Publications
NSW National Parks and Wildlife Service, July 2001
ISBN 0 731 36213 6
Observations made between 1955 and 1970 indicate that the Southern Corroboree Frog was abundant within its limited geographic range. High numbers of individuals were frequently recorded at suitable breeding sites (see Osborne 1988 for a historical review). However, by 1986 it was apparent that the species had declined considerably in abundance (Osborne 1989). Subsequent monitoring over a thirteen year period (1986-1999) indicated that the populations continued to decline and that remaining populations are now severely depleted (Hunter et al. 1997; Osborne and Davis 1997, Osborne et al. in press). The species has disappeared from 70% of the sites at which it formerly occurred, and the number of remaining adult males is estimated to be in the order of 300-400 individuals (Hunter et al. 1997; Hunter in prep.).
In the summers of 1985 and 1986 Osborne (1988, 1989) undertook the first extensive surveys of the distribution and relative abundance of the Southern Corroboree Frog. He delineated the geographic distribution of the species and compared current population levels to observations made by earlier workers (including his own observations in the late 1970's). Osborne (1988, 1989) concluded that the numbers of frogs present at most breeding sites were very low; at 74% of the sites in the Snowy Mountains ten or less calling males were recorded at each site. Choruses estimated at greater than 25 calling males were recorded at only 15% of the sites with frogs. These larger sites occurred near Mt Jagungal and along the Toolong Range south of Round Mountain in the Jagungal Wilderness area.
For many of the historical records of the Southern Corroboree Frog in the Snowy Mountains there is little information available on numbers of individuals observed. In such cases any assessment of changes in population status can only compare the present distribution to past known occurrences. At best this can provide an indication of broad changes in distribution. However, additional information on relative abundance is available for a number of sites (summarised in Osborne 1988, 1991) providing a baseline against which an assessment of whether or not the species is declining can be made.
Population monitoring programs at several locations have been under way for twelve years (1986-1998) (Osborne 1991, Hunter et al. 1997; Osborne and Davis 1997; Osborne and Hunter unpublished data). This has indicated that populations have not recovered, and that there has been a gradual loss of local breeding aggregations and a corresponding contraction of geographic range (Figures 2 and 3). In the Smiggin Holes and Guthega region all eleven remaining breeding sites (see Osborne 1991 for further details) have been monitored using non-invasive techniques annually since the summer of 1985/86. Each of these small local populations has declined to extinction. This pattern of severe declines has been repeated at most other sites subject to monitoring. Of the 60 former sites surveyed only eight were found to still have frogs. At sites where frogs still occur, the numbers of adult males remaining is very low; 50 of the 63 sites had fewer than six adult males present in 1996-98. Only one large population remained in 1998 (95 adult males in 1998).
Amphibian populations may undergo significant natural fluctuations in time and space. Detection of frogs can vary markedly with season or local climatic conditions, which may affect calling and other activity levels (Duellman and Trueb 1986). Studies undertaken overseas indicate that amphibian populations may exhibit considerable fluctuation over several years, and that patterns of population change can vary between species (Berven 1990; Pechmann et al. 1991). The initial cause of the decline in the Southern Corroboree Frog was thought to be the 1982-83 drought (Osborne 1989), and a recent analysis of the long-term trends in precipitation indicates that declining autumn and winter precipitation (particularly snow cover) may have been implicated in the decline (Osborne and Davis 1997). However, the link to precipitation is not readily obvious and the actual cause of the decline is still not known, particularly since the populations have not recovered in response to more-favourable weather conditions.
Understanding what constitutes natural population changes due to normal climatic or environmental variation is necessary before the significance of the decline in the Southern Corroboree Frog can be fully assessed. This can only be achieved by monitoring relative abundance of populations over time periods long enough to establish the extent of ecological stability or expected patterns of fluctuation. This is generally considered to be at least one generation turnover (Connell and Sousa 1983; Blaustein et al. 1994; Pechmann and Wilbur 1998). It is likely that the Southern Corroboree Frog takes three years to reach reproductive maturity (Pengilley 1966 and pers. comm.; Osborne 1991; D. Hunter pers. comm.), and may live for up to eight years (D. Hunter pers. comm). Thus, the current monitoring which includes 12 consecutive years is likely to be long enough to have detected demographic changes. It also includes a complete El Nino cycle. It is highly unlikely that the extent of decline in the Southern Corroboree Frog is simply a reflection of a normal population fluctuation.
What is not known is whether the observed decline is part of a very long-term population cycle, although this seems unlikely given the disappearance of the frogs from so many sites, including from regions that would be difficult to re-colonise (eg. south of the Snowy River where the species now appears to be extinct). Long-term monitoring is required to determine if population declines are continuing, or are part of a long-term cycle. Such monitoring is also essential for determining the success of any experimental trials involving habitat manipulation, recruitment enhancement or translocation.
The cause of the present serious decline in populations of the Southern Corroboree Frog is unknown. As mentioned above, it was originally assumed that the decline was the result of drought that affected the region in the early 1980's, and that once conditions had improved, the frog population would recover (Osborne 1989). However, this has not been the case; local populations have continued declining, or remained low, during a decade which has included many potentially good breeding seasons (Osborne and Davis 1997). The frog is faced with a considerable risk because of its specialised life history and its high degree of breeding habitat specificity. It has a very low clutch size, each female breeds only once each season, and the tadpoles are slow-growing, spending over six months in the shallow pools. Such a strategy reduces the ability of the species to recover quickly during favourable seasons, and places it at risk from any long-term disturbance which affects the breeding sites.
Whilst the principal cause of the population decline and range contraction has not been identified, it is clear that its effect is acting at least on a regional basis since all high altitude populations, and some low altitude populations, of Northern Corroboree Frogs are also in decline (ACT Government 1997; Hunter and Osborne unpublished data). Whilst activities such as ski-resort development, road construction and the operation of a hydro-electrical scheme undoubtedly had major localised influences on frog breeding habitat in KNP (see below), the likely cause of the current decline is obviously a response to a wider-reaching phenomenon.
Concern about global warming (Pearman 1989; Galloway 1989; Whetton et al. 1996) has a particular significance for the conservation of cool-adapted species such as the Southern Corroboree Frog (Bennett et al. 1991). Due to its restricted subalpine distribution the species is likely to be particularly susceptible to climate change. Global warming has the potential to alter the breeding season and change the period required for eggs and tadpoles to develop; this may lead to these events occurring earlier or later than at an optimum time. A change in regional temperature and precipitation is also likely to influence the hydrology of the breeding pools, and affect the growth and dynamics of vegetation in the breeding habitat.
During recent monitoring of breeding sites in the Snowy Mountains, it has been observed that some pools within breeding sites have become overgrown by Sphagnum moss. Although such hummock-hollow development in bogs is a natural part of the bog dynamics (Clark 1980), other factors including reduced snow cover, and warming may allow the moss to grow more prolifically (Tallis 1994). It should be noted that loss of breeding pools (at least the larger, more obvious pools) is not yet evident at most of the sites monitored in the Snowy Mountains, although this needs to be examined in detail during further surveys. Perhaps of greater importance is the possibility, as mentioned above, that with warmer temperatures, or longer periods of drier weather during spring and early summer, the pools still containing tadpoles may dry (Osborne 1990; Pengilley 1992).
Of equal importance would be a shift in the seasonality of precipitation or a decline in winter snow cover. A preliminary analysis of long-term trends in snow-cover, precipitation and temperature indicated that declining winter snow cover and precipitation may be a factor contributing to the decline (Osborne and Davis 1997; Osborne et al. 1998). In 1997 over-winter survival of developing encapsulated tadpoles of the Southern Corroboree Frog and of subalpine populations of Northern Corroboree Frog was extremely low, with tadpoles surviving through the winter at only a few of the sites where eggs were observed in the previous summer (Hunter, et al. 1999). The reason for this high mortality is not known, but two hypotheses have been proposed: (1) due to reduced snow cover, tadpoles were exposed to sub-zero temperatures; and (2) due to a low water table and minimal autumn and winter precipitation nest sites did not flood and hatched tadpoles were unable to move to the nearby pools. Testing of these hypotheses requires further research, which directly compares tadpole survivorship with hydrological and meteorological measurements.
Ultraviolet radiation (UV-B) has increased significantly in recent years due to increasing ozone (eg. Jones and Shanklin 1995), and is likely to increase as reduction in ozone in the upper atmosphere continues. Some amphibians are known to have increased levels of developmental abnormalities after exposure to artificial UV-B (Grant and Licht 1995; Licht and Grant 1997) and increased ultraviolet radiation is implicated in frog declines at high altitudes where there is less atmosphere for UV-B to pass through (Blaustein et al. 1994). Recent research indicates that the developing eggs of some alpine amphibians are damaged by ambient ultraviolet radiation (Blaustein et al. 1994, 1995).
It is well known that UV-B radiation causes developmental abnormalities in amphibians (see review by Grant and Licht, 1995) but few field-based studies have addressed the possible effects. Experiments in North America (Blaustein et al. 1994; Blaustein et al. 1995) demonstrated species-specific differences in the ability of eggs to repair radiation damage to DNA and differential hatching success of embryos exposed to solar radiation at natural egg laying sites when compared to controls shielded from UV radiation. Blaustein et al. (1994) argue that at high elevations, development times in amphibians may be extended because of lower ambient temperatures, thereby exposing the developing eggs and tadpoles for longer periods. Moreover, high altitude amphibians are often heliothermic and actively seek out sunlight to increase developmental temperatures (Stebbins and Cohen, 1995). Although higher elevation populations may be more resistant to UV or may have behavioural adaptations to minimise UV damage, these may be insufficient to cope with the dramatic increases in UV levels currently experienced at high altitudes (Blaustein et al. 1995).
Ultraviolet radiation can reach significant depths in natural waters (Smith, 1989), and in clear freshwater lakes potentially harmful intensities of UV-B radiation can penetrate to several metres (Schindler et al. 1996). With respect to the potential for causing damage to biological organisms, shallow alpine lakes, pools and streams are likely to be of greatest concern, as they are likely to have very low amounts of dissolved organic carbon and are exposed to high levels of ambient UV-B. In the shallow, extremely clear pools found in alpine regions, high levels of UV-B are likely to penetrate to all depths used as egg development sites by amphibians.
In a study conducted along an elevational gradient at Thredbo in Kosciuszko National Park, Broomhall et al. (in press) found that newly-hatched tadpoles of a declining frog (Alpine Tree Frog Litoria verreauxii alpina) showed significantly higher levels of mortality when not protected by UV-B filters, when compared to tadpoles protected by mylar UV-B filters. Moreover, a non-declining species used as a control (Common Eastern Froglet Crinia signifera) had significantly lower mortality under all treatments. Whilst it could be argued that ultraviolet radiation is unlikely to affect the adults or eggs of P. corroboree because they are hidden within the moss and are unlikely to be exposed, the larval life stage of the frog may be at risk because they are confined to highly-exposed shallow, clear pools. Research on the susceptibility of the Southern Corroboree Frog eggs, embryos and tadpoles should be undertaken.
As mentioned in section 3.5.4 the decline in both species of Corroboree Frog could also be a result of disease. The hypothesis that at least some declines in amphibians have been caused by pathogens (eg. Carey 1993; Laurance 1996; Laurance et al. 1996) has recently gained favour, amongst at least some researchers (however, see Hero and Gillespie 1997; Alford and Richards 1997), following identification of a highly virulent fungal pathogen (Chytridiomycosis) in moribund frogs at sites where declines have occurred (Berger and Speare 1998; Berger et al. 1999).
Chytridiomycosis has been detected in Spotted Tree Frogs (Litoria spenceri) that have died in the field in the Snowy Mountains (G. Gillespie pers. comm.) and the carcasses of captive adult Northern Corroboree Frogs that died over winter at the Amphibian Research Centre in Melbourne were found to contain a range of pathogens including a fungus similar to Chytridiomycosis (G. Marantelli pers. comm.). Nevertheless, it is likely that partially decomposed carcasses will contain a range of microbial decomposers, so that this observation is not unexpected.
Rick Speares (James Cook University, pers. comm.) has recently detected the pathogen in museum specimens of the Southern Corroboree Frog collected in 1992 and 1993. However, there is no field evidence to date that either species has suffered rapid population decline as a result of fast-acting pathogens. Some researchers suggest that it is likely that more than one causal agent is involved in frog declines in Australia (and elsewhere) and that these may in fact act in a synergistic manner. A fast-spreading disease is less-likely to be the cause of the decline in Northern Corroboree Frogs because the decline has occurred slowly, with populations gradually becoming depleted before local extinctions have occurred (Osborne et al. in press). There is, however, very good reason for concern. As further information on disease in Australian frogs comes to light, tests should be made to examine the susceptibility of both species of Corroboree Frogs.
An extensive experimental bushfire at Bushrangers Catchment in the Brindabella Range eliminated a very small local population of Northern Corroboree Frogs (Osborne, unpublished data). However, near Round Mountain, and in the Maragle Range in Kosciuszko National Park, subalpine bushfires that burnt to near the edges of Sphagnum bog pools during the breeding season did not prevent the frogs from returning to the pools the following season. These fires occurred at a time when adult males and females were in or near the moist breeding habitat and thus their immediate impact on adults may not have been great. However, autumn fires burning through woodland and heath surrounding breeding sites are likely to have a greater potential influence. At this time adult and sub-adult frogs have moved distances of up to 300 m or more, away from the bogs, to feed and to find suitable over-wintering sites. Extensive burning of understorey litter and grass cover in these areas, such as occurs during prescribed burns, is likely to reduce the shelter available to the frogs and make them more vulnerable to dehydration or freezing. Earlier this century, an increased frequency of autumn burning in association with grazing activity may have had a significant impact (K. McDougall, pers. comm.).
Livestock grazing may also have been a historical cause of habitat deterioration. Pengilley (1966) suggested that the Snowy Mountains had been the most severely affected by modification of its breeding habitat due to livestock grazing. He considered that it was possible that some contraction of the range of the frogs may have occurred as a result of drainage of sphagnum bogs and their conversion to grazing land. This suggestion is reinforced by well documented observations of a deterioration in vegetation cover and a significant increase in soil erosion during the grazing era in the Snowy Mountains (Byles 1932; Costin 1954; Wimbush and Costin 1979; McDougall 1989). Costin (1954) noted that trampling by livestock rapidly breaks the ground surface in moist peatlands leading to incision of the bogs with deeper drainage channels. These deeper channels then hasten the drying out and humification of the peat, which may be further degraded by wind and water erosion. Grazing in subalpine grasslands and bogs was often accompanied by burning, which probably further reduced the protective cover in both the breeding habitat and non-breeding habitat of the frogs.
Given the importance of shallow seepage and bog pools as breeding sites for the frogs, it is likely that disturbance to these areas during the high-country grazing era (1889 to 1958) resulted in the loss of some local populations. It is now over 30 years since the cessation of grazing and the condition of bogs and other potential breeding habitats is reported to have improved (A.B. Costin pers. comm.; R.B. Good pers. comm.), and the effects are now likely to be greatly diminished. Nevertheless, long-term monitoring studies such as those of Wimbush and Costin (1979) should be encouraged as they provide valuable information on the recovery of these areas. For example, it is possible that, due to extensive former disturbance, incision and lowering of the water table, seepages and bogs no longer support the range of pools that existed before stock grazing. This hypothesis requires further consideration and could be tested experimentally by construction of a variety of artificial pools and monitoring of their suitability.
Significant losses of Southern Corroboree Frog breeding habitat probably occurred during the construction of the Snowy Mountains Hydro-Electric Scheme between 1948 and 1972. Near Smiggin Holes, Pipers Saddle, Wragges Creek and Guthega, extensive road-making, aqueduct construction and other activities influencing drainage are likely to have resulted in the drying out of some areas, particularly where water was diverted away from suitable breeding habitats or where the water table was lowered. The gradual loss of some populations may have been exacerbated by run-off of road silt and gravel. Areas impacted upon by construction have now been stabilised by soil conservation measures and natural rehabilitation and, unless substantial new earth works are proposed, are unlikely to present an additional impact to frog populations.
Southern Corroboree Frogs were once common in at least three ski resort areas: Smiggin Holes, Blue Cow and Guthega (Osborne 1988). They almost certainly also occurred at Perisher Valley but there have been no reliable records of the species from this area. Osborne (1988 and 1991) presents a detailed account of the potential impacts of ski developments on the frogs; this will not be repeated here. It now seems certain that the Southern Corroboree Frog no longer occurs in any resort areas (W. Osborne unpublished data; see also Osborne 1988), but the disappearance of the frogs from these areas is obviously symptomatic of the more widespread decline.
Apart from direct mechanical impacts on habitat, the construction activities outlined in Osborne (1991) may have other impacts that apply to a wider range of situations. These are: (1) increased erosion and siltation, (2) increased nutrient run-off, and (3) stream or seepage diversion or drainage. These are briefly discussed below:
Erosion and subsequent siltation resulting from construction activities would be expected to have a significant impact on breeding sites. The pools used by Southern Corroboree Frogs as breeding sites are generally shallow, clear pools with few dissolved ions and a moderately low pH. Any sediment run-off and increase in turbidity in the pools would potentially threaten tadpoles by mechanically blocking their gills, smothering the detritus in the pools which provide the food source of the tadpoles, or filling the pools with sediment to the extent that they no longer contain water.
Fertilisers are commonly used during rehabilitation of disturbed areas when agronomic plant species are being sown for revegetation. There is often considerable run-off of nutrients from these sites as evidenced by the spread of exotic plant species down drainage lines below buildings, roads and carparks. The influence of higher levels of nutrient input into breeding pools is not known, but would probably present problems if the native vegetation cover was replaced by taller plants which shaded the pools, or if the higher nutrient levels caused excessive algal build-up and eutrophication of the pool.
Ski-slope development and hydro-electric construction may alter the drainage characteristics of an area, commonly initiating erosion and drainage of bogs and wet heaths. Activities such as the construction of underground sewers, the laying of buried cables and the building of drains could detrimentally affect the frogs if the construction diverts water away from breeding sites, causes a lowering of the water table in the breeding area, or leads to a concentration of the flow.
Given the potential influences described above, it is unlikely that the Southern Corroboree Frog can be maintained in areas that have been extensively disturbed as a result of ski resort development. Osborne (1991) suggests a need to protect the hydrological integrity of the breeding habitat by maintaining an extensive protective buffer around breeding sites. The requirement for a total prohibition on disturbance within the breeding areas would be difficult to meet given the present extent of development in the resort lease areas.
Suitable breeding sites still occur within the resort lease boundaries, although the condition of potential breeding habitat in these areas varies considerably (Osborne 1988). Pools remote from groomed ski runs and roads appear to be in relatively good condition and provide potential habitat. These sites have been mapped (unpublished maps prepared by W.S. Osborne and held by NPWS at the Snowy Mountains Regional Office, Jindabyne). Pools close to, or within, ski slopes and lift lines, and near roads, are usually highly disturbed and do not provide potential breeding habitat. Many of these pools within summer-groomed ski slopes, close to lift lines and near roads, have been highly disturbed as a result of past developments and do not provide potential breeding habitat. They may contain rubbish, grooming rubble and tree cuttings. Vehicle activity has also damaged some of the sites and some pools appear to contain sediment washed from nearby roads or lift access tracks. While current practices for ski resort development (eg. Perisher Blue Pty Limited 2000b) are aimed at protecting Southern Corroboree Frog habitat from further disturbance, those pools which have suffered past impacts are unlikely to recovery in the foreseeable future.
Although the prospects for Corroboree Frogs in the more intensively developed areas of ski resorts appear to be poor, intact areas of habitat remain in other parts of the resort areas. It should be noted that, until recently, at Mt. Baw Baw in Victoria the endemic Baw Baw Frog, Philoria frosti, still occurred in and near managed ski slopes (Malone 1985). Therefore, the possibility of a recovery of Corroboree Frog populations in the Smiggin Holes-Guthega-Blue Cow area should not be excluded completely, and conservation plans for the species should include a consideration of potential breeding sites in the resort areas. Such provisions are included in the Ski Slope Plan for the Perisher Blue Ski Resort (Perisher Blue Pty Limited 2000a).
The Southern Corroboree Frog is an attractive species and because of its appeal is often searched for, and sometimes removed, by visitors. However, because of their specialised diet the frogs may be difficult to maintain for long periods in captivity and thus have low potential value for the pet trade.
The physical intervention involved in searching for Corroboree Frogs usually results in disturbance to the nest site. After being disturbed, males often begin calling again at the same site or move a short distance away (D. Hunter unpublished data). However, the activities of people trying to find the frogs could be expected to seriously disrupt reproductive success at sites with low numbers of frogs. In addition, the pulling apart of moss and other vegetation by people searching for frogs may expose some of the clutches to dehydration during dry summer conditions.
The second aspect of collecting is the potential for over-collecting for scientific purposes. In the past very large samples of Corroboree Frogs have been removed from various readily accessible sites in Kosciuszko National Park and in the Brindabella Range (see comments in Osborne 1988). However, these early scientific collections apparently did not permanently affect the size of local breeding groups, with most collecting sites at a later stage still supporting large numbers of frogs (Dr. R.K. Pengilley pers. comm.; D.J. Wimbush pers. comm.; Dr. R.E. Barwick pers. comm.). It is possible that intense illegal collecting could deplete local populations, but it is unlikely that scientific collecting would be undertaken without application first being made for research permits. Under the present extremely low population numbers any research on the Southern Corroboree Frog that involves removal of specimens has the potential to disturb already seriously depleted populations, including the possibility of introducing exotic diseases. Therefore, any new research proposals that involve removal of frogs require careful consideration by the NPWS and by the Recovery Team.
Sites supporting Southern Corroboree Frogs are currently virtually free of weeds. Plantings of exotic trees, such as willows Salix spp., for soil conservation purposes by the Snowy Mountains Hydro-Electricity Authority, have been widespread in the Snowy Mountains. These plantings occur along roads, aqueducts and other areas disturbed during construction activities. Although no breeding sites are directly threatened by willow invasion, in the longer-term the spread of willows by sexual reproduction and the production of seed and also through vegetative growth along seepages and streams may present a problem for the management of some sites. In Victoria, some bogs are being invaded by Mimulus moschatus, Mentha spicata and Juncus effusus (K. McDougall, pers comm.). Of these, Mimulus moschatus appears to be the biggest threat as it grows well in the sphagnum and may form quite large patches. All three species are present in Kosciuszko National Park.
Feral Pigs (Sus scrofa) sometimes damage breeding areas used by Southern Corroboree Frogs. They may also damage over-wintering habitat by overturning shelter sites such as logs and destroying dense ground cover. This is mainly confined to areas between Kiandra and Mt Jagungal and on the Toolong Range, but pigs appear to be slowly spreading to other subalpine areas (C. Smith NPWS pers. comm.). Additional research on the interaction between pigs and Corroboree Frogs would be useful to better predict the likely impacts.