., Bosch, J. & Fisher, M.C. (2007) Environmental detection of Batrachochytrium dendrobatidis in a temperate climate. Diseases of Aquatic Organisms, 77, 105112.
  • Wayne, A.F., Liddelow, G.L. & Williams, M.R. (2011) FORESTCHECK: terrestrial vertebrate associations with fox control and silviculture in jarrah (Eucalyptus marginata) forest. Australian Forestry, 74, 336349.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
jpe12091-sup-0001-AppendixS1-S2_TableS1-S6.docxWord document40K

Appendix S1. DNA extraction and quantification of chytrid load.

Appendix S2. Climate data sources and missing climate data.

Table S1. Frog species from south-western Australia, chytrid status and breeding biology.

Table S2. Sampling sites, locality data, date sampled in 2008, mean snout-vent length (SVL mm).

Table S3. Short term climate summaries by site for 15, 30 and 45 days prior to sampling.

Table S4. Long term climate averages by site, 2003–2007 inclusive.

Table S5. Short-term climate and occurrence of chytrid fungus – PCA results.

Table S6. Long-term climate and occurrence of chytrid fungus – PCA results.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.